Adjunctive colposcopy technologies for examination of the uterine cervix - DySIS, LuViva Advanced Cervical Scan and Niris Imaging System: a systematic review and economic evaluation

Notes

Permissions

Chapter 1 Background and definition of the decision problem

Chapter 2 Assessment design and results by condition or aetiology

Systematic review of clinical effectiveness

Background

A systematic review was undertaken to assess the clinical effectiveness of adjunctive colposcopy technologies DySIS, LuViva Advanced Cervical Scan and Niris Imaging System for patients referred for colposcopy through the NHS Cervical Screening Programme.

The original scope for the assessment also included the APX 100 device (developed by Zilico Ltd, Manchester, UK). 3 However, this technology was removed from the assessment in December 2011, after the inclusion screening stage of the assessment.

Methods for reviewing clinical effectiveness

The systematic review was conducted following the general principles recommended in the Centre for Reviews and Dissemination (CRD) guidance13 and the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement. 14

Search strategy

The literature search aimed to systematically identify research related to the clinical effectiveness and cost-effectiveness of adjunctive colposcopy technologies.

The base search strategy was constructed using MEDLINE and then adapted to the other resources searched. The search included the following components:

1. terms for cervix, and

2. terms for colposcopy (including both general colposcopy terms as well as specific technologies).

Searches of major bibliographic databases were limited by date (2000 onwards) reflecting the date of development of the new technologies. No language, study design or other limits were applied. Reference lists of all included studies were hand-searched to identify further relevant studies. Where necessary, authors of eligible studies were contacted for further information.

Search strategies were developed by an information specialist with input from the project team. The search strategy was checked by a second information specialist. Sources of information were identified by an information specialist with input from the project team.

As the technologies involved are relatively new, particular attention was given to identifying sources for ongoing trials and conference reports (by searching specialist sources such as Inside Conferences and ClinicalTrials.gov). Details of the search strategies are presented in Appendix 1.

The following resources were searched for relevant clinical effectiveness and cost-effectiveness research:

• Allied and Complementary Medicine Database (AMED): via OvidSP, using the segment 1985 to September 2011, searched on 22 September 2011

• BIOSIS Previews: via Dialog, using the segment 1993 to 2011 week 2 October, searched on 19 October 2011

• Cochrane Database of Systematic Reviews (CDSR): via Wiley Cochrane Library website, Issue 9 of 12, September 2011, searched on 22 September 2011

• Cochrane Central Register of Controlled Trials (CENTRAL): via Wiley Cochrane Library website, Issue 3 of 4, July 2011, searched on 22 September 2011

• Cumulative Index to Nursing and Allied Health Literature (CINAHL): via EBSCO, using the segment 1981 to 16 September 2011, searched on 22 September 2011

• ClinicalTrials.gov: via website www.clinicaltrials.gov/, using the segment to September 2011, searched on 28 September 2011

• Current Controlled Trials (CCT): via website www.controlled-trials.com/, using the segment to September 2011, searched on 28 September 2011

• Database of Abstracts of Reviews of Effects (DARE): via Wiley Cochrane Library website Issue 3 of 4, July 2011, searched on 22 September 2011

• EMBASE: via OvidSP, using the segment 1996 to week 37 2011, searched on 22 September 2011

• Health Management Information Consortium (HMIC): via OvidSP, using the segment 1985 to September 2011, searched on 22 September 2011

• Health Technology Assessment (HTA) database (via Wiley Cochrane Library website Issue 3 of 4, July 2011, searched on 22 September 2011

• Inspec: via OvidSP, using the segment 1969 to week 36 2011, searched on 22 September 2011

• Inside Conferences: via Dialog, using the segment 1993 to 18 October 2011, searched on 19 October 2011

• MEDLINE: via OvidSP, using the segment 1948 to September week 2 2011, searched on 22 September 2011

• NHS Economic Evaluation Database (NHS EED): via Wiley Cochrane Library website Issue 3 of 4, July 2011, searched on 22 September 2011

• PASCAL: via Dialog, using the segment 1973 to 2011 week 2 October, searched on 19 October 2011

• Science Citation Index Expanded (SCIE): via Web of Knowledge, using the segment 2000 to 22 September 2011, searched on 23 September 2011

• Science Citation Index (SCI) – Conference Proceedings: via Web of Knowledge, using the segment 1990 to 22 September 2011, searched on 23 September 2011.

Additional searches were conducted to identify systematic reviews of colposcopy in an attempt to ascertain the diagnostic accuracy of colposcopy:

• CDSR: via Wiley Cochrane Library website Issue 10 of 12, October 2011, searched on 25 October 2011

• DARE: via CRD administration database, using the segment to 25 October 2011, searched on 25 October 2011

• DARE: via Wiley Cochrane Library website Issue 4 of 4, October 2011, searched on 25 October 2011.

The following websites were searched for guidelines and care pathways:

• Scottish Intercollegiate Guidelines Network (SIGN) (www.sign.ac.uk/, searched on 16 June 2011)

• National Institute for Health and Clinical Excellence (NICE) (www.nice.org.uk/, searched on 16 June 2011)

• National Guideline Clearinghouse (www.guidelines.gov/, searched on 16 June 2011)

• National Institute for Health Research (NIHR) Health Technology Assessment programme (www.hta.ac.uk/, searched on 16 June 2011)

• NHS Evidence (www.evidence.nhs.uk/, searched on 16 June 2011)

• TRIP database (www.tripdatabase.com/, searched on 16 June 2011).

Inclusion and exclusion criteria

Two reviewers independently screened all titles and abstracts. Full paper manuscripts of any titles/abstracts that appeared to be relevant were obtained, where possible, and the relevance of each study independently assessed by two reviewers according to the inclusion and exclusion criteria below. Studies that did not meet all of the criteria were excluded and their bibliographic details listed with reasons for exclusion. Any discrepancies were resolved through consensus, with involvement of a third reviewer when necessary.

As stated earlier, the original scope for the assessment also included the APX 100 device, developed by Zilico Ltd. 3 Since this technology was removed from the assessment in December 2011, after the inclusion screening stage of the assessment, inclusion criteria refer to the APX 100 device.

• Study design Comparative studies, including diagnostic test accuracy studies and controlled trials, were included in the evaluation of clinical effectiveness, as this study design allows a comparison to be made between the new technology and current practice, which is essential for the economic model.

• Intervention Studies assessing DySIS, LuViva Advanced Cervical Scan, Niris Imaging System or APX 100, alone or alongside colposcopy, were included in the evaluation of clinical effectiveness.

• Comparators Studies that compared one of the adjunctive colposcopy technologies with standard colposcopy were included in the evaluation of clinical effectiveness.

• Participants The population of interest is women referred for colposcopy through the NHS Cervical Screening Programme. Therefore, studies of women referred for colposcopy because of an abnormal cytology result were included in the evaluation of clinical effectiveness. Studies that also included women referred for colposcopy because of symptoms indicative of cervical cancer (e.g. postcoital bleeding) or women referred for colposcopy for follow-up of CIN were also eligible for inclusion; however, studies that included only women referred for symptoms or for follow-up were not eligible for inclusion.

• Outcomes The clinical outcomes of interest were diagnostic test accuracy outcomes (e.g. sensitivity and specificity), adverse effects and patient experience. Where other patient health outcomes were reported (e.g. morbidity and mortality from cancer or treatment), these were also included in the assessment.

Data extraction strategy

Data on study and participant characteristics and outcomes were extracted by one reviewer using a standardised data extraction form and independently checked for accuracy by a second reviewer. Disagreements were resolved through consensus, with involvement of a third reviewer when necessary.

Where sufficient data were available, the following diagnostic accuracy statistics [with 95% confidence intervals (CIs)] were calculated, for each study, using the Canadian Institute of Health Research's Knowledge Translation statistics calculator:15 sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR), and negative LR. Subsequently, accuracy was also calculated (the proportion of true-positive and true-negative results).

To allow consistency when comparing studies, in our results section we have reported our calculated results, rather than those reported in the study reports (as our results sometimes differed slightly from those in the study reports). Where data were missing from publications or other study reports, the authors were contacted (via NICE in the case of the manufacturers of the technologies). Data from multiple publications of the same study were extracted as a single study. The data extraction tables are presented in Appendix 2.

Quality assessment strategy

The quality of the included studies was assessed using the QUADAS-2 quality assessment tool for diagnostic studies. 16 As well as adding review-specific questions to domains 2 and 3, three further quality-related questions were assessed (see Appendix 3 for details). The assessment was performed by one reviewer and independently checked by a second. Disagreements were resolved through consensus, with involvement of a third reviewer when necessary. Further details about QUADAS-2 and results of the quality assessment are presented in Chapter 2 (see Quality of research available) and Appendix 3.

Data analysis

In view of the heterogeneity between the included studies, in terms of participant characteristics and the different comparator technologies used, formal meta-analysis was not appropriate. Therefore, the studies were grouped according to the adjunctive technology used and a narrative synthesis was presented.

Results of the review of clinical effectiveness

Quantity of research available

A total of 7835 records were identified from the clinical effectiveness searches and an additional 69 records were identified via hand-searching or contact with the manufacturers (via NICE) (Figure 1).

FIGURE 1.

Flow diagram of the study selection process.

Seven studies (reported in 31 references) met the inclusion criteria. Details of studies excluded at the full publication stage are provided in Appendix 4.

On 21 December 2011, after we had finished screening studies for inclusion, we were informed by NICE that the APX 100 device, developed by Zilico Ltd, should be omitted from the assessment (one study, reported in four references). Therefore, six studies (reported in 27 references) were included in the review.

There were two main studies of the DySIS technology6,17 and two additional subgroup assessments; the two main studies6,17 were published in full, whereas one of the subgroup assessments was an unpublished draft manuscript (Zaal et al. , The VU University Medical Center, Amsterdam, the Netherlands, 2011) and the other subgroup assessment18 was reported in a conference abstract.

There was one study (Flowers et al. , University of Emory School of Medicine, Atlanta, GA, 2011) and one subgroup assessment19 of the LuViva Advanced Cervical Scan. The main study was an unpublished draft manuscript, whereas the subgroup assessment was reported in a conference poster. 19 he remaining 11 records were conference abstracts,20–23 presentations,24–26 a flyer,27 a ClinicalTrials.gov record,28 the manufacturer's presentation for NICE29 and the manufacturer's response to a question from the US Food and Drug Administration (FDA). 30 In addition, we received further clarification of methods and additional results via personal correspondence with the manufacturer on a number of occasions. However, there were some inconsistencies in the information we received; therefore, we are not entirely confident in the accuracy of these additional data. Results data received via personal correspondence have been highlighted as such in the summary of study characteristics and results (see Table 6) and the data extraction tables in Appendix 2.

There were three studies of the Niris Imaging System, all published in full. 31–33 The remaining seven records were conference abstracts,34–35 presentations36 and posters,37–38 the draft manuscript for one of the published papers (Liu et al. , Peking University Shenzhen Hospital, Shenzhen, China, 2009) and a draft book chapter that described one of the published studies. 39

Quality of research available

The QUADAS-2 tool, developed to improve, and to allow greater rating transparency than the original QUADAS tool, separates the evaluation of study quality into two main areas: risk of bias, and concerns regarding applicability. The tool consists of four domains: patient selection, index test, reference standard, and flow and timing. For individual studies each domain is assessed as being at a high, low, or unclear risk of bias, with the first three domains also assessed in terms of applicability concerns (also using high, low, or unclear ratings). The domains are supported by signalling questions, to help judge risk of bias. 16

Table 3 summarises the results of the QUADAS-2 assessments. Across almost all of the studies there were few applicability concerns in relation to appropriate patient recruitment and reference standard use. However, for the majority of studies, there were often difficulties in appraising risk of bias due to poor reporting, and there were also various applicability concerns about the conduct or interpretation of the adjunctive technologies. In general, study quality differed according to the type of adjunctive technology.

TABLE 3 - Summary of QUADAS 2 assessment results

In the NHS, histopathologists do not normally use colposcopy results to inform their diagnoses, therefore knowledge of colposcopy results before interpreting biopsy results was not assessed, although knowledge of adjunctive technology results was assessed.

Note: based on subsequent information from DySIS Medical/Imalux Corporation, this should read ‘Low’.

Study	Concerns
	Risk of bias					Applicability
		Index test					Index test
	Patient selection	Adjunct	Comparator	Reference standarda	Flow and timing	Patient selection	Adjunct	Comparator	Reference standard
Soutter,17 DySIS	Low	Low	Low	Low	High	Low	High	High	Low
Louwers,6 DySIS	Low	Low	Low	Unclearb	Low	Low	Unclearb	High	Low
Flowers, unpublished, LuViva	Unclear	Unclear	Unclear	Low	High	Low	High	Low	Low
Escobar,31 Niris	Unclear	Unclear	Unclear	Low	Low	Low	High	Low	Low
Liu,32 Niris	Unclearb	Unclearb	Unclearb	Unclearb	Unclear	Low	High	Low	Low
Gallwas,33 Niris	Unclear	Unclear	Unclear	Low	High	Low	High	Unclear	Unclear

DySIS

The two DySIS studies were judged to be at low risk of bias in terms of both patient selection and conduct and interpretation of the DySIS and colposcopy examinations. 6,17 However, there were applicability concerns in both studies relating to the conduct of colposcopy; video colposcopy using the DySIS colposcope was used, rather than the conventional colposcopy methods and equipment used in the NHS. The accuracy of colposcopy in these studies may therefore not be an accurate reflection of current NHS practice. Furthermore, in the earlier study a precommercial model was used, raising both applicability and bias concerns; around one-third of patients were excluded, largely due to equipment or software developmental problems. 17 These problems lessened during the later study, although 13% of patients were still excluded. 6 The earlier study clearly reported that histopathologists were unaware of DySIS results prior to assessing biopsies;17 details were unclear for the later study. 6 [Note: DySIS Medical have subsequently confirmed that histopathologists were unaware of DySIS results prior to assessing biopsies for this study also.]

LuViva Advanced Cervical Scan

The only study of LuViva (Flowers et al. , University of Emory School of Medicine, Atlanta, GA, 2011, unpublished) utilised two prototype systems that were referred to as LightTouch, rather than LuViva. The risk of bias assessment was hindered by poor reporting; it was unclear whether patients were enrolled consecutively, and there were uncertainties regarding possible bias arising from the conduct of the tests (most importantly, there was a lack of reporting on the level of training LightTouch assessors had been given). It was unclear whether the standard of care results could possibly have been influenced by knowledge of the biopsy results. The reference standard biopsy procedure was also poorly reported. After seeking further clarification from the manufacturers, it became apparent that only areas seen as being abnormal according to colposcopy were biopsied, with endocervical curettage and/or diagnostic excision biopsy being used for other patients. Applicability concerns regarding the conduct and interpretation of the tests were low for standard of care (where results were interpreted in the knowledge of both cytology and HPV test results) and high for LightTouch (where the cytology and HPV results were not used).

Niris Imaging System

For all three studies of the Niris Imaging System there was an unclear risk of bias in terms of patient selection (none of the studies indicated whether or not patients were recruited consecutively). 31–33 [Note: Imalux Corporation have subsequently confirmed that patients were enrolled consecutively in the study by Liu et al. 32] Similarly, all three studies were at an unclear risk of bias arising from the conduct of the tests (arising particularly from the absence of reporting on the level of training Niris assessors had been given). [Note: Imalux Corporation have subsequently confirmed that in the study by Liu et al. 32 expert colposcopists undertook the colposcopy examination and an OCT expert provided the OCT impression.] The risk of bias relating to the conduct and interpretation of biopsies was low in the two studies reporting that Niris images were anonymised,31,33 but was unclear in the remaining study. 32 [Note: Imalux Corporation have subsequently confirmed that histopathologists were unaware of Niris results prior to assessing biopsies for this study also.] The most recent study was at a high risk of bias for the flow and timing domain, since biopsies were taken only from suspicious areas (meaning false-negative results would not be identified). 33 For the earliest study the risk was low (random biopsies were performed). 31 For the remaining study the risk was unclear (it was unclear whether all recruited patients were included in the analyses). 32 Applicability concerns were high for all three studies regarding the conduct and interpretation of the Niris test. In both the earlier studies the Niris system could not provide cut-offs more specific than being ‘normal’, ‘abnormal’ or ‘indeterminate’ (see the data extraction table for the Escobar et al. study,31 in Appendix 2, for definitions),31–32 whereas for the latest study although results using CIN1+, CIN2+, and CIN3+ as cut-offs were provided, the images were not interpreted during the colposcopic examination. 33 Applicability concerns relating to colposcopy were low for the two earlier studies where the procedure was clearly described,31–32 and unclear for the later study where few details were provided. 33

Synthesis of the included studies

Table 4 displays the participant characteristics and comparator technologies used in the included studies. There was considerable heterogeneity between the included studies, in terms of participant characteristics and comparator technologies used, therefore no quantitative synthesis has been undertaken. The studies have been synthesised, narratively, for each adjunctive technology separately.

TABLE 4 - Summary of participant characteristics and comparator technologies used in the studies

AiC, academic in confidence.

	Study
	Louwers et al., 20116	Zaal et al., unpublished	Soutter et al., 200917	Soutter et al., conference abstract18
	Study
	Flowers et al., unpublished	Flowers and Tadros, conference poster19
	Study
	Gallwas et al., 201133	Liu et al., 201032	Escobar et al., 200631
Participant characteristics	239 women with abnormal cervical cytology or follow-up of a CIN1 or 2 lesion	Subgroup assessment of women in Louwers study6 who had an adequate HPV test result (n = 177)	308 women with abnormal cervical cytology or symptoms suggesting the possibility of cervical neoplasia	Subgroup assessment of women in Soutter study17 in which the grade of the abnormal smear was known (n = 299)
	Prevalence of CIN2+ = 45.2%	Prevalence of CIN2+ = 48%	Prevalence of CIN2+ = 23.4%	Prevalence of CIN2+ in women referred with a low-grade smear = 13.8% Prevalence of CIN2+ in women referred with a high-grade smear = 53.3%
	Analysis: per patient	Analysis: per patient	Analysis: per patient	Analysis: per patient
Comparator technology	Colposcopy using DSI colposcope	Colposcopy using DSI colposcope	Colposcopy using DSI colposcope	Colposcopy using DSI colposcope
	Diagnostic accuracy (CIN2+):	Diagnostic accuracy (CIN2+):	Diagnostic accuracy (CIN2+):	Diagnostic accuracy (CIN2+):
	Sensitivity = 51.9%	Sensitivity = 55%	Sensitivity = 48.6%	Women referred with a low-grade smear:
	Specificity = 81.7%	Specificity = 85%	Specificity = 89.4%
				Sensitivity = 19.4%
				Specificity = 93.3%
				Women referred with a high-grade smear:
				Sensitivity = 72.5%
				Specificity = 68.6%
Participant characteristics	AiC information removed	Subgroup assessment of women in Flowers et al., unpublished study; women aged 16–20 years (n = 245)
		Prevalence of CIN2+ = 18.8%
		Analysis: per patient
Comparator technology	AiC information removed	Current standard of care (consisting of Pap result, HPV and colposcopically directed biopsy)
		Diagnostic accuracy (CIN2+):
		Sensitivity = 80%
Participant characteristics	Women with abnormal cervical cytology (number unknown)	299 women with abnormal cervical cytology or HPV positive for one of the hrHPV types (1237 paired images)	212 women with abnormal cervical cytology or suspicious lesions (1215 images)
	Prevalence of CIN2+ = 52.9%	Prevalence of CIN2+ = 18%	Prevalence of CIN2+ = 15.3%
	Analysis: per image	Analysis: per patient, per lesion and per ‘most severe biopsy per woman’	Analysis: per patient and per lesion
Comparator technology	Conventional colposcopy	Conventional colposcopy	Conventional colposcopy
	Diagnostic accuracy (CIN2+):	Diagnostic accuracy (CIN2+):	Diagnostic accuracy (CIN2+):
	Sensitivity = 99%	Low grade:	Sensitivity = 37.5%
	Specificity = 61%	Sensitivity = 74%	Specificity = 70.6%
		Specificity = 67%
		High grade:
		Sensitivity = 22.6%
		Specificity =  96.3%

DySIS

The main characteristics and results of the included DySIS studies are presented in Table 5; further details are presented in Appendix 2. There were two main studies of the DySIS technology6,17 and two additional subgroup assessments; one subgroup assessment of women according to their hrHPV type [HPV16 vs non-16 hrHPV (Zaal et al. , The VU University Medical Center, Amsterdam, the Netherlands, 2011, unpublished)] and one subgroup assessment18 of women according to the cytology test result (high grade vs low grade).

TABLE 5 - Summary of study characteristics and results: DySIS studies

AGUS, atypical glandular cells of undetermined significance; lrHPV, low-risk human papillomavirus; NR, not reported; PCM, pseudocolour map.

	Study
	Louwers et al., 20116	Zaal et al., unpublished	Soutter et al., 200917	Soutter et al., conference abstract18
Recruitment dates	1 July 2008 to 1 September 2009		August 2004 to July 2005
Number recruited	275		447
Number analysed	239	177	308	299
Patient inclusion criteria	Women with abnormal cervical cytology or follow-up of a CIN1 or 2 lesion	Subgroup assessment of women in Louwers et al.6 study assessed as per protocol, who had an adequate HPV test result	Women with abnormal cervical cytology or symptoms suggesting the possibility of cervical neoplasia	Subgroup assessment of women in Soutter et al.17 for whom the grade of the abnormal smear was known
Patient age	Mean: 36.7 (range 18.7–62.6) years	Median: 33.6 (range 18.7–62.6) years	Median 37 years (upper and lower quartiles 29–46)	NR
Other relevant patient information	Result of last smear:	Result of last smear:	No women with clinically apparent cancer were included	224 women were referred with a low-grade smear, 75 women were referred with a high-grade smear
	normal = 5 (2.1%), borderline or mild dyskaryosis = 153 (64.0%), worse than borderline/mild dyskaryosis = 81 (33.9%)	normal = 4 (2.3%), borderline or mild dyskaryosis = 113 (63.8%), worse than borderline/mild dyskaryosis = 60 (33.9%)	Four women were referred with AGUS cervical cytology result
	hrHPV test: negative = 73 (30.5%), positive = 158 (66.1%), not performed = 8 (3.3%)	hrHPV test: positive = 133 women; 10 lrHPV+, 80 non-16 hrHPV+, 42 hrHPV16+, in one case the typing was inconclusive
Adjunctive technology characteristics	DSI colposcope – DySIS v2.1, with colour-coded map	DSI colposcope – DySIS v2.1, with colour-coded map	Precommercial DySIS model (FPC-03), with PCM	Precommercial DySIS model (FPC-03), with PCM
Colposcopy characteristics	Colposcopic examination using DySIS as a regular video colposcope	Colposcopic examination using DySIS as a regular video colposcope	Colposcopic examination using DySIS as a regular video colposcope	Colposcopic examination using DySIS as a regular video colposcope
Reference standard	Histology result. Biopsies were taken from all suspicious areas identified by the DySIS colour-coded map or colposcopic impression. If both tests evaluated the cervix as normal, one biopsy was taken from the transformation zone at the 12 o'clock position	Histology result. Biopsies were taken from all suspicious areas identified by the DySIS colour-coded map or colposcopic impression. If both tests evaluated the cervix as normal, one biopsy was taken from the transformation zone at the 12 o'clock position	Histology result. Biopsies were taken from all suspicious areas identified by the DySIS colour-coded map or colposcopic impression, and also from sites that did not seem to contain CIN in order to reduce verification bias	Histology result. Biopsies were taken from all suspicious areas identified by the DySIS colour-coded map or colposcopic impression, and also from sites that did not seem to contain CIN, in order to reduce verification bias
Analysis presented	Per patient, ITT	Analysis: per patient, per protocol	Per patient	Per patient
Primary outcome	Histologically confirmed high-grade cervical disease (CIN2+)	Difference in colposcopic impression and histological outcome in women positive for HPV16 (HPV16+) vs women negative for HPV16 but positive for at least one hrHPV type (non-16 hrHPV+)	Incremental DySIS test characteristics over conventional colposcopy, using histology as a reference	The sensitivities and specifcities of DySIS and conventional colposcopy were calculated separately for patients referred with low-grade smears and patients referred with high-grade smears
Diagnostic accuracy results for DySIS adjunctive technology	CIN2+	CIN2+	High-grade disease	Patients referred for a low-grade smear
	Sensitivity = 64.8% (95% CI 55.4 to 73.2)	Total population: Sensitivity = 80% (95% CI 70 to 88)	Sensitivity = 79.2% (95% CI 68.4 to 86.9)	Sensitivity = 77.4%
	Specificity = 77.2%	Specificity = 70.2% (95% CI 61.9 to 77.4)	Specificity = 77% (95% CI 67 to 85)	Specificity = 75.8% (95% CI 70.0 to 80.9)
				Patients referred for a high-grade smear:
	PPV = 64.2% (95% CI 54.9 to 72.6)	Non-16 hrHPV+ population: Sensitivity = 74% (95% CI 57 to 87)	PPV = 50.0% (95% CI 41.0 to 59.0)	Sensitivity = 80.0%
	NPV = 70.8% (95% CI 62.4 to 77.9)	Specificity = 67% (95% CI 50 to 80)	NPV = 92.3% (95% CI 87.6 to 95.3) Accuracy = 76.6%	Specificity = 74.3%
	Accuracy = 67.8% LR+ = 2.18 (95% CI 1.62 to 2.93)	HPV16+ population: Sensitivity = 97% (95% CI 84 to 100)	LR+ = 3.28 (95% CI 2.54 to 4.23)
	LR− = 0.50 (95% CI 0.38 to 0.66)	Specificity = 100% (95% CI 69 to 100)	LR− = 0.28 (95% CI 0.17 to 0.43)
Prevalence = 45.2%		Prevalence = 23.4%
Diagnostic accuracy results for colposcopy	CIN2+	CIN2+	High-grade disease	Patients referred with a low-grade smear
	Sensitivity = 51.9% (95% CI 42.5 to 61.0)	Total population: Sensitivity = 55% (95% CI 44 to 66)	Sensitivity = 48.6% (95% CI 37.4 to 59.9)	Sensitivity = 19.4%
	Specificity = 81.7% (95% CI 74.2 to 87.4)	Specificity = 85% (95% CI 76 to 91)	Specificity = 89.4% (95% CI 84.8 to 92.7)	Specificity = 93.3%
	PPV = 70.0% (95% CI 59.2 to 78.9)	Non-16 hrHPV+ population: Sensitivity = 61% (95% CI 43 to 76)	PPV = 58.3% (95% CI 45.7 to 69.9)	Patients referred with a high-grade smear:
	NPV = 67.3% (95% CI 59.7 to 74.1)	Specificity = 83% (95% CI 69 to 93)	NPV = 85.1% (95% CI 80.1 to 89.0)	Sensitivity = 72.5%
	Accuracy = 68.2%	HPV16+ population: Sensitivity = 53% (95% CI 35 to 71)	Accuracy = 79.9%	Specificity = 68.6%
	LR+ = 2.83 (95% CI 1.89 to 4.24)		LR+ = 4.59 (95% CI 2.96 to 7.13)
	LR− = 0.59 (95% CI 0.48 to 0.73)	Specificity = 90% (95% CI 55 to 100)	LR− = 0.58 (95% CI 0.46 to 0.72)
	Prevalence = 45.2%		Prevalence = 23.4%
Diagnostic accuracy results for DySIS adjunctive technology and colposcopy combined	Sensitivity = 79.6% (95% CI 71.1 to 86.1)
	Specificity = 62.6% (95% CI 54.1 to 70.4)
	PPV = 63.7% (95% CI 55.3 to 71.3)
	NPV = 78.8% (95% CI 70.0 to 85.6)
	Accuracy = 70.3%
	LR+ = 2.13 (95% CI 1.67 to 2.71)
	LR− = 0.33 (95% CI 0.22 to 0.48)
	Prevalence = 45.2%
Adverse effects	No adverse events were reported during the study period	NR	No adverse events were reported	NR
Patient satisfaction	DSI colposcopy, compared with conventional colposcopy, was no extra burden for the majority of women	NR	NR	NR

The participants in the main studies were similar: women referred for colposcopy with an abnormal cervical cytology result or follow-up of a CIN1 or CIN2 lesion,6 or women referred with an abnormal cervical cytology result or symptoms suggesting the possibility of cervical neoplasia. 17 However, the prevalence of CIN2+ was considerably higher in the study by Louwers et al. ,6 at 45%, than in the study by Soutter et al. 17 (23%). The average age of participants was 37 years in both of the main studies. The Louwers et al. 6 results presented below are those for the ‘intention-to-treat’ (ITT) cohort of patients, rather than the ‘according-to-protocol’ (ATP) cohort, from which 56 women were excluded as their management did not strictly adhere to the protocol. 6 Results for the ATP cohort are reported in Appendix 2.

The DySIS technology used in the earlier study by Soutter et al. 17 was a precommercial model (FPC-03), which had some technical problems relating to the software, speculum and a batch of faulty disposable nozzles, leading to the exclusion of a large proportion of participants from the analyses. 17 DySIS v2.1 was used in the later study by Louwers et al. ; 6 therefore, this study is the most relevant for clinical practice. Both studies used the DySIS colposcope as a regular video colposcope as the comparator technology, and histology result was the reference standard. All patients underwent both DySIS colposcopy and the comparator colposcopic examination during the same appointment.

The sensitivity of DySIS was higher than that of conventional colposcopy (using the DySIS colposcope as a regular video colposcope) for distinguishing between normal or low-grade (CIN 0–1) and high-grade (CIN2+) disease: 64.8% compared with 51.9% in the study by Louwers et al. 6 and 79.2% compared with 48.6% in the study by Soutter et al. 17 However, the specificity was lower with DySIS; 70.2% compared with 81.7% in the study by Louwers et al. 6 and 75.8% compared with 89.4% in the study by Soutter et al. 17 The sensitivity and specificity of DySIS (the DSI colour-coded map) combined with conventional colposcopy were 79.6% and 62.6% respectively, compared with 51.9% and 81.7% for conventional colposcopy alone. 6 The differences in sensitivity and specificity between DySIS and conventional colposcopy and between DySIS combined with conventional colposcopy and conventional colposcopy alone were statistically significant (asymptotic McNemar test in the study by Louwers et al. , 6 Fisher's exact two-sided test in the study by Soutter et al. 17).

In the study by Louwers et al. 6 the overall diagnostic accuracy of DySIS was similar to that of conventional colposcopy: 67.8% compared with 68.2%. In the study by Soutter et al. 17 the overall diagnostic accuracy of DySIS was slightly lower than that of conventional colposcopy: 76.6% compared with 79.9%. The accuracy of DySIS combined with conventional colposcopy was also assessed using data from the study by Louwers et al. ,6 and was similar to that of conventional colposcopy alone, 70.3%.

In a subgroup assessment of women referred with a high-grade cytology test result, both sensitivity and specificity were higher with DySIS than conventional colposcopy; 80% compared with 72.5% for sensitivity and 74.3% compared with 68.6% for specificity, although this was based on a subgroup assessment of just 75 women. 18 In a subgroup of women referred with a low-grade cytology test result, sensitivity was higher with DySIS (77.4% compared with 19.4%), but specificity was lower (77.2% compared with 93.3%), based on a subgroup assessment of 224 women. 18

In a subgroup assessment of women with hrHPV16, both sensitivity and specificity were higher with DySIS than conventional colposcopy: 97% compared with 53% for sensitivity and 100% compared with 90% for specificity, although this was based on a subgroup assessment of just 42 women. In the subgroup of women with non-16 hrHPV, sensitivity was higher with DySIS (74% vs 61%), but specificity was lower (67% vs 83%), based on a subgroup assessment of 80 women (Zaal et al. , unpublished).

The two main studies stated that no adverse events were reported. 6,17

The study by Louwers et al. 6 assessed patient satisfaction using a questionnaire; the majority of women reported that DySIS was no extra burden compared with conventional colposcopy.

LuViva Advanced Cervical Scan

The main characteristics and results of the included LuViva study are presented in Table 6; further details are presented in Appendix 2. There was one main study of LuViva (Flowers et al. , unpublished) and one additional subgroup assessment of women aged 16–20 years. 19 However, women in England are not invited for cervical screening under the NHS Cervical Screening Programme until the age of 25 years;2 therefore, the subgroup population is not of direct relevance to this assessment.

TABLE 6 - Summary of study characteristics and results: LuViva studies

MHS, multimodal hyperspectroscopy; NR, not reported.

	Study
	Flowers et al., unpublished	Flowers and Tadros, conference poster19
Recruitment dates	AiC information has been removed
Number recruited
Number analysed		245
Patient inclusion criteria		Subgroup assessment of women in Flowers et al., unpublished study; women aged 16–20 years
Patient age		16–20 years
Other relevant patient information
Adjunctive technology characteristics		MHS LightTouch
Comparator technology characteristics		Current standard of care (consisting of Pap result, HPV and colposcopically directed biopsy)
Reference standard		Histology result and clinical follow-up
Analysis presented		Per patient
Primary outcome		Prevalence of CIN2+ or worse disease in women of < 21 years and performance of MHS in this population
Diagnostic accuracy results for LuViva adjunctive technology		CIN2+
		Sensitivity = 91.3% (95% CI 79.7 to 96.6)
		Specificity = 28.6% (95% CI 22.8 to 35.3)
		PPV = 22.8% (95% CI 17.4 to 29.4)
		NPV = 93.4% (95% CI 84.3 to 97.4)
		Accuracy = 40.4%
		LR+ = 1.28 (95% CI 1.13 to 1.45)
		LR− = 0.30 (95% CI 0.12 to 0.79)
		Prevalence = 18.8%
Diagnostic accuracy results for the current standard of care		CIN2+
		Sensitivity = 80%
Adverse effects		NR
Patient satisfaction		NR

The main study of LuViva was reported in an academic-in-confidence (AiC) unpublished report; therefore, the data cannot be presented in this report.

The name of the technology has been changed since the study was conducted; at the time of the study the LuViva Advanced Cervical Scan was called LightTouch. The comparator used in the study was the ‘current standard of care’, consisting of the cytology test result, HPV test result and colposcopically directed biopsy. Histology result was the reference standard; however, this was based on biopsy for abnormal-looking areas, and endocervical curettage when no lesion was seen on colposcopy [although if patients had been referred with low-grade squamous intraepithelial lesion (LSIL), atypical squamous cells with possible high-grade squamous intraepithelial lesion (ASC-H) or high-grade squamous intraepithelial lesion (HSIL), diagnostic excision biopsy was performed]. In addition, around half of the patients had 2-year clinical follow-up. All patients underwent the LightTouch scan during the standard colposcopy appointment.

Niris Imaging System

The main characteristics and results of the included Niris studies are presented in Table 7; further details are presented in Appendix 2. There were three studies of the Niris Imaging System. 31–33

TABLE 7 - Summary of study characteristics and results: Niris studies

ASCUS, atypical squamous cells with uncertain significance; NR, not reported.

	Study
	Gallwas et al., 201133	Liu et al., 201032	Escobar et al., 200631
Recruitment dates	July 2008 to May 2010	NR	NR
Number recruited	Unclear, although 1375 images were taken from 120 women (1165 images were from unsuspicious areas, and 210 were compared with histology)	Unclear	220
Number analysed	210 images (number of women unknown)	299 women (1237 paired diagnoses)	212 (1215 images)
Patient inclusion criteria	Women with abnormal cervical cytology	Women with abnormal cervical cytology or a positive test for one of the high-risk types of HPV	Women with abnormal cervical cytology or suspicious lesions
Patient age	Mean: 31.1 (range 18–46) years	Median: 36.7 (range 19.2–67.9) years	Mean: 35.5 (range 18–80) years
Other relevant patient information	Result of last smear	10% of women were menopausal	Result of last smear
	PAP II, 19; PAP IIW, 14; PAP III, 5; PAP IIID, 44; PAP IVA, 32; PAP IVB, 5; PAP V, 1		48 (23%) had ASCUS, 142 (67%) had LSIL, 22 (10%) had HSIL
	hrHPV test		189 were premenopausal and 23 were postmenopausal
	93 women tested positive
Adjunctive technology characteristics	Colposcopy-guided OCT using the Niris Imaging system	Niris Imaging System	Imalux OCT device
Colposcopy characteristics	Conventional colposcopy	Conventional colposcopy	Conventional colposcopy
Reference standard	Histology result. Biopsies were taken from suspicious areas identified using OCT. (Biopsy procedure details were unclear for the colposcopy assessment.)	Histology result. Biopsies were taken at all positive areas and at the 2, 4, 8 and 10 o'clock positions at the squamocolumnar junction. Endocervical curettage was also performed on every patient	Histology result. Biopsies were taken at all positive areas and at the 2, 4, 8 and 10 o'clock positions at the squamocolumnar junction. Endocervical curettage was also performed on every patient
Analysis presented	Per image	Per patient, per lesion and per ‘most severe biopsy per woman’	Per patient and per lesion
Primary outcome	CIN using cut-offs at CIN1+, CIN2+ and CIN3+	CIN using cut-offs at indeterminate or abnormal	CIN using cut-offs at indeterminate or abnormal
Diagnostic accuracy results for Niris adjunctive technology	CIN1+	Per-patient analysis	Per-patient analysis
	Sensitivity = 97.9% (95% CI 94.1 to 99.3)	CIN2+ Indeterminate/abnormal	CIN2+ Indeterminate/abnormal
	Specificity = 39.1% (95% CI 28.1 to 51.3)	Sensitivity = 45.3% (95% CI 32.7 to 58.5)	Sensitivity = 93.8% (95% CI 79.9 to 98.3)
	PPV = 78.6% (95% CI 72.1 to 83.9)	Specificity = 86.1% (95% CI 81.2 to 89.9)	Specificity = 10.7% (95% CI 7.0 to 16.2)
	NPV = 89.3% (95% CI 72.8 to 96.3)	PPV = 41.4% (95% CI 29.6 to 54.2)	PPV = 16.0% (95% CI 11.4 to 21.9)
	Accuracy = 80.0%	NPV = 87.9% (95% CI 83.2 to 91.5)	NPV = 90.5% (95% CI 71.1 to 97.3)
	LR+ = 1.61(95% CI 1.32 to 1.96)	Accuracy = 78.9%	Accuracy = 23.4%
	LR− = 0.05(95% CI 0.02 to 0.17)	LR+ = 3.26(95% CI 2.12 to 5.02)	LR+ = 1.05(95% CI 0.95 to 1.16)
	Prevalence = 69.5%	LR− = 0.64(95% CI 0.50 to 0.82)	LR− = 0.58(95% CI 0.14 to 2.38)
	CIN2+
	Sensitivity = 86.5% (95% CI 78.9 to 91.6)	Prevalence = 17.8%	Prevalence = 15.3%
		Abnormal	Abnormal
	Specificity = 63.6% (95% CI 53.8 to 72.4)	Sensitivity = 32.1% (95% CI 21.1 to 45.5)	Sensitivity = 56.3% (95% CI 39.3 to 71.8)
	PPV = 72.7% (95% CI 64.6 to 79.6)	Specificity = 93.1% (95% CI 89.2 to 95.6)	Specificity = 59.3% (95% CI 52.0 to 66.3)
	NPV = 80.8% (95% CI 70.7 to 88.0)	PPV = 50% (95% CI 34.1 to 65.9)	PPV = 20.0% (95% CI 13.0 to 29.4)
	Accuracy = 75.7%	NPV = 86.4% (95% CI 81.7 to 90.0)	NPV = 88.2% (95% CI 81.2 to 92.9)
	LR+ = 2.38(95% CI 1.81 to 3.12)	Accuracy = 82.2%	Accuracy = 58.9%
	LR− = 0.21(95% CI 0.13 to 0.35)	LR+ = 4.62(95% CI 2.53 to 8.45)	LR+ = 1.38(95% CI 0.97 to 1.97)
	Prevalence = 52.9%	LR− = 0.73(95% CI 0.61 to 0.88)	LR− = 0.74(95% CI 0.49 to 1.11)
	CIN3+
	Sensitivity = 87.2% (95% CI 78.0 to 92.9)	Prevalence = 17.8%	Prevalence = 15.3%
	Specificity = 81.1% (95% CI 73.5 to 86.8)
	PPV = 73.1% (95% CI 63.3 to 81.1)
	NPV = 91.5% (95% CI 85.0 to 95.3)
	Accuracy = 83.3%
	LR+ = 4.60(95% CI 3.20 to 6.62)
	LR− = 0.16(95% CI 0.09 to 0.28)
	Prevalence = 37.1%
Diagnostic accuracy results for colposcopy	CIN1+	Per-patient analysis	CIN2+
	Sensitivity = 99%	CIN2+	Sensitivity = 37.5% (95% CI 22.9 to 54.7)
	Specificity = 19%	Low grade
	CIN2+	Sensitivity = 74% (95% CI 60 to 84)	Specificity = 70.6% (95% CI 63.5 to 76.8)
	Sensitivity = 99%
	Specificity = 61%	Specificity = 67% (95% CI 61 to 73)	PPV = 18.8% (95% CI 11.1 to 30.0)
	CIN3+	High grade	NPV = 86.2% (95% CI 79.7 to 90.9)
	Sensitivity = 78%
	Specificity = 74%	Sensitivity = 22.6% (95% CI 13.5 to 35.5)	Accuracy = 65.6%
		Specificity = 96.3% (95% CI 93.2 to 98.1)	LR+ = 1.28(95% CI 0.77 to 2.11)
		PPV = 57.1% (95% CI 36.5 to 75.5)	LR− = 0.89(95% CI 0.67 to 1.18)
		NPV = 85.3% (95% CI 80.6 to 88.9)	Prevalence = 15.3%
		Accuracy = 83.3%
		LR+ = 6.19(95% CI 2.75 to 13.94)
		LR− = 0.80(95% CI 0.69 to 0.93)
		Prevalence = 17.7%
Adverse effects	NR	NR	NR
Patient satisfaction	NR	NR	NR

The participants were similar in all three studies: women referred for colposcopy with an abnormal cervical cytology result,33 women referred with an abnormal cervical cytology result or hrHPV,32 or women referred with an abnormal cervical cytology result or suspicious lesions. 31 However, the prevalence of CIN2+ was considerably higher in the study by Gallwas et al. ,33 at 53%, than in the study by Liu et al. 32 (18%) and the study by Escobar et al. (15%). 31 The average age of participants in the studies was between 31 and 37 years.

In the study by Gallwas et al. ,33 Niris images were evaluated as normal, inflammation, CIN1, CIN2, CIN3 or squamous carcinoma. 33 In the earlier study by Liu et al. ,32 Niris images were evaluated as normal, indeterminate or abnormal. 32 In the earliest study by Escobar et al. ,31 the system was referred to as the Imalux OCT device, it had different technical specifications to the other two studies,32,33 and this study31 also evaluated images as normal, indeterminate or abnormal. Images were evaluated as being normal if a well-organised, simple two-layer structure with a sharp interface between the surface epithelium and underlying stromal layer was seen. Images were evaluated as being abnormal if the tissue was unstructured with no apparent interface present. Images were evaluated as being indeterminate if irregularities on the images suggested artefacts or physiological alterations and did not meet criteria for normal or abnormal. The study by Gallwas et al. 33 is the most relevant for clinical practice because of the cut-offs used for categorising images.

All three studies31–33 used conventional colposcopy as the comparator technology, and histology result was the reference standard. However, biopsies were taken from only the suspicious areas in the study by Gallwas et al. ,33 therefore the results from this study are unreliable. All patients underwent OCT imaging using the Niris technology during the standard colposcopy appointment.

In the study by Gallwas et al. 33 the sensitivity of Niris was lower than that of conventional colposcopy for detecting CIN2+ disease: 86.5% and 99%, respectively. However, the sensitivity of 99% for conventional colposcopy is not representative of colposcopy in practice. In this study, biopsies for reference standard assessment were taken only from suspicious areas; thus, false-negative results would not have been detected, resulting in a falsely increased sensitivity result. 33 Therefore, the results from this study are unreliable. The specificity of Niris was similar to that of colposcopy: 63.6% and 61% respectively. The overall accuracy of Niris was 75.7%; it was not possible to calculate overall accuracy for conventional colposcopy. However, the lack of reference standard assessment of patients for whom no suspicious areas were identified also affects the specificity and overall accuracy results.

For detecting CIN1+ disease, the sensitivity of Niris was 97.9% compared with 99% for colposcopy, specificity was 39.1% for Niris and 19% for colposcopy, and accuracy was 80% for Niris. For detecting CIN3+, disease the sensitivity of Niris was 87.2% compared with 78% for colposcopy, specificity was 81.1% for Niris and 74% for colposcopy, and accuracy was 83.3% for Niris. 33 Again, these results are unreliable, owing to biopsies for reference standard assessment being taken only from suspicious areas.

In the study by Liu et al. ,6 the sensitivity of Niris was higher than that of conventional colposcopy for distinguishing between normal/indeterminate and abnormal lesions: 32.1% compared with 22.6%. 32 The specificity of Niris was slightly lower than that of conventional colposcopy: 93.1% compared with 96.3%. The overall diagnostic accuracy of Niris was similar to that of conventional colposcopy: 82.2% compared with 83.3% for conventional colposcopy (for determining ‘high-grade’ lesions with colposcopy).

For distinguishing between normal and indeterminate/abnormal lesions, the sensitivity of Niris was lower than that of colposcopy; 45.3% compared with 74% for conventional colposcopy (for determining ‘low-grade’ lesions with colposcopy). The specificity of Niris was higher than that of colposcopy: 86.1% compared with 67% for conventional colposcopy. The overall diagnostic accuracy of Niris was 78.9%; overall accuracy was not reported for conventional colposcopy.

In the study by Escobar et al. ,31 the sensitivity of Niris was higher than that of colposcopy for distinguishing between normal/indeterminate and abnormal lesions: 56.3% compared with 37.5%. However, the specificity of Niris was lower than that of colposcopy: 59.3% compared with 70.6%. The overall diagnostic accuracy of Niris was lower than that of conventional colposcopy: 58.9% compared with 65.6% for conventional colposcopy.

For distinguishing between normal and indeterminate/abnormal lesions, the sensitivity of Niris was 93.8% and the specificity was 10.7%. The overall diagnostic accuracy of Niris was 23.4%.

Summary of results for the most clinically relevant studies

Table 8 summarises the diagnostic accuracy results for the three most clinically relevant studies: the study of the most recent model of the DySIS technology by Louwers et al. ,6 the study of the LuViva Advanced Cervical Scan (under its former name of LightTouch) by Flowers et al. (unpublished) and the study of the most recent model of the Niris Imaging System by Gallwas et al. 33 (the only Niris study to report results using a CIN2 cut-off).

TABLE 8 - Summary of diagnostic accuracy results for determining CIN2+: results of the most clinically relevant studies

NR, not reported (data not available).

Study
Louwers et al., 20116			Flowers et al., unpublished			Gallwas et al., 201133
DySIS alone	DySIS + conventional colposcopy	Conventional colposcopy	AiC information has been removed	AiC information has been removed	AiC information has been removed	Niris Imaging System	Conventional colposcopy
Sensitivity = 64.8% (95% CI 55.4 to 73.2)	Sensitivity = 79.6% (95% CI 71.1 to 86.1)	Sensitivity = 51.9% (95% CI 42.5 to 61.0)				Sensitivity = 86.5% (95% CI 78.9 to 91.6)	Sensitivity = 99%
Specificity = 70.2% (95% CI 61.9 to 77.4)	Specificity = 62.6% (95% CI 54.1 to 70.4)	Specificity = 81.7% (95% CI 74.2 to 87.4)				Specificity = 63.6% (95% CI 53.8 to 72.4)	Specificity = 61%
Accuracy = 67.8%	Accuracy = 70.3%	Accuracy = 68.2%				Accuracy = 75.7%	NR
LR+ = 2.18 (95% CI 1.62 to 2.93)	LR+ = 2.13 (95% CI 1.67 to 2.71)	LR+ = 2.83 (95% CI 1.89 to 4.24)				LR+ = 2.38 (95% CI 1.81 to 3.12)	NR
LR− = 0.50 (95% CI 0.38 to 0.66)	LR− = 0.33 (95% CI 0.22 to 0.48)	LR− = 0.59 (95% CI 0.48 to 0.73)				LR− = 0.21 (95% CI 0.13 to 0.35)	NR
Prevalence = 45.2%	Prevalence = 45.2%	Prevalence = 45.2%				Prevalence = 52.9%	NR

The results of the studies suggest that the sensitivity of the adjunctive technologies is higher for DySIS, DySIS plus conventional colposcopy, and LuViva than conventional colposcopy alone, and for LuViva sensitivity is also higher than the current standard of care (consisting of the cytology test result, HPV test result and colposcopically directed biopsy). For DySIS the specificity is lower for DySIS and DySIS plus conventional colposcopy than conventional colposcopy alone; resulting in an overall diagnostic accuracy similar to that of conventional colposcopy alone. The specificity of LuViva is lower than that of colposcopy alone, although the specificity of LuViva cannot be compared against the standard of care, as the relevant data were not reported. The sensitivity of Niris was found to be lower than that of conventional colposcopy and the specificity of Niris appears to be similar to that of conventional colposcopy. However, the results from this study33 are unreliable because biopsies for reference standard assessment were taken only from suspicious areas.

Discussion

Interpretation of study results and quality assessment

The systematic review identified a limited amount of evidence on the three adjunctive colposcopy technologies: two studies of the DySIS colposcope,6,17 one study of the LuViva Advanced Cervical Scan (Flowers et al. , unpublished) and three studies of the Niris Imaging System. 31–33

Both studies of the DySIS colposcope6,17 found that the sensitivity of DySIS was statistically significantly higher than that of conventional colposcopy for identifying CIN2+ disease, although specificity was significantly lower with DySIS. 6,17 Taking both sensitivity and specificity into account, the overall diagnostic accuracy was similar to that of conventional colposcopy. The LRs indicated that DySIS was only a fair predictor of how much a test result will change the (pre-test) odds of having CIN2+. The combination of the DSI colour-coded map and conventional colposcopy resulted in the highest result for sensitivity, although specificity was lowered further. 6 The authors did not define what was meant by ‘DSI colour-coded map and conventional colposcopy combined’, although it appears that patients were considered positive if either the DSI colour-coded map or conventional colposcopy were positive. [Note: DySIS Medical has subsequently confirmed that this assumption is correct.] It appears that this would be workable in clinical practice, with the colposcopist performing the examination using DySIS as a conventional colposcope, followed by assessment using the DSI colour-coded map.

The sensitivity of DySIS remained high in the subgroup of women referred for colposcopy with a low-grade cytology test result, whereas the sensitivity of conventional colposcopy was low in this subgroup of women. 18

In a subgroup analysis, the sensitivity of DySIS was higher in women with hrHPV16 than in women with non-16 hrHPV. Therefore, when the prevalence of hrHPV16 reduces in the screening population, as females who have been vaccinated against this strain of HPV reach the age for cervical cancer screening, DySIS sensitivity will reduce. However, the sensitivity of DySIS was still higher than that of conventional colposcopy in women with non-16 hrHPV, as well as women with hrHPV16 (Zaal et al. , unpublished).

The study of the LuViva Advanced Cervical Scan (Flowers et al., unpublished) reported higher sensitivity than the standard of care (consisting of the cytology test result, HPV test result and colposcopically directed biopsy) for identifying CIN2+ disease, although the specificity of LuViva was low. The authors of the study (Flowers et al. , unpublished) stated that the study evaluated the potential of the new technology to effectively triage women at risk for moderate and high-grade dysplasia rather than as an adjunct to colposcopy.

The most recent study33 of the Niris Imaging System was the most relevant for clinical practice because of the cut-offs used for categorising patients. This study reported a lower sensitivity for Niris for identifying CIN2+ disease than with conventional colposcopy, and a similar specificity.

From the results of our quality assessment, it appears that only the results of the DySIS study by Louwers et al. 6 can be interpreted as being both reliable and clinically applicable. The only concern with this study6 was whether conventional colposcopy was represented appropriately, although the authors pointed out that the results were similar to other studies evaluating conventional colposcopy. The authors also noted a limitation in that a second DySIS examination could not be performed after the first (the acetowhitening effect can last up to 45 minutes, which can interfere with DySIS measurements). This would restrict the use of DySIS when a repeat examination was required e.g. when only part of the transformation zone could be visualised in the first examination. For most of the other studies, the lack of clear reporting meant that the risk of bias was often ‘unclear’, although the reported issues that cast doubt on their reliability or relevance included a high dropout rate;17 use of different reference standard procedures across the population (Flowers et al. , unpublished); lack of a clinically relevant cut-off;31–32 reference standard not performed for all patients; and results not provided in real time. 33

Test accuracy may be overestimated in studies at risk of bias. 40 The STARD (STAndards for the Reporting of Diagnostic accuracy studies) statement was produced with the aim of improving the quality of reporting of diagnostic accuracy studies;41–42 although it appears so far to have had a minimal tangible effect on reporting quality, even in papers published in journals which explicitly endorse the STARD statement. 43–44 This has led to a call for authors, editors and peer reviewers to adhere to, and enforce, STARD statement guidelines. 44

Strengths and limitations

This systematic review addressed a clear research question using predefined inclusion criteria. Comprehensive literature searches were performed to locate all relevant published and unpublished studies without any language restrictions, thereby minimising the potential for publication bias and language bias. Hand-searching was also performed in order to identify additional relevant studies and the manufacturers were asked whether any other potentially relevant studies were available. Study selection was undertaken independently by two reviewers. Data extraction and quality assessment were checked by a second reviewer to minimise the potential for reviewer bias or error. The authors of studies were contacted, when necessary, for clarification of study details and for additional diagnostic accuracy data. Study quality assessment was undertaken using a validated checklist for diagnostic studies, with additional review-specific quality assessment items added. We are therefore confident that we have identified all relevant evidence and have appropriately critically appraised and synthesised the included studies.

However, the studies included in the review were clinically and methodologically heterogeneous, which meant that statistical pooling of results was not appropriate. The ways in which the studies varied, and the implications of the variation, are discussed below.

Participants

Women in England are not invited for cervical screening under the NHS Cervical Screening Programme until the age of 25 years. The studies included in the review included women aged 18 years, so the youngest women included in the studies would not be seen in practice. The mean or median age of participants was > 35 years for all studies except the study by Gallwas et al. 33 (mean age 31 years) and the study by Flowers et al. (unpublished), in which the mean age was not reported. From the data presented in the study by Flowers et al. (unpublished) it was apparent that around one-third of the participants were aged < 25 years. This limits the applicability of this study's results to an NHS setting.

The prevalence of CIN2+ varied considerably between studies, demonstrating heterogeneity between participants in the included studies. The reasons for this variation are not clear, although the inclusion criteria differed slightly between studies, and there was some variation in the setting of the included studies; the studies were conducted in the Netherlands, England and Greece, the USA, Germany, China, and the USA and the Dominican Republic. The two studies with the highest prevalence of CIN2+ were conducted in the Netherlands and Germany. 6,33

The implications of this variation in prevalence of CIN2+ on the results are that the sensitivity may be reduced in studies with a lower prevalence of CIN2+, as colposcopists who are less familiar with the characteristics of CIN2+ may be less able to recognise them on colposcopic examination.

Intervention

Some studies of both DySIS17 and the Niris Imaging System31,32 related to earlier versions of the technology, meaning that their results were of limited clinical value and/or more prone to bias; in addition, the earlier Niris studies31,32 did not use clinically appropriate cut-offs for categorising patients. In clinical practice, patient management decisions are made based on the colposcopist's impression of CIN grade and the reason for referral for colposcopy from the NHS Cervical Screening Programme.

The authors of the study of the LuViva Advanced Cervical Scan (Flowers et al. , unpublished) suggest that the intended use of the technology is to triage women for colposcopy, rather than as an adjunct to colposcopy. Therefore, this technology has a different place in the care pathway from the other technologies included in this assessment.

In clinical practice, colposcopists have access to cytology test results and are aware of other patient characteristics, such as age, etc. However, in most of the included studies it was unclear whether these data were available to colposcopists when interpreting the results of the new technology. Only two studies17,31 reported that cytology test results were available when interpreting the results of the new technology. [Note: Based on subsequent information from DySIS Medical and Imalux Corporation, four studies6,17,31,32 reported that cytology test results were available when interpreting the results of the new technology.]

Comparator

The comparators used varied across the technologies. In the studies of DySIS,6,17 the comparator was video colposcopy using the DySIS colposcope, rather than the conventional colposcopy methods and equipment used in the NHS. Therefore, the accuracy of conventional colposcopy in these studies may not be an accurate reflection of current NHS practice. In the study of the LuViva Advanced Cervical Scan (Flowers et al. , unpublished) the comparator was ‘standard of care’, which consisted of the cytology test result, HPV test result and colposcopically directed biopsy. The comparator used in the studies31–33 of the Niris Imaging System was conventional colposcopy.

The accuracy of colposcopy varied considerably between studies, which may reflect differences in colposcopic examination and biopsy procedures, the expertise of colposcopists and also the differences in prevalence of CIN2+ between studies.

Reference standard

The reference standard was histology for all of the included studies, although in the study of the LuViva Advanced Cervical Scan (Flowers et al. , unpublished) some patients also had 2-year clinical follow-up. Both of the studies of DySIS6,17 used random biopsies to assess negative colposcopy results. In the study of the LuViva Advanced Cervical Scan the histology result was based on biopsy for abnormal-looking areas and endocervical curettage when no lesion was seen on colposcopy (although some women may have had diagnostic excision biopsy), which has implications for the reliability of the reference standard. In the most clinically relevant study of the Niris Imaging System by Gallwas et al. ,33 biopsies were taken only from suspicious areas, so false-negative results would not be identified.

It is difficult to obtain a definite reference standard for patients with negative index test results; random biopsies are likely to be the most accurate, although they may miss diseased areas. Long-term follow-up may result in high dropout rates and the possibility that disease spotted at long-term follow-up began in the interim period, i.e. may not have been present at initial assessment. The LuViva study (Flowers et al. , unpublished) followed up around only half of the participants at 2 years, although the reasons for participants not receiving a 2-year follow-up were not explicit.

Outcomes

In order to re-calculate and confirm the reported results, full 2 × 2 data were required. However, these data were reported for only the adjunctive technology and the comparator in the two studies of DySIS. 6,17 Full 2 × 2 data were provided by the study authors for two further studies, on request. 31–32

All except one of the studies reported results ‘per patient’; the study of the Niris Imaging System by Gallwas et al. 33 reported results ‘per image’, meaning that not all of the data were independent observations as some women may have contributed multiple images. Furthermore, in this study it was unclear whether all participants contributed to the analysis.

As discussed earlier, in clinical practice patient management decisions are made based on the colposcopist's diagnosis and the reason for referral for colposcopy from the NHS Cervical Screening Programme. The majority of studies used the cut-off of CIN2+ for determining the sensitivity and specificity of the adjunctive technology. However, management guidelines are different for women with CIN1 from women with no CIN. 5 Therefore, the ability to distinguish between normal, CIN1, CIN2 and CIN3 is important in practice.

Conclusions

The DySIS colposcope is significantly more sensitive than conventional colposcopy for identifying CIN2+ disease, although specificity is significantly lower. The combination of the DSI colour-coded map and conventional colposcopy results in the highest sensitivity, although specificity is lowered further. Based on study quality assessment, these results are likely to be reliable.

The study of the LuViva Advanced Cervical Scan (Flowers et al. , unpublished) and the clinically relevant study of the Niris Imaging System33 contain significant biases and uncertainties; therefore, their results can not be relied on. In addition, the authors of the study of LuViva (Flowers et al. , unpublished) suggest that the intended use of the technology is to triage women for colposcopy, rather than as an adjunct to colposcopy.

Description of decision-analytic model

Overview

A decision-analytic model was developed to assess the cost-effectiveness of the three devices (DySIS, LuViva Advanced Cervical Scan and Niris Imaging System). It compared these with standard colposcopy for examination of the uterine cervix, for the detection of cancerous and precancerous cervical tissue in patients referred for colposcopy through the NHS Cervical Screening Programme. As a result of the weaknesses in the studies of Niris and LuViva (discussed in detail in Systematic review of clinical effectiveness), these devices were excluded in the base-case analysis. The analysis adopted the perspective of the UK NHS. The model provides a framework for the synthesis of data from the review of clinical effectiveness (see Systematic review of clinical effectiveness) and other relevant parameters.

Outcomes in the model are expressed in terms of quality-adjusted life-years (QALYs). Costs are evaluated from the perspective of the NHS and Personal Social Services, expressed in UK pounds sterling at 2011 prices. Both costs and outcomes are discounted using an annual discount rate of 3.5%, in line with current methods guidelines. 53 All stages of the work were informed by discussion with our clinical advisor and members of the specialist committee to provide feedback on specific aspects of the analysis, such as the modelling approach, data inputs and assumptions.

The following sections outline the structure of the model and provide details of the key assumptions and data sources used to populate the model.

Modelling approach

The decision-analytic model involved two stages. First, a decision tree to model the diagnostic and treatment pathways for patients referred to colposcopy from the NHS Cervical Screening Programme. Second, a Markov model, which simulates the natural history of patients and captures future cytological screening and referrals to colposcopy, to estimate the outcomes of the initial diagnosis and treatment choices. The decision tree has been developed for this appraisal, whereas the Markov model is based on a revised version of the model used by Hadwin et al. ,51 henceforth referred to as the Sheffield model. 52

Diagnostic and treatment decision tree

The diagnostic and treatment decision tree was developed to model the short-term diagnostic and treatment pathways and the outcomes of patients referred to colposcopy from the NHS Cervical Screening Programme. Patients are referred for colposcopy through the NHS Cervical Screening Programme for a variety of reasons (e.g. moderate or severe cytology). 2 For any given referral reason there is a distribution of the true underlying health states (this is discussed in further detail in Model inputs). The diagnostic treatment decision tree first allocates patients to their true underlying health state, with the distribution being dependent on their reason for referral, and then sends them down the diagnostic and treatment pathways dependent on probabilities for diagnostic accuracy, treatment and treatment effectiveness. Examples of parts of the decision trees are shown in Figure 2, showing the distribution of true underlying health state by reason for referral; Figure 3, showing the diagnostic and treatment pathways for a patient with a true underlying health state of CIN1; and Figure 4, showing the diagnostic and treatment pathways for a patient with cervical cancer.

FIGURE 2.

Distribution of true underlying health state by reason for referral.

FIGURE 3.

Diagnostic decision tree for patients with CIN1.

FIGURE 4.

Diagnostic decision tree for patients with cancer.

The decision tree captures the initial diagnosis of the patient by the colposcopist and any subsequent treatments or screening options based on their diagnosis at colposcopy and the reason for referral from the NHS Cervical Screening Programme. The effectiveness of any treatment is based on the true underlying condition of the patient. Treatment and screening options available include:

1. return to the NHS Cervical Screening Programme

2. refer for rescreen at 6 months: patients can be referred for rescreen with a cytological smear and HPV test, or can be referred for rescreen with colposcopy or adjunct

3. a diagnostic biopsy

4. a treatment biopsy

5. a treatment biopsy followed by cancer treatment.

Figure 3 shows the diagnostic and treatment decision tree for a patient whose true health state is CIN1. The patient receives an initial diagnosis by the colposcopist and can be incorrectly identified as clear, correctly identified as CIN1, incorrectly identified as CIN2/3 or incorrectly identified as having invasive cervical cancer (patients are not diagnosed as HPV+ by the colposcopist). The initial diagnosis is based on the diagnostic accuracy of the device (this is discussed in more detail in Model inputs). Following identification, the patient will be assigned to one of the five treatment and screening options discussed above (although, in Figure 3, option 5 is excluded as it is assumed that no patient with CIN1 can receive cancer therapy incorrectly, as histology resulting from the treatment biopsy is assumed to be 100% sensitive and specific). The patient's allocation to the treatment/screening option is based on the colposcopist's diagnosis and the reason for their referral for colposcopy from the NHS Cervical Screening Programme (this is discussed in more detail in Model inputs).

Patients referred for rescreen at 6 months or returned to the NHS Cervical Screening Programme without receiving treatment enter the Markov model (described in detail in Description of decision-analytic model) in the same underlying health state in which they entered the diagnostic and treatment decision tree (in the case of Figure 3, CIN1). Patients receiving diagnostic biopsy will then receive a subsequent treatment or screening option (treatment biopsy, referred for rescreen at 6 months or returned to the NHS Cervical Screening Programme) based on their true underlying histology (as diagnostic biopsy and subsequent histology is assumed to be 100% sensitive and specific). Patients receiving treatment biopsy, at the initial colposcopy, or at a subsequent colposcopy as the result of a treatment decision based on a diagnostic biopsy, will be either ‘cured’ or ‘not cured’, i.e. the treatment biopsy has a failure rate that is described in more detail later in this chapter. Those patients who are cured will be referred for rescreen at 6 months or returned to the NHS Cervical Screening Programme and will enter the Markov model in the ‘clear’ health state. Those patients who are not cured will be referred for rescreen at 6 months or returned to the NHS Cervical Screening Programme and will enter the Markov model in their original health state, in the case of Figure 3, CIN1.

In our model we have split the types of biopsies into treatment and diagnostic biopsies. Different types of biopsies may be used for either reason, but the important distinction in the model is that a diagnostic biopsy is not curative and provides further information on the patient (in the model it is assumed to be perfect information). A treatment biopsy is undertaken with curative intent. Treatment biopsy may be a LLETZ, but in some cases less invasive treatment may be used.

Figure 4 shows the diagnostic decision tree for a patient with invasive cervical cancer. Similarly to Figure 3 for patients with CIN1, the patient receives an initial diagnosis by the colposcopist and can be incorrectly identified as clear, incorrectly identified as CIN1, incorrectly identified as CIN2/3 or correctly identified as having invasive cervical cancer. In contrast with patients with CIN1, patients with invasive cervical cancer who receive a treatment biopsy will also receive appropriate cancer treatment. As a result of the cancer therapy, they can be cured and returned to the NHS Cervical Screening Programme, or referred for rescreen at 6 months, entering the Markov model as ‘clear’ but receiving a QALY decrement and cost associated with survival of cervical cancer (this is discussed in detail in Model inputs) or they will die of cancer. Those patients who die of cancer do not enter the Markov model but instead receive an expected QALY ‘pay-off’ and costs associated with dying from cancer (this is discussed in detail in Model inputs).

Natural history and screening model

The natural history and screening model is based on the Sheffield model, a revised version of the model used by Hadwin et al. 51

The natural history model consists of nine states: clear, HPV, CIN1, CIN2/3, invasive cancer stages 1, 2, 3 and 4, and death (Figure 5). Patients enter the natural history model in the state based on their outcome from the diagnostic and treatment decision tree described above. Patients were allowed to progress and regress between these states every 6 months, based on age-related transition probabilities. 47 Possible transitions during any period in the model are represented by the arrows in the figure. Although the transition probabilities in the Hadwin et al. paper51 were largely based on an earlier study,54 the revised version of the model used the probabilities from Myers et al. 47 This model allowed for the regression of CIN2/3 to the less severe states of CIN1 and clear, transitions which were not allowed for in the earlier model.

FIGURE 5.

Natural history model. 47,51–52

At different time points during the model, the patients will also enter a screening pathway model (shown in Figure 6). For those returned to the NHS Cervical Screening Programme, screening will take place every 3 years between the ages of 25 and 49 years, and every 5 years between the ages of 50 and 64 years. For those referred for rescreen by cytology at 6 months, this will occur 6 months after the initial colposcopy. Following cytological screening, and HPV screening where required, a patient may be re-referred for colposcopy, based on the reasons for referral for colposcopy from the NHS Cervical Screening Programme. 5 When they are re-referred for colposcopy, they re-enter the diagnostic and treatment decision tree described previously.

FIGURE 6.

Screening pathway model. 52

It should be noted that not all patients with invasive cervical cancer will be identified as a result of cytological screening or colposcopy. These patients will be missed by screening but may subsequently be identified as having cancer, as a result of their cancer becoming symptomatic. These patients would then be treated appropriately, and some will be cured, and will transition to the ‘clear’ health state in the natural history model but receive an appropriate QALY decrement and cost associated with cancer treatment, whereas some will not be cured and will die of cancer, and will exit the model immediately but receive a QALY pay-off and cost associated with cancer treatment. As previously stated, these pay-offs and decrements associated with cancer are described in detail in Model inputs.

Chapter 3 Discussion

Chapter 4 Conclusions

DySIS, particularly when combined with colposcopy, has higher sensitivity than conventional colposcopy alone. There is no reliable evidence on the clinical effectiveness of the other adjunctive colposcopy technologies: the LuViva Advanced Cervical Scan and the Niris Imaging System.

From the economic analysis, the EAG concludes that the results of the analysis suggest that DySIS plus colposcopy is less costly and more effective than both DySIS alone and colposcopy alone, and that these results are robust to the numerous sensitivity analyses that were undertaken. The effectiveness evidence on LuViva and Niris is not considered sufficiently reliable to be included in the economic analysis.

Acknowledgements

We would like to thank the following for providing information and advice: Dr Maggie Cruickshank, Senior Lecturer in Gynaecology, University of Aberdeen; and Dr Hazel Squires and Dr Jim Chilcott, University of Sheffield.

We would also like to thank the following members of the Diagnostics Advisory Committee for additional information and advice: Dr Karin Denton, Consultant Cytopathologist, North Bristol NHS Trust; Mrs Phyllis Dunn, Clinical Lead Nurse, University Hospital of North Staffordshire; Mr Andrew Fish, Consultant Gynaecologist, Brighton and Sussex University Hospitals NHS Trust; Dr Sadaf Ghaem-Maghami, Senior Lecturer and Honorary Consultant in Gynaecological Oncology, Imperial College London; Mr Pierre Martin-Hirsch, Consultant Gynaecological Oncologist, Lancashire Teaching Hospitals NHS Foundation Trust; Mr Charles Redman, Consultant in Obstetrics and Gynaecology, University Hospital of North Staffordshire NHS Trust; and Dr Miren Turner, GP/Colposcopist.

References

1. Cancer Research UK . Cervical Cancer Statistics-UK 2012. http://info.cancerresearchuk.org/cancerstats/types/cervix/ (accessed 3 August 2011).
2. Colposcopy and programme management: guidelines for the NHS Cervical Screening Programme. NHSCSP; 2010.
3. Adjunctive colposcopy technologies for examination of the uterine cervix – DySIS, LuViva Advanced Cervical Scan, Niris Imaging System and Zilico APX-100: final scope. September 2011. London: NICE; 2011.
4. The NHS Information Centre – Public Health Indicators and Population Statistics team . Cervical Screening Programme England 2009–10 2010. www.ic.nhs.uk/webfiles/publications/008_Screening/cervscreen0910/2009_10_Cervical_Bulletin_Final_Report_AI_v1F.pdf (accessed 13 January 2012).
5. HPV triage and test of cure implementation guidance. NHSCSP good practice guide number 3. Sheffield: NHSCSP; 2011.
6. Louwers JA, Zaal A, Kocken M, ter Harmsel WA, Graziosi GCM, Spruijt JWM, et al. Dynamic spectral imaging colposcopy: higher sensitivity for detection of premalignant cervical lesions. BJOG 2011;118:309-18. http://dx.doi.org/10.1111/j.1471-0528.2010.02806.x.
7. LuViva™ Advanced Cervical Scan – adjunct to screening and colposcopy for cervical dyskaryosis. Birmingham: National Horizon Scanning Centre, University of Birmingham; 2010.
8. Etherington IJ, Luesley DM, Shafi MI, Dunn J, Hiller L, Jordan JA. Observer variability among colposcopists from the West Midlands region. Br J Obstet Gynaecol 1997;104:1380-4. http://dx.doi.org/10.1111/j.1471-0528.1997.tb11007.x.
9. Mitchell MF, Schottenfeld D, Tortolero-Luna G, Cantor SB, Richards-Kortum R. Colposcopy for the diagnosis of squamous intraepithelial lesions: a meta-analysis. Obstet Gynecol 1998;91:626-31. http://dx.doi.org/10.1016/S0029-7844(98)00006-4.
10. Jordan J, Arbyn M, Martin-Hirsch P, Schenck U, Baldauf JJ, Da Silva D, et al. European guidelines for quality assurance in cervical cancer screening: recommendations for clinical management of abnormal cervical cytology, part 1. Cytopathology 2008;19:342-54. http://dx.doi.org/10.1111/j.1365-2303.2008.00623.x.
11. Pretorius RG, Zhang W-H, Belinson JL, Huang M-N, Wu L-Y, Zhang X, et al. Colposcopically directed biopsy, random cervical biopsy, and endocervical curettage in the diagnosis of cervical intraepithelial neoplasia II or worse. Am J Obstet Gynecol 2004;191:430-4. http://dx.doi.org/10.1016/j.ajog.2004.02.065.
12. ASCUS-LSIL Triage Study (ALTS) Group . A randomized trial on the management of low-grade squamous intraepithelial lesion cytology interpretations. Am J Obstet Gynecol 2003;188:1393-400.
13. Systematic reviews: CRD&#039;s guidance for undertaking reviews in health care. York: University of York; 2009.
14. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JPA, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009;339. http://dx.doi.org/10.1136/bmj.b2700.
15. Centre for Evidence-Based Medicine, Toronto . Stats Calculator n.d. http://ktclearinghouse.ca/cebm/practise/ca/calculators/statscalc (accessed 18 January 2012).
16. Whiting PF, Rutjes AWS, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med 2011;155:529-36.
17. Soutter WP, Diakomanolis E, Lyons D, Ghaem-Maghami SA, Tosin A, Haidopoulos D, et al. Dynamic spectral imaging: improving colposcopy. Clin Cancer Res 2009;15:1814-20. http://dx.doi.org/10.1158/1078-0432.CCR-08-1636.
18. Soutter WP, Diakomanolis E, Lyons D, Haidopoulos D, Balas C. Comparison of the diagnostic performance characteristics of dynamic spectral imaging and colposcopy in low and high grade referral smear populations, separately [abstract]. J Low Genit Tract Dis 2010;14.
19. Flowers LC, Tadros TS. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results for adolescent patients. Norcross, GA: Guided Therapeutics; 2010.
20. Chakhtoura NA, Twiggs LB, Wilkinson EJ, Raab SS. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results of follow up data [abstract]. J Low Genit Tract Dis 2010;14.
21. Twiggs LB, Chakhtoura NA, Werner CL, Griffith WF, Flowers LC, Lashgari M, et al. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results of a 1,607 subject pivotal trial. J Low Genit Tract Dis 2010;14:248-9.
22. Flowers LC, Tadros TS. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results for adolescent subjects [abstract]. J Low Genit Tract Dis 2010;14.
23. Wilkinson EJ, Raab SS. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: histopathology review procedures and results. J Low Genit Tract Dis 2010;14:265-6.
24. Chakhtoura N, Twiggs L, Wilkinson E, Raab S. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results of follow up data. Norcross, GA: Guided Therapeutics; 2010.
25. Twiggs LB. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results of a 1,607 subject pivotal trial. Norcross, GA: Guided Therapeutics; 2010.
26. Chakhtoura N, Twiggs LB. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: results of a 1,607 subject pivotal trial. Norcross, GA: Guided Therapeutics; 2010.
27. Point-of-care, non-invasive cervical disease detection. Norcross, GA: Guided Therapeutics; 2010.
28. Spectroscopic evaluation of cervical neoplasia. Bethesda, MD: National Institutes of Health; n.d.
29. Distributor presentation United Kingdom and Ireland. Norcross, GA: Guided Therapeutics; 2010.
30. Lerner HP. PMA P100037 Clinical Module Amendment 2. Norcross, GA: Guided Therapeutics; 2011.
31. Escobar PF, Rojas-Espaillat L, Tisci S, Enerson C, Brainard J, Smith J, et al. Optical coherence tomography as a diagnostic aid to visual inspection and colposcopy for preinvasive and invasive cancer of the uterine cervix. Int J Gynecol Cancer 2006;16:1815-22. http://dx.doi.org/10.1111/j.1525-1438.2006.00665.x.
32. Liu Z, Belinson SE, Li J, Yang B, Wulan N, Tresser NJ, et al. Diagnostic efficacy of real-time optical coherence tomography in the management of preinvasive and invasive neoplasia of the uterine cervix. Int J Gynecol Cancer 2010;20:283-7. http://dx.doi.org/10.1111/IGC.0b013e3181cd1810.
33. Gallwas JKS, Turk L, Stepp H, Mueller S, Ochsenkuehn R, Friese K, et al. Optical coherence tomography for the diagnosis of cervical intraepithelial neoplasia. Lasers Surg Med 2011;43:206-12. http://dx.doi.org/10.1002/lsm.21030.
34. Gallwas J, Turk L, Stepp H, Muller S, Friese K, Dannecker C. Visualization of cervical intraepithelial neoplasia (CIN) by optical coherence tomography (OCT) [abstract]. Arch Gynecol Obstet 2010;282.
35. Wulan N, Belinson JL, Belinson SE, Rong X, Zhang W, Zhu YS, et al. Study of the diagnostic efficacy of ‘real time’ optical coherence tomography as an adjunct to colposcopy for cervical neoplasia. Cleveland, OH: Imalux Corporation; 2010.
36. Rasool N, Wulan N, Belinson SE, Belinson JL, Wang C, Rong X, et al. Study of the diagnostic efficacy of ‘real time&#039; optical coherence tomography as an adjunct to VIA in managing cervical neoplasia. Cleveland, OH: Imalux Corporation; 2008.
37. Apisdorf J, Sperling J, Rojas-Espaillat L, Escobar P, Feldchtein FI, Tresser N, et al. Niris optical coherence tomography as an adjunct to cervical cancer diagnosis and treatment: model for a single episode of care. Cleveland, OH: Imalux Corporation; 2007.
38. Rojas-Espaillat L, Escobar PF, Belinson JL, Tisc S, Brainard J, De la Rosa V, et al. Optical coherence tomography as a diagnostic adjunct to visual inspection and colposcopy for pre-invasive and invasive cancer of the uterine cervix. Cleveland, OH: Imalux Corporation; n.d.
39. Kuznetsova IA, Gladkova ND, Gelikonov VM, Belinson JL, Shakhova NM, Feldchtein FI, et al. Optical coherence tomography: technology and applications. Berlin: Springer; 2008.
40. Rutjes AWS, Reitsma JB, Di Nisio M, Smidt N, van Rijn JC, Bossuyt PMM. Evidence of bias and variation in diagnostic accuracy studies. CMAJ 2006;174:469-76. http://dx.doi.org/10.1503/cmaj.050090.
41. Bossuyt PM, Reitsma JB. The STARD initiative. Lancet 2003;361. http://dx.doi.org/10.1016/S0140-6736(03)12122-8.
42. Bossuyt PMM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, et al. The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Clin Chem 2003;49:7-8. http://dx.doi.org/10.1373/49.1.7.
43. Bossuyt PMM. STARD statement: still room for improvement in the reporting of diagnostic accuracy studies. Radiology 2008;248:713-14. http://dx.doi.org/10.1148/radiol.2483080868.
44. Wilczynski NL. Quality of reporting of diagnostic accuracy studies: no change since STARD statement publication: before-and-after study. Radiology 2008;248:817-23. http://dx.doi.org/10.1148/radiol.2483072067.
45. Chuck A. Cost-effectiveness of 21 alternative cervical cancer screening strategies. Value Health 2010;13:169-79. http://dx.doi.org/10.1111/j.1524-4733.2009.00611.x.
46. Legood R, Gray A, Wolstenholme J, Moss S. Lifetime effects, costs, and cost effectiveness of testing for human papillomavirus to manage low grade cytological abnormalities: results of the NHS pilot studies. BMJ 2006;332:79-85. http://dx.doi.org/10.1136/bmj.38698.458866.7C.
47. Myers ER, McCrory DC, Nanda K, Bastian L, Matchar DB. Mathematical model for the natural history of human papillomavirus infection and cervical carcinogenesis. Am J Epidemiol 2001;151:1158-71. http://dx.doi.org/10.1093/oxfordjournals.aje.a010166.
48. Bergeron C, Largeron N, McAllister R, Mathevet P, Remy V. Cost-effectiveness analysis of the introduction of a quadrivalent human papillomavirus vaccine in France. Int J Technol Assess Health Care [serial Online] 2008:10-9.
49. Kulasingam SL, Benard S, Barnabas RV, Largeron N, Myers ER. Adding a quadrivalent human papillomavirus vaccine to the UK cervical cancer screening programme: a cost-effectiveness analysis. Cost Effectiveness and Resource Allocation 2008;6. http://dx.doi.org/10.1186/1478-7547-6-4.
50. Mandelblatt JS, Lawrence WF, Womack SM, Jacobsen D, Yo B, Hwang Y, et al. Benefits and costs of using HPV testing to screen for cervical cancer. JAMA 2002;287:2372-81. http://dx.doi.org/10.1001/jama.287.18.2372/NHSEED-22002008122/frame.html.
51. Hadwin R, Eggington S, Brennan A, Walker P, Patnick J, Pilgrim H. Modelling the cost-effectiveness and capacity impact of changes to colposcopy referral guidelines for women with mild dyskaryosis in the UK Cervical Screening Programme. BJOG 2008;115:749-57. http://dx.doi.org/10.1111/j.1471-0528.2008.01683.x.
52. Kim E-J. Modelling the Cost-Effectiveness of Human Papillomavirus (HPV) Testing for Triage of Women With Low-Grade Abnormal Cervical Smears: A Study Within the TOMBOLA Trial 2010.
53. Guide to the Methods of Technology Appraisal. London: NICE; 2008.
54. Sherlaw-Johnson C, Gallivan S, Jenkins D, Jones MH. Cytological screening and management of abnormalities in prevention of cervical cancer: an overview with stochastic modelling. J Clin Pathol 1994;47:430-5. http://dx.doi.org/10.1136/jcp.47.5.430.
55. Ghaem-Maghami S, De-Silva D, Tipples M, Lam S, Perryman K, Soutter W. Determinants of success in treating cervical intraepithelial neoplasia. BJOG 2011;118:679-84. http://dx.doi.org/10.1111/j.1471-0528.2010.02770.x.
56. Ghaem-Maghami S, Sagi S, Majeed G, Soutter WP. Incomplete excision of cervical intraepithelial neoplasia and risk of treatment failure: a meta-analysis. Lancet Oncol 2007;8:985-93.
57. Health and life events guidance and metadata. ONS; n.d.
58. Subtopic: population estimates by age and sex. ONS; n.d.
59. Cervical cancer statistics and outlook. UK: Cancer Research; 2010.
60. Chuck A. Cost-effectiveness of 21 alternative cervical cancer screening strategies. Value Health 2010:169-79. http://dx.doi.org/10.1111/j.1524-4733.2009.00611.x.
61. Karnon J, Peters J, Platt J, Chilcott J, McGoogan E, Brewer N. Liquid-based cytology in cervical screening: an updated rapid and systematic review and economic analysis. Health Technol Assess 2004;8.
62. Insinga RP, Glass AG, Myers ER, Rush BB. Abnormal outcomes following cervical cancer screening: event duration and health utility loss. Med Decis Making 2007;27:414-22. http://dx.doi.org/10.1177/0272989X07302128.
63. Birch S, Melnikow J, Kuppermann M. Conservative versus aggressive follow up of mildly abnormal Pap smears: testing for process utility. Health Econ 2003;12:879-84.
64. Group T, Sharp L, Cotton S, Cochran C, Gray N, Little J, et al. After-effects reported by women following colposcopy, cervical biopsies and LLETZ: results from the TOMBOLA trial. BJOG 2009;116:1506-14. http://dx.doi.org/10.1111/j.1471-0528.2009.02263.x.
65. Arbyn M, Kyrgiou M, Simoens C, Raifu AO, Koliopoulos G, Martin-Hirsch P, et al. Perinatal mortality and other severe adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: meta-analysis. BMJ 2008;337. http://dx.doi.org/10.1136/bmj.a1284.
66. Group T. Options for managing low grade cervical abnormalities detected at screening: cost effectiveness study. BMJ 2009;339. http://dx.doi.org/10.1136/bmj.b2549.
67. Wolstenholme JL, Whynes DK. Stage-specific treatment costs for cervical cancer in the United Kingdom. Eur J Cancer 1998;34:1889-93. http://dx.doi.org/10.1016/S0959-8049(98)00232-9.
68. Kelly RS, Patnick J, Kitchener HC, Moss SM. HPV testing as a triage for borderline or mild dyskaryosis on cervical cytology: results from the Sentinel Sites study. Br J Canc 2011;105:983-8. http://dx.doi.org/10.1038/bjc.2011.326.
69. Eggington S, Hadwin R, Brennan A, Walker P. Modelling the impact of referral guideline changes for mild dyskaryosis on colposcopy services in England. Sheffield: NHS Cancer Screening Programmes; 2006.
70. Ferris DG, Macfee MS, Miller JA, Litaker MS, Crawley D, Watson D. The efficacy of telecolposcopy compared with traditional colposcopy. Obstet Gynecol 2002;99:248-54. http://dx.doi.org/10.1016/S0029-7844(01)01671-4.
71. 2000.
72. 2002.
73. SpectRx cervical cancer test moves toward approval. Photomed Laser Surg 2004;22:259-60. http://dx.doi.org/10.1089/1549541041438650.
74. American Society for Colposcopy and Cervical Pathology (ASCCP), Hagerstown, MD. Society news. J Low Genit Tract Dis 2006;10:69-71.
75. New imaging system helps detect cervical precancer. AWHONN Lifelines 2006;10.
76. American Society for Colposcopy and Cervical Pathology (ASCCP) . Biennial Meeting, Las Vegas, NV 24–28 March 2010, abstracts. J Low Genit Tract Dis 2010;14:241-74.
77. Clinical positioning of APX 100. Zilico; 2011.
78. Abdul S, Brown BH, Milnes P, Tidy JA. A clinical study of the use of impedance spectroscopy in the detection of cervical intraepithelial neoplasia (CIN). Gynecol Oncol 2005;99:S64-6. http://dx.doi.org/10.1016/j.ygyno.2005.07.046.
79. Abdul S, Brown BH, Milnes P, Tidy JA. The use of electrical impedance spectroscopy in the detection of cervical intraepithelial neoplasia. Int J Gynecol Cancer 2006;16:1823-32. http://dx.doi.org/10.1111/j.1525-1438.2006.00651.x.
80. Acosta-Mesa H-G, Cruz-Ramirez N, Hernandez-Jimenez R. Aceto-white temporal pattern classification using k-NN to identify precancerous cervical lesion in colposcopic images. Comput Biol Med 2009;39:778-84. http://dx.doi.org/10.1016/j.compbiomed.2009.06.006.
81. Agrawal A, Harrell T, Bambot S, Faupel M, Ferris D. Multimodal multispectral imaging of the cervix in vivo for the detection of neoplasia. Proc SPIE 2001;4259:68-74. http://dx.doi.org/10.1117/12.432482.
82. Alush A, Greenspan H, Goldberger J. Automated and interactive lesion detection and segmentation in uterine cervix images. IEEE Trans Med Imaging 2010;29:488-501. http://dx.doi.org/10.1109/TMI.2009.2037201.
83. Alvarez RD, Wright TC. Optical Detection Group . Effective cervical neoplasia detection with a novel optical detection system: a randomized trial. Gynecol Oncol 2007;104:281-9. http://dx.doi.org/10.1016/j.ygyno.2006.08.056.
84. Anastasiadou M, De Martino A, Clement D, Liege F, Laude-Boulesteix B, Quang N, et al. Polarimetric imaging for the diagnosis of cervical cancer. Physica Status Solidi C 2008;5:1423-6. http://dx.doi.org/10.1002/pssc.200777805.
85. Atkinson EN. Age and FSH effects in fluorescence spectra from the cervix: an exploratory analysis. Gynecol Oncol 2005;99:S95-7. http://dx.doi.org/10.1016/j.ygyno.2005.07.051.
86. Azar EG, Intes X. Bellingham, WA: SPIE; 2009.
87. Badizadegan K, Backman V, Boone CW, Crum CP, Dasari RR, Georgakoudi I, et al. Spectroscopic diagnosis and imaging of invisible pre-cancer. Faraday Discuss 2004;126:265-79. http://dx.doi.org/10.1039/b305410a.
88. Balas C. A novel optical imaging method for the early detection, quantitative grading, and mapping of cancerous and precancerous lesions of cervix. IEEE Trans Biomed Eng 2001;48:96-104. http://dx.doi.org/10.1109/10.900259.
89. Balas C. The potential of spectral imaging in the non-invasive diagnostics of tissue cancers and pre-cancers. Proc SPIE 2002;5123:287-92.
90. Balas C, Simos C, Sciadas Y. Dynamic spectral imaging for in vivo diagnostics, screening and guided therapeutics of cervical neoplasia. Indian J Med Res 2005;121.
91. Balas CJ, Papoutsoglou G, Potirakis A. In vivo molecular imaging of cervical neoplasia using acetic acid as biomarker. IEEE J Sel Top Quantum Electron 2008;14:29-42. http://dx.doi.org/10.1109/JSTQE.2007.913396.
92. Balas C, Papoutsoglou G, Diakomanolis E, Haidopoulos D, Soutter WP. Quantitative assessment of the AW phenomenon and determination of its biophysical origins as a means for in vivo assessing CIN-related micro-structural and functional changes. J Low Genit Tract Dis 2010;14:268-9.
93. Balasubramani L, Brown BH, Healey J, Tidy JA. The detection of cervical intraepithelial neoplasia by electrical impedance spectroscopy: the effects of acetic acid and tissue homogeneity. Gynecol Oncol 2009;115:267-71. http://dx.doi.org/10.1016/j.ygyno.2009.08.010.
94. Bazant-Hegemark F, Stone N, Read MD, McCarthy K, Wang RK. Optical coherence tomography (OCT) imaging and computer aided diagnosis of human cervical tissue specimens. Proc SPIE 2007;6627.
95. Belinson J, Qiao YL, Pretorius R, Zhang WH, Elson P, Li L, et al. Shanxi Province Cervical Cancer Screening Study: a cross-sectional comparative trial of multiple techniques to detect cervical neoplasia. Gynecol Oncol 2001;83:439-44. http://dx.doi.org/10.1006/gyno.2001.6370.
96. Belinson JL. n.d.
97. Belinson S, Ledford K, Wulan N, Rasool N, Wang C, Rollins A, et al. Epithelial brightness by optical coherence tomography distinguishes grades of dysplasia. Cleveland, OH: Imalux Corporation; 2009.
98. Benavides JM, Chang SC, Park SY, Richards-Kortum R, Mackinnon N, MacAulay C, et al. Multispectral digital colposcopy for in vivo detection of cervical cancer. Opt Express 2003;11:1223-36. http://dx.doi.org/10.1364/OE.11.001223.
99. Bogaards A, Aalders MC, Zeyl CC, de Blok S, Dannecker C, Hillemanns P, et al. Localization and staging of cervical intraepithelial neoplasia using double ratio fluorescence imaging. J Biomed Opt 2002;7:215-20. http://dx.doi.org/10.1117/1.1463045.
100. Brookner C, Utzinger U, Follen M, Richards-Kortum R, Cox D, Atkinson EN. Effects of biographical variables on cervical fluorescence emission spectra. J Biomed Opt 2003;8:479-83. http://dx.doi.org/10.1117/1.1578642.
101. Brown BH, Milnes P, Abdul S, Tidy JA. n.d.
102. Brown BH, Tidy JA, Boston K, Blackett AD, Smallwood RH, Sharp F. Relation between tissue structure and imposed electrical current flow in cervical neoplasia. Lancet 2000;355:892-5. http://dx.doi.org/10.1016/S0140-6736(99)09095-9.
103. Brown BH, Milnes P, Abdul S, Tidy JA. Detection of cervical intraepithelial neoplasia using impedance spectroscopy: a prospective study. BJOG 2005;112:802-6. http://dx.doi.org/10.1111/j.1471-0528.2004.00530.x.
104. Bush J, Davis PG, Marcus MA. All-fiber optic coherence domain interferometric techniques. Boston, MA: Optiphase, Inc.; 2001.
105. Cantor SB, Levy LB, Cardenas-Turanzas M, Basen-Engquist K, Le T, Beck JR, et al. Collecting direct non-health care and time cost data: application to screening and diagnosis of cervical cancer. Med Decis Making 2006;26:265-72. http://dx.doi.org/10.1177/027298906288679.
106. Cantor SB, Yamal J-M, Guillaud M, Cox DD, Atkinson EN, Benedet JL, et al. Accuracy of optical spectroscopy for the detection of cervical intraepithelial neoplasia: testing a device as an adjunct to colposcopy. Int J Cancer 2011;128:1151-68. http://dx.doi.org/10.1002/ijc.25667.
107. Chance B, Alfano RR, Tromberg BJ, Tamura M, Sevick-Muraca EM. Bellingham, WA: SPIE; 2005.
108. Chang SK, Follen M, Malpica A, Utzinger U, Staerkel G, Cox D, et al. Optimal excitation wavelengths for discrimination of cervical neoplasia. IEEE Trans Biomed Eng 2002;49:1102-10. http://dx.doi.org/10.1109/TBME.2002.803597.
109. Chang SK, Mirabal YN, Atkinson EN, Malpica A, Follen M, Richards-Kortum R. n.d.
110. Chang SK, Pavlova I, Marin NM, Follen M, Richards-Kortum R. Fluorescence spectroscopy as a diagnostic tool for detecting cervical pre-cancer. Gynecol Oncol 2005;99:S61-3. http://dx.doi.org/10.1016/j.ygyno.2005.07.045.
111. Chang SK, Mirabal YN, Atkinson EN, Cox D, Malpica A, Follen M, et al. Combined reflectance and fluorescence spectroscopy for in vivo detection of cervical pre-cancer. J Biomed Opt 2005;10. http://dx.doi.org/10.1117/1.1899686.
112. Chang VT-C, Cartwright PS, Bean SM, Palmer GM, Bentley RC, Ramanujam N. Quantitative physiology of the precancerous cervix in vivo through optical spectroscopy. Neoplasia 2009;11:325-32.
113. Chang VT-C, Bean SM, Cartwright PS, Ramanujam N. Visible light optical spectroscopy is sensitive to neovascularization in the dysplastic cervix. J Biomed Opt 2010;15. http://dx.doi.org/10.1117/1.3495730.
114. Chang VT-C, Merisier D, Yu B, Walmer DK, Ramanujam N. Calibration schemes of a field-compatible optical spectroscopic system to quantify neovascular changes in the dysplastic cervix. Proc SPIE 2011;7891.
115. Cheung TH, Wu T, Lo WK, Yu MY, Qu J, Monsonego J. 5th International Multidisciplinary Congress (EUROGIN 2003), Paris, France, 13–16 April, 2003. Bologna: Medimond Inc.; n.d.
116. Claude I, Pouletaut P, Huault S, Boulanger JC. n.d.
117. Measurement of digital colposcopy for fluorescence spectroscopy of cervical intraepithelial neoplasia. Bethesda, MD: National Institutes of Health; n.d.
118. Spectroscopy in diagnosing dysplasia and neoplasia in the cervix in women undergoing colposcopy. Bethesda, MD: National Institutes of Health; n.d.
119. Measurement of digital colposcopy for fluorescence spectroscopy of TNC using a second device. Bethesda, MD: National Institutes of Health; n.d.
120. Digital colposcopy in finding cervical intraepithelial neoplasia in patients with an abnormal pap smear. Bethesda, MD: National Institutes of Health; n.d.
121. Digital colposcopy in finding cervical intraepithelial neoplasia. Bethesda, MD: National Institutes of Health; n.d.
122. Spectroscopy versus standard care in cervical cancer patients. Bethesda, MD: National Institutes of Health; n.d.
123. Combined digital colposcopy analysis to improve cervical precancer and cancer detection. Bethesda, MD: National Institutes of Health; n.d.
124. Cervical cancer detection using optical spectroscopy. Bethesda, MD: National Institutes of Health; n.d.
125. Non-invasive, real-time technology for diagnosis of cervical tissue. Bethesda, MD: National Institutes of Health; n.d.
126. Patient satisfaction &amp; quality of life in patients with cervical cancer. Bethesda, MD: National Institutes of Health; n.d.
127. Digital imaging aid for assessment of cervical dysplasia. Bethesda, MD: National Institutes of Health; n.d.
128. Investigation of the uterine cervix using CervicalMD. Bethesda, MD: National Institutes of Health; n.d.
129. High resolution digital imaging of the uterine cervix. Bethesda, MD: National Institutes of Health; n.d.
130. Reflectance confocal imaging in cervical cancer patients. Bethesda, MD: National Institutes of Health; n.d.
131. Collier T, Guillaud M, Follen M, Malpica A, Richards-Kortum R. Real-time reflectance confocal microscopy: comparison of two-dimensional images and three-dimensional image stacks for detection of cervical precancer. J Biomed Opt 2007;12. http://dx.doi.org/10.1117/1.2717899.
132. Coppolillo EF, Eliseth MC, Malamud de Ruda Vega H, Perazzi BE. Three-dimensional digital colposcopy. Acta Obstet Gynecol Scand 2009;88:120-1. http://dx.doi.org/10.1080/00016340802534802.
133. Dattamajumdar AK, Wells D, Parnell J, Lewis JT, Ganguly D, Wright TC. n.d.
134. Dattamajumdar AK, Wells D, Parriell J, Ganguly D, Wright TC. Preliminary results from multi-center clinical trials for detection of cervical squamous intraepithelial lesions using a novel full-field evoked tissue fluorescence based imaging instrument. Proc SPIE 2001;4432:312-13. http://dx.doi.org/10.1117/12.447157.
135. Dattamajumdar AK, Ferenczy A, Tuggy M, Rosing M, Waldman J, Hunter V, et al. 5th International Multidisciplinary Congress (EUROGIN 2003), Paris, France, 13–16 April 2003. Bologna: Medimond Inc.; n.d.
136. DeSantis T, Chakhtoura N, Twiggs L, Ferris DG, Lashgari M, Flowers L, et al. Spectroscopic imaging as a triage test for cervical disease: a prospective multicenter clinical trial. J Low Genit Tract Dis 2007;11:18-24. http://dx.doi.org/10.1097/01.lgt.0000230207.50495.05.
137. DeWeert MJ, Oyama J, McLaughlin E, Jacobson E, Hakansson J, Bignami GS, et al. Analysis of spatial variability in hyperspectral imagery of the uterine cervix in vivo. Proc SPIE 2003;4959:67-76. http://dx.doi.org/10.1117/12.479509.
138. Dominik P, Witold K, Marcin P, Agata J, Maciej P, Marek S. Assessment of optoelectronic method and molecular test usefulness for cervical intraepithelial neoplasia and cervical cancer detection. Ginekol Pol 2010;81:426-30.
139. Drezek R, Follen M, Richards-Kortum R. 2002.
140. Escobar PF, Belinson JL, White A, Shakhova NM, Feldchtein FI, Kareta MV, et al. Diagnostic efficacy of optical coherence tomography in the management of preinvasive and invasive cancer of uterine cervix and vulva. Int J Gynecol Cancer 2004;14:470-4. http://dx.doi.org/10.1111/j.1048-891x.2004.14307.x.
141. Escobar PF, Rojas-Espaillat L, Belinson JL. Optical diagnosis of cervical dysplasia. Int J Gynaecol Obstet 2005;89:63-4. http://dx.doi.org/10.1016/j.ijgo.2004.12.029.
142. Feldchtein FI, Gladkova ND, Snopova L, Zagaynova E, Streltzova O, Shakhov A, et al. Blinded recognition of optical coherence tomography images of human mucosa precancer. Proc SPIE 2003;4956:89-94. http://dx.doi.org/10.1117/12.478926.
143. Feldchtein F, Bush J, Gelikonov G, Gelikonov V, Piyevsky S. Cost effective, all-fiber autocorrelator based 1300 nm OCT system. Proc SPIE 2005;5690:349-55. http://dx.doi.org/10.1117/12.589502.
144. Ferris DG, Lawhead RA, Dickman ED, Holtzapple N, Miller JA, Grogan S, et al. Multimodal hyperspectral imaging for the noninvasive diagnosis of cervical neoplasia. J Low Genit Tract Dis 2001;5:65-72.
145. Ferris DG, Litaker MS, Dickman ED, Allmond LM, Smith KM, Arrington TL. Women&#039;s responses to cervical interrogation by fluorescent and reflective spectroscopy. J Low Genit Tract Dis 2003;7:299-303. http://dx.doi.org/10.1097/00128360-200310000-00013.
146. Ferris DG, Li W, Gustafsson U, Lieberman RW, Galdos O, Santos C. Enhancing colposcopy with polarized light. J Low Genit Tract Dis 2010;14:149-54. http://dx.doi.org/10.1097/LGT.0b013e3181d52f59.
147. Freeberg JA, Serachitopol DM, McKinnon N, Price R, Atkinson EN, Cox DD, et al. Fluorescence and reflectance device variability throughout the progression of a phase II clinical trial to detect and screen for cervical neoplasia using a fiber optic probe. J Biomed Opt 2007;12. http://dx.doi.org/10.1117/1.2750332.
148. Fujii T, Nakamura M, Kameyama K, Saito M, Nishio H, Ohno A, et al. Digital colposcopy for the diagnosis of cervical adenocarcinoma using a narrow band imaging system. Int J Gynecol Cancer 2010;20:605-10. http://dx.doi.org/10.1111/IGC.0b013e3181d98da9.
149. Gage JC, Safaeian M, Jeronimo J, Schiffman M. Spectroscopic imaging as triage test for cervical disease. J Low Genit Tract Dis 2008;12:52-3. http://dx.doi.org/10.1097/lgt.0b013e318149e232.
150. Gallwas J, Turk L, Stepp H, Friese K, Dannecker C. Visualization of cervical intraepithelial neoplasias by optical coherence tomography: early results. Geburtshilfe Frauenheilkd 2009;69:931-4. http://dx.doi.org/10.1055/s-0029-1186174.
151. Gallwas J, Chiapponi C, Turk L, Ochsenkuehn R, Friese K, Dannecker C. Optical coherence tomography: preliminary results with a new noninvasive technique for evaluating uterine cervical tissue and vulvar epithelium. Minerva Ginecol 2010;62:395-401. http://dx.doi.org/10.1002/uog.7656.
152. Gallwas J, Turk L, Friese K, Dannecker C. Optical coherence tomography as a non-invasive imaging technique for preinvasive and invasive neoplasia of the uterine cervix. Ultrasound Obstet Gynecol 2010;36:624-9.
153. Gandhi SV, Walker D, Milnes P, Mukherjee S, Brown BH, Anumba DOC. Electrical impedance spectroscopy of the cervix in non-pregnant and pregnant women. Eur J Obstet Gynecol Reprod Biol 2006;129:145-9. http://dx.doi.org/10.1016/j.ejogrb.2005.12.029.
154. Georgakoudi I, Sheets EE, Crum CP, Mueller MG, Backman V, Feld MS. Tri-modal spectroscopy as a tool for detecting cervical squamous intraeptihelial lesions in vivo. Proc SPIE 2001;2:1-9. http://dx.doi.org/10.1117/12.447118.
155. Georgakoudi I, Sheets EE, Muller MG, Backman V, Crum CP, Badizadegan K, et al. Trimodal spectroscopy for the detection and characterization of cervical precancers in vivo. Am J Obstet Gynecol 2002;186:374-82. http://dx.doi.org/10.1067/mob.2002.121075.
156. Ghanate AD, Kothiwale S, Singh SP, Bertrand D, Murali Krishna C. Comparative evaluation of spectroscopic models using different multivariate statistical tools in a multicancer scenario. J Biomed Opt 2011;16. http://dx.doi.org/10.1117/1.3548303.
157. Gladkova ND, Shakhova NM, Shakhov BY, Snopova LB. Optic coherent tomography: a new high-resolution technology of visualization of tissue structures. Communication II. Optical images of benign and malignant entities. Vestn Roentgenol Radiol 2004;2:44-5.
158. Gudibande R, Mozumder M, Singh R, Panigrahi PK, Gupta S, Pradhan A. Differentiating human cervical dysplastic and normal tissue through wavelet domain characterization of intrinsic fluorescence. Proc SPIE 2011;7902.
159. What is LuViva?. Norcross: GA: Guided Therapeutics; n.d.
160. Gustafsson U, McLaughlin E, Jacobson E, Hakansson J, Troy P, DeWeert MJ, et al. Fluorescence and reflectance monitoring of human cervical tissue in vivo: a case study. Proc SPIE 2003;4959:100-10. http://dx.doi.org/10.1117/12.479495.
161. Gustafsson U, McLaughlin E, Jacobson E, Hakansson J, Troy P, DeWeert MJ, et al. In vivo fluorescence and reflectance imaging of human cervical tissue. Proc SPIE 2003;5031:521-30. http://dx.doi.org/10.1117/12.480413.
162. Harper DM, Schmid-Saugeon P, Zelenchuk A, Pitts JD, Kaufman HB. Quantitative measurements of acetowhitening may provide an adjunct to colposcopy to identify cervical intraepithelial neoplasia 2 and 3. Obstet Gynecol 2004;103:S18-19.
163. Hsiung P, Chudoba C, Pitris C, Li XD, Ko T, Fujimoto JG, et al. In vivo colposcopic imaging of neoplastic tissues using Optical Coherence Tomography. Proc SPIE 2001;4254:58-60. http://dx.doi.org/10.1117/12.427947.
164. Huang D, Swanson EA, Lin CP, Schuman JS, Stinson WG, Chang W, et al. Optical coherence tomography. Science 1991;254:1178-81. http://dx.doi.org/10.1126/science.1957169.
165. Huang Z, Mo J, Zheng W, Low J, Ng J, Ilancheran A. Combining near-infrared autofluorescence and Raman spectroscopy improves the in vivo detection of cervical precancer. 2008 Quantum Electronics and Laser Science Conference (QELS). Piscataway, NJ: IEEE; 2008.
166. Huh WK, Cestero RM, Garcia FA, Gold MA, Guido RS, McIntyre-Seltman K, et al. Optical detection of high-grade cervical intraepithelial neoplasia in vivo: results of a 604-patient study. Am J Obstet Gynecol 2004;190:1249-57. http://dx.doi.org/10.1016/j.ajog.2003.12.006.
167. Huh WK. Optical detection of cervical neoplasia: results from an internally-controlled multicenter study. Gynecol Oncol 2005;99. http://dx.doi.org/10.1016/j.ygyno.2005.07.041.
168. Report no. 0400–000018 Rev A. Cleveland, OH: Imalux Corporation; 2010.
169. Report no. 0400–000019 Rev A. Cleveland, OH: Imalux Corporation; 2010.
170. Cervical cancer [economic effects]. Cleveland, OH: Imalux Corporation; n.d.
171. OCT technology: Imalux. Cleveland, OH: Imalux Corporation; n.d.
172. Report No.: 0400–000025–0. Cleveland, OH: Imalux Corporation; n.d.
173. Report No.: 0400–000024–0. Cleveland, OH: Imalux Corporation; n.d.
174. Imalux . Niris Imaging System [Powerpoint Presentation Prepared for the National Institute for Health and Clinical Excellence] 2011.
175. Jeronimo J, Schiffman M. Colposcopy at a crossroads. Am J Obstet Gynecol 2006;195:349-53. http://dx.doi.org/10.1016/j.ajog.2006.01.091.
176. Jeronimo J, Massad LS, Schiffman M. Visual appearance of the uterine cervix: correlation with human papillomavirus detection and type. Am J Obstet Gynecol 2007;197.
177. Johansson A, Kromer K, Sroka R, Stepp H. Clinical optical diagnostics – status and perspectives. Medical Laser Application 2008;23:155-74. http://dx.doi.org/10.1016/j.mla.2008.08.002.
178. Kang W. Diagnostic efficacy of computer-extracted image features using optical coherence tomography in pre- cancerous cervix. Cleveland, OH: Imalux Corporation; 2008.
179. Kang W, Qi X, Kareta M, Belinson JL, Rollins AM. Automated processing and classification of cervical images using optical coherence tomography. Washington, DC: Optical Society of America; 2008.
180. Kang W, Qi X, Tresser NJ, Kareta M, Belinson JL, Rollins AM. Computer-aided diagnosis in OCT of the precancerous cervix. Cleveland, OH: Imalux; 2010.
181. Kang W, Qi X, Tresser NJ, Kareta M, Belinson JL, Rollins AM. Diagnostic efficacy of computer extracted image features in optical coherence tomography of the precancerous cervix. Med Phys 2011;38:107-13. http://dx.doi.org/10.1118/1.3523098.
182. Kanter EM, Majumder S, Kanter GJ, Woeste EM, Mahadevan-Jansen A. Effect of hormonal variation on Raman spectra for cervical disease detection. Am J Obstet Gynecol 2009;200:512-3. http://dx.doi.org/10.1016/j.ajog.2008.11.024.
183. Knapp P, Zbroch T, Kobylec M, Knapp P. Computerized fluorocolposcopy: the significance in evaluation and therapy of early cervical lesions. Przegl Lek 2004;61:1393-9.
184. Kuznetzova IA, Shakhova NM, Kachalina TS, Gladkova ND, Myakov AV, Iksanov RR, et al. Optical coherence tomography in diagnosing of cervical cancer. Proc SPIE 2000;3097:440-7. http://dx.doi.org/10.1117/12.386285.
185. Lange H, Baker R, Hakansson J, Gustafsson UP. Reflectance and fluorescence hyperspectral elastic image registration. Proc SPIE 2004;5370:335-45. http://dx.doi.org/10.1117/12.535720.
186. Lange H. Automatic detection of multi-level acetowhite regions in RGB color images of the uterine cervix. Proc SPIE 2005;5747:1004-17. http://dx.doi.org/10.1117/12.596064.
187. Lange H. Automatic glare removal in reflectance imagery of the uterine cervix. Proc SPIE 2005;5747:2183-92. http://dx.doi.org/10.1117/12.596012.
188. Lange H, Ferris DG. Computer-aided-diagnosis (CAD) for colposcopy. Proc SPIE 2005;5747:71-84. http://dx.doi.org/10.1117/12.594658.
189. Ledford K, Tresser N. Epithelium brightness by optical coherence tomography (OCT) may distinguish low-grade from high-grade cervical dysplasia in real-time n.d.
190. Lee JS, Follen M, MacAulay C, Pikkula B, Serachitopol D, Price R, et al. Sources of variability in fluorescence spectroscopic measurements in a phase II clinical trial of 850 patients. Gynecol Oncol 2007;107:S260-9. http://dx.doi.org/10.1016/j.ygyno.2007.07.012.
191. Lee S-W, Kang J-H, Yoo J-Y, Kim B-M. n.d.
192. Lee S-W, Yoo J-Y, Kang J-H, Kang M-S, Jung S-H, Chong Y, et al. n.d.
193. Lee S-W, Yoo J-Y, Kang J-H, Kang M-S, Jung S-H, Chong Y, et al. Optical diagnosis of cervical intraepithelial neoplasm (CIN) using polarization-sensitive optical coherence tomography. Opt Express 2008;16:2709-19. http://dx.doi.org/10.1364/OE.16.002709.
194. Li W, Van Raad V, Gu J, Hansson U, Hakansson J, Lange H, et al. Computer Vision for Biomedical Image Applications. First International Workshop, CVBIA 2005. Proceedings. Berlin: Springer-Verlag Berlin; 2005.
195. Li W, Poirson A. Advances in visual computing. Second International Symposium, ISVC 2006, Lake Tahoe, NV, USA, November 6–8, 2006, Proceedings, Part II. Berlin: Springer-Verlag; 2006.
196. Li W, Soto-Thompson M, Gustafsson U. A new image calibration system in digital colposcopy. Opt Express 2006;14:12887-901. http://dx.doi.org/10.1364/OE.14.012887.
197. Li W, Soto-Thompson M, Yizhi X, Lange H. A new image calibration technique for colposcopic images. Proc SPIE 2006;6144.
198. Li W, Gu J, Ferris D, Poirson A. Automated image analysis of uterine cervical images. Proc SPIE 2007;6514.
199. Li W, Ferris DG, Lieberman RW. Computerized image analysis for acetic acid induced intraepithelial lesions. Proc SPIE 2008;6914.
200. Li W, Lieberman RW, Nie S, Xie Y, Eldred M, Oyama J. Histopathology reconstruction on digital imagery. Proc SPIE 2009;7263.
201. Li W, Venkataraman S, Gustafsson U, Oyama JC, Ferris DG, Lieberman RW. Using acetowhite opacity index for detecting cervical intraepithelial neoplasia. J Biomed Opt 2009;14. http://dx.doi.org/10.1117/1.3079810.
202. Loning M, Lankenau E, Strunck C, Krokowski M, Hillbricht S, Diedrich K, et al. Optical coherence tomography: a noninvasive high resolution cross-sectional tissue imaging in addition to colposcopy. Geburtshilfe Frauenheilkd 2003;63:1158-61. http://dx.doi.org/10.1055/s-2003-815196.
203. Luck B, Carlson K, Collier T, Sung K-B. Confocal microscopy. IEEE Potentials 2004;23:14-7. http://dx.doi.org/10.1109/MP.2004.1266933.
204. Lukic A, Iannaccio S, Di Properzio M, Carico E, Camboni A, Vecchione A, et al. Microcolposcopy in the diagnostic evaluation of abnormal cervical cytology: when and why to do it. Clin Exp Obstet Gynecol 2009;36:26-30.
205. MacKinnon NB, Cardeno M, Au S, MacAulay CE, Pikkula BM, Serachitopol D, et al. Design of a multispectral digital colposcope. Proc SPIE 2007;6430.
206. Malpica A, Zuluaga A, Boiko I, Richards-Kortum R, Follen M. Assessment of cervical squamous intraepithelial lesions by optical coherence tomography and colposcopy [abstract]. Mod Pathol 2001;14.
207. Malpica A, Zuluaga A, Boiko I, Richards-Kortum R, Follen M. Assessment of cervical squamous intraepithelial lesion by optical coherence tomography and colposcopy. Lab Invest 2001;81:1331-8.
208. Margariti V, Zervakis M, Balas C. n.d.
209. Marin NM, Milbourne A, Rhodes H, Ehlen T, Miller D, Benedet L, et al. Diffuse reflectance patterns in cervical spectroscopy. Gynecol Oncol 2005;99:S116-20. http://dx.doi.org/10.1016/j.ygyno.2005.07.054.
210. Marsa L. Optical wand used to detect cervical cancer. Los Angeles Times 2004.
211. Martinho HD. Optical biopsy of pre-malignant or degenerative lesions: the role of the inflammatory process. Proc SPIE 2011;7890.
212. Massad LS, Collins YC. Strength of correlations between colposcopic impression and biopsy histology. Gynecol Oncol 2003;89:424-8. http://dx.doi.org/10.1016/S0090-8258(03)00082-9.
213. McLaren W, Tan J, Quinn M, Monsonego J. 5th International Multidisciplinary Congress (EUROGIN 2003), Paris, France, 13–16 April 2003. Bologna: Medimond Inc.; n.d.
214. Mehlhorn G, Munzenmayer C, Kage A, Benz M, Koch MC, Winter C, et al. Computer-assisted diagnosis (CAD) for colposcopy: evaluation of a pilot study. Endoskopie Heute 2010;23:290-6. http://dx.doi.org/10.1055/s-0030-1262651.
215. Mikhail MS, Palan PR, Basu J, Romney SL. Computerized measurement of intercapillary distance using image analysis in women with cervical intraepithelial neoplasia: correlation with severity. Acta Obstet Gynecol Scand 2004;83:308-10.
216. Milbourne A, Park SY, Benedet JL, Miller D, Ehlen T, Rhodes H, et al. Results of a pilot study of multispectral digital colposcopy for the in vivo detection of cervical intraepithelial neoplasia. Gynecol Oncol 2005;99:S67-75. http://dx.doi.org/10.1016/j.ygyno.2005.07.047.
217. Mirabal YN, Chang SK, Atkinson EN, Malpica A, Follen M, Richards-Kortum R. Reflectance spectroscopy for in vivo detection of cervical precancer. J Biomed Opt 2002;7:587-94. http://dx.doi.org/10.1117/1.1502675.
218. Mirkovic J, Lau C, McGee S, Yu C-C, Nazemi J, Galindo L, et al. Effect of anatomy on spectroscopic detection of cervical dysplasia. J Biomed Opt 2009;14. http://dx.doi.org/10.1117/1.3194142.
219. Mo J, Zheng W, Low J, Ng J, Ilancheran A, Huang Z. Near-infrared autofluorescence spectroscopy for in vivo diagnosis of cervical precancer. Proc SPIE 2008;6853. http://dx.doi.org/10.1117/12.762181.
220. Mo J, Zheng W, Low JJH, Ng J, Ilancheran A, Huang Z. High wavenumber Raman spectroscopy for in vivo detection of cervical dysplasia. Anal Chem 2009;81:8908-15. http://dx.doi.org/10.1021/ac9015159.
221. Mourant J, Kunapareddy P, Carpenter S, Freyer JP. Vibrational spectroscopy for identification of biochemical changes accompanying carcinogenesis and the formation of necrosis. Gynecol Oncol 2005;99:S58-60. http://dx.doi.org/10.1016/j.ygyno.2005.07.044.
222. Mourant JR, Bocklage TJ, Powers TM, Greene HM, Bullock KL, Marr-Lyon LR, et al. In vivo light scattering measurements for detection of precancerous conditions of the cervix. Gynecol Oncol 2007;105:439-45. http://dx.doi.org/10.1016/j.ygyno.2007.01.001.
223. Mourant JR, Bocklage TJ, Powers TM, Greene HM, Dorin MH, Waxman AG, et al. Detection of cervical intraepithelial neoplasias and cancers in cervical tissue by in vivo light scattering. J Low Genit Tract Dis 2009;13:216-23. http://dx.doi.org/10.1097/LGT.0b013e318195d91b.
224. Mourant JR, Powers TM, Bocklage TJ, Greene HM, Dorin MH, Waxman AG, et al. In vivo light scattering for the detection of cancerous and precancerous lesions of the cervix. Appl Opt 2009;48:D26-35. http://dx.doi.org/10.1364/AO.48.000D26.
225. Muller MG, Georgakoudi I, Feld MS, Chu S, Vuletic V, Kerman AJ, et al. Proceedings of the 15th International Conference on LASER Spectroscopy, Snowbird, UT, 10–15 June 2001. Singapore: World Scientific; 2002.
226. Murali Krishna C, Prathima NB, Malini R, Vadhiraja BM, Bhatt RA, Fernandes DJ, et al. Raman spectroscopy studies for diagnosis of cancers in human uterine cervix. Vib Spectrosc 2006;41:136-41. http://dx.doi.org/10.1016/j.vibspec.2006.01.011.
227. Muthuvelu K, Shanmugam S, Koteeswaran D, Srinivasan S, Venkatesan P, Aruna P, et al. Synchronous luminescence spectroscopic characterization of blood elements of normal and patients with cervical cancer. Proc SPIE 2011;7895.
228. Nath A, Rivoire K, Chang S, West L, Cantor SB, Basen-Engquist K, et al. A pilot study for a screening trial of cervical fluorescence spectroscopy. Int J Gynecol Cancer 2004;14:1097-107. http://dx.doi.org/10.1111/j.1048-891X.2004.14607.x.
229. Nordstrom RJ, Burke L, Niloff JM, Myrtle JF. Identification of cervical intraepithelial neoplasia (CIN) using UV-excited fluorescence and diffuse-reflectance tissue spectroscopy. Lasers Surg Med 2001;29:118-27. http://dx.doi.org/10.1002/lsm.1097.
230. Nour El-Din N, Mohii El-Din AR, Sanad AS, Hegazy AR. The use of colposcopy versus colposcopy and optical detection system in diagnosis of squamous intraepithelial lesions of the cervix: a clinical comparative study. Int J Gynaecol Obstet 2009;107. http://dx.doi.org/10.1016/S0020-7292(09)61059-6.
231. O’Connell MP, Tidy J, Wisher SJ, Avis NJ, Brown BH, Lindow SW. An in vivo comparative study of the pregnant and nonpregnant cervix using electrical impedance measurements. Br J Obstet Gynaecol 2000;107:1040-1. http://dx.doi.org/10.1111/j.1471-0528.2000.tb10410.x.
232. Okimoto GS, Parker MF, Mooradian GC, Saggese SJ, Grisanti A, O’Connor DM, et al. New features for detecting cervical pre-cancer using hyperspectral diagnostic imaging. Proc SPIE 2001;4255:67-79. http://dx.doi.org/10.1117/12.426747.
233. Orfanoudaki IM, Themelis GC, Sifakis SK, Fragouli DH, Panayiotides JG, Vazgiouraki EM, et al. A clinical study of optical biopsy of the uterine cervix using a multispectral imaging system. Gynecol Oncol 2005;96:119-31. http://dx.doi.org/10.1016/j.ygyno.2004.09.013.
234. Papoutsoglou G, Potirakis A, Balas C. Modeling of epithelia transport phenomena related with the acetowhitening optical changes: potential for the in-vivo diagnosis of cervical neoplasia. Proc SPIE 2008;6991.
235. Park SY, Follen M, Milbourne A, Rhodes H, Malpica A, MacKinnon N, et al. Automated image analysis of digital colposcopy for the detection of cervical neoplasia. J Biomed Opt 2008;13. http://dx.doi.org/10.1117/1.2830654.
236. Park SY, Sargent D, Lieberman R, Gustafsson U. Domain-specific image analysis for cervical neoplasia detection based on conditional random fields. IEEE Trans Med Imaging 2011;30:867-78. http://dx.doi.org/10.1109/TMI.2011.2106796.
237. Park SY, Sargent D, Wolters R, Lieberman RW. n.d. http://dx.doi.org/10.1109/ICSC.2010.85.
238. Parker MF, Mooradian GC, Karins JP, O’Connor DM, Speer BA, Owensby PD, et al. Hyperspectral diagnostic imaging of the cervix: report on a new investigational device. J Low Genit Tract Dis 2000;4:119-24.
239. Parker MF, Mooradian GC, Okimoto GS, O’Connor DM, Miyazawa K, Saggese SJ. Initial neural net construction for the detection of cervical intraepithelial neoplasia by fluorescence imaging. Am J Obstet Gynecol 2002;187:398-402. http://dx.doi.org/10.1067/mob.2002.123940.
240. Parker MF. Emerging technology in cervical cancer screening: spectroscopy. Clin Obstet Gynecol 2005;48:209-17. http://dx.doi.org/10.1097/01.grf.0000151586.23981.74.
241. Pfefer TJ, Sharma D, Agrawal A, Matchette LS. Evaluation of a fiberoptic-based system for optical property measurement in highly attenuating turbid media. Proc SPIE 2005;5691:163-71. http://dx.doi.org/10.1117/12.591101.
242. Pitris C, Drexler W, Ghanta RK, Kartner FX, Morgner U, Fujimoto JG, et al. n.d. http://dx.doi.org/10.1109/CLEO.2000.907315.
243. Pitris C, Hsiung P-L, Li X, Drexler W, Fujimoto JG, Goodman A, et al. n.d.
244. Pogue BW, Kaufman HB, Zelenchuk A, Harper W, Burke GC, Burke EE, et al. Analysis of acetic acid-induced whitening of high-grade squamous intraepithelial lesions. J Biomed Opt 2001;6:397-403. http://dx.doi.org/10.1117/1.1412850.
245. Porras C, Wentzensen N, Rodríguez AC, Morales J, Burk RD, Alfaro M, et al. Switch from cytology-based to human papillomavirus test-based cervical screening: implications for colposcopy. Int J Cancer 2012;130:1879-87. http://dx.doi.org/10.1002/ijc.26194.
246. Pretorius RG, Bao Y-P, Belinson JL, Burchette RJ, Smith JS, Qiao Y-L. Inappropriate gold standard bias in cervical cancer screening studies. Int J Cancer 2007;121:2218-24. http://dx.doi.org/10.1002/ijc.22991.
247. Qiang J, Engel J, Craine E. Classifying cervix tissue patterns with texture analysis. Pattern Recognition 2000;33:1561-73. http://dx.doi.org/10.1016/S0031-3203(99)00123-5.
248. Qu JY, Cheung TH, Hua J, Lo WK. Calibrated autofluorescence imaging for classification of tissue and detection of early cervical cancer. Proc SPIE 2001;4707:97-102.
249. Redden Weber C, Schwarz RA, Atkinson EN, Cox DD, MacAulay C, Follen M, et al. Model-based analysis of reflectance and fluorescence spectra for in vivo detection of cervical dysplasia and cancer. J Biomed Opt 2008;13. http://dx.doi.org/10.1117/1.3013307.
250. Robichaux A, Lieber CA, Shappell H, Huff B, Jones H, Mahadevan-Jansen A. In vivo detection of cervical dysplasia with near-infrared Raman spectroscopy. Proc SPIE 2002;4614:145-51. http://dx.doi.org/10.1117/12.460793.
251. Robichaux-Viehoever A, Kanter E, Shappell H, Billheimer D, Jones H. Characterization of Raman spectra measured in vivo for the detection of cervical dysplasia. Appl Spectrosc 2007;61:986-93. http://dx.doi.org/10.1366/000370207781746053.
252. Sanad AS, Nour El-Din N, Mohii El-Din AR, Hegazy AR. The use of colposcopy versus colposcopy and optical detection system in diagnosis of squamous intraepithelial lesions of the cervix: a clinical comparative study. Middle East Fertil Soc J 2011;16:137-42. http://dx.doi.org/10.1016/j.mefs.2010.12.005.
253. Sapozhnikova VV, Shakhova NM, Kamensky VA, Kuranov RV, Loshenov VB, Petrova SA. Complementary use of optical coherence tomography and 5-aminolevulinic acid induced fluorescent spectroscopy for diagnosis of neoplastic processes in cervix and vulva. Proc SPIE 2003;4956:81-8. http://dx.doi.org/10.1117/12.477738.
254. Sapozhnikova VV, Shakhova NM, Kamensky VA, Petrova SA, Snopova LB, Kuranov RV. Capabilities of fluorescence spectroscopy using 5-ALA and optical coherence tomography for diagnosis of neoplastic processes in the uterine cervix and vulva. Laser Physics 2005;15:1664-73.
255. Schomacker KT, Meese TM, Jiang C, Abele CC, Dickson K, Sum ST, et al. Novel optical detection system for in vivo identification and localization of cervical intraepithelial neoplasia. J Biomed Opt 2006;11. http://dx.doi.org/10.1117/1.2208987.
256. Sergeev AM, Gelikonov VM, Gelikonov GV, Feldchtein FI, Kuranov RV, Gladkova ND, et al. In vivo endoscopic OCT imaging of precancer and cancer states of human mucosa. Opt Express 1997;1:432-40. http://dx.doi.org/10.1364/OE.1.000432.
257. Shakhova NM, Kuznetsova IA, Kachalina TS, Yunusova EE. New approach to cervical cancer diagnosing using OCT-colposcopy [poster]. Cleveland, OH: Imalux Corporation; n.d.
258. Shakhova N, Kuznetzova I, Kachalina T, Gladkova N, Gelikonov V, Kamensky V, et al. 5th International Multidisciplinary Congress (EUROGIN 2003), Paris, France, 13–16 April 2003. Bologna: Medimond Inc.; n.d.
259. Shakhova NM, Gelikonov GV, Kamensky VA, Kuranov RV, Sapozhnikova VV, Turchin IV, et al. New advances in optical coherence tomography for diagnostics of cervical neoplasia. Proc SPIE 2003;5140:138-43. http://dx.doi.org/10.1117/12.500095.
260. Shakhova NM, Sapozhnikova VV, Kamensky VA, Kuranov RV, Losheno VB, Petrova SA, et al. Optical methods for diagnosis of neoplastic processes in the uterine cervix and vulva. J Appl Res 2003;3:144-55.
261. Shinn E, Le T, Gallegos J, Basen-Engquist K. A pilot analysis of multispectral digital colposcopy for women with high-grade squamous intraepithelial lesion (HGSIL) Pap smear results. Gynecol Oncol 2007;107:S83-5. http://dx.doi.org/10.1016/j.ygyno.2007.07.046.
262. Sokolov K, Nieman LT, Myakov A, Gillenwater A. Polarized reflectance spectroscopy for pre-cancer detection. Technol Cancer Res Treat 2004;3:1-14.
263. Srinivasan Y, Corona E, Nutter B, Mitra S, Bhattacharya S. A unified model-based image analysis framework for automated detection of precancerous lesions in digitized uterine cervix images. IEEE J Sel Top Signal Process 2009;3. http://dx.doi.org/10.1109/JSTSP.2008.2011102.
264. Sung K-B, Richards-Kortum R, Follen M, Malpica A, Liang C, Descour MR. Fiber optic confocal reflectance microscopy: a new real-time technique to view nuclear morphology in cervical squamous epithelium in vivo. Opt Express 2003;11:3171-81. http://dx.doi.org/10.1364/OE.11.003171.
265. Tan J, Quinn MA, Pyman JM, Delaney PM, McLaren WJ. Detection of cervical intraepithelial neoplasia in vivo using confocal endomicroscopy. BJOG 2009;116:1663-70. http://dx.doi.org/10.1111/j.1471-0528.2009.02261.x.
266. Trokhanova OV, Chijova YA, Okhapkin MB, Korjenevsky AV, Tuykin TS. Using of electrical impedance tomography for diagnostics of the cervix uteri diseases. JPCS 2010;224.
267. Tromberg BJ, Yodh AG, Tamura M, Sevick-Muraca EM. n.d.
268. Utzinger U, Heintzelman DL, Mahadevan-Jansen A, Malpica A, Follen M, Richards-Kortum R. Near-infrared Raman spectroscopy for in vivo detection of cervical precancers. Appl Spectrosc 2001;55:955-9. http://dx.doi.org/10.1366/0003702011953018.
269. Van Raad V. n.d.
270. Van Raad V, Hamza MH. IASTED International Conference on Biomedical Engineering; Salzburg, Austria, 25–27 June 2003. Anaheim, CA: ACTA Press; n.d.
271. Van Raad V, Liu Y, Jiang T, Zhang C. Computer vision for biomedical image applications. First international workshop, CVBIA 2005, Beijing, China, 21 October 2005. Proceedings. Berlin: Springer-Verlag; 2005.
272. Van Raad V, Xue Z, Lange H. Lesion margin analysis for automated classification of cervical cancer lesions. Proc SPIE 2006;6144.
273. Vargas G, Shilagard T, Johnston R, Bell B, Stegall RL, Vincent K, et al. Use of high-resolution confocal imaging of the vaginal epithelial microstructure to detect microbicide toxicity. J Infect Dis 2009;199:1546-52. http://dx.doi.org/10.1086/598221.
274. Vargis E, Byrd T, Khabele D, Mahadevan-Jansen A. Using Raman spectroscopy to detect cervical dysplasia in minority populations. Gynecol Oncol 2010;116.
275. Vargis E, Byrd T, Reese J, Khabele D, Al-Hendy A, Mahadevan-Jansen A. Detecting changes in the cervix with Raman spectroscopy. AIP Conf Proc 2010;1267:441-2. http://dx.doi.org/10.1063/1.3482605.
276. Vargis E, Kanter E, Kanter G, Rao G, Mahadevan-Jansen A. Proximity to disease has a significant impact on automated disease classification of normal tissues. Gynecol Oncol 2010;116:S33-4.
277. Vargis E, Kanter EM, Majumder SK, Keller MD, Beaven RB, Rao GG, et al. Effect of normal variations on disease classification of Raman spectra from cervical tissue. Analyst 2011;136:2981-7. http://dx.doi.org/10.1039/c0an01020k.
278. Vengadesan N, Muthuvelu K, Anbupalam T, Madhuri S, Parmeswaran D, Aruna P, et al. Optical biopsy of normal and cancerous cervical tissues by native fluorescence spectroscopy. Proc SPIE 2001;4261:145-9. http://dx.doi.org/10.1117/12.424530.
279. Vengadesan N, Anbupalam T, Hemamalini S, Ebenezar J, Muthvelu K, Koteeswaran D, et al. Characterization of cervical normal and abnormal tissues by synchronous luminescence spectroscopy. Proc SPIE 2002;4613:13-7. http://dx.doi.org/10.1117/12.465247.
280. Vincent KL, Bell BA, Rosenthal SL, Stanberry LR, Bourne N, Sweeney YTC, et al. Application of optical coherence tomography for monitoring changes in cervicovaginal epithelial morphology in macaques: potential for assessment of microbicide safety. Sex Transm Dis 2008;35:269-75. http://dx.doi.org/10.1097/OLQ.0b013e31815abad8.
281. Vincent KL, Bourne N, Bell BA, Vargas G, Tan A, Cowan D, et al. High resolution imaging of epithelial injury in the sheep cervicovaginal tract: a promising model for testing safety of candidate microbicides. Sex Transm Dis 2009;36:312-18. http://dx.doi.org/10.1097/OLQ.0b013e31819496e4.
282. Weingandt H, Stepp H, Baumgartner R, Diebold J, Xiang W, Hillemanns P. Autofluorescence spectroscopy for the diagnosis of cervical intraepithelial neoplasia. BJOG 2002;109:947-51. http://dx.doi.org/10.1111/j.1471-0528.2002.01311.x.
283. Werner CL, Griffith WF, Ashfaq R, Gossett D, Wilkinson E, Raab S, et al. Comparison of human papilloma virus testing and spectroscopy combined with cervical cytology for the detection of high-grade cervical neoplasia. J Low Genit Tract Dis 2007;11:73-9. http://dx.doi.org/10.1097/01.lgt.0000230208.58118.58.
284. Wilkinson EJ, Raab SS. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: histopathology review procedures and results. Norcross, GA: Guided Therapeutics; 2010.
285. Winter ML. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: experience with a low cost commercial prototype. Norcross, GA: Guided Therapeutics; 2010.
286. Winter ML, Sternfeld DR. Multimodal spectroscopy as a triage test for women at risk for cervical neoplasia: experience with a low cost commercial prototype. J Low Genit Tract Dis 2010;14.
287. Wu T, Cheung TH, Lo KW, Qu JY. Detecting neoplastic growths in vivo with autofluorescence imaging. Proc SPIE 2003;4958:71-7. http://dx.doi.org/10.1117/12.479804.
288. Wu T, Qu JY, Cheung T-H, Lo KW-K, Yu M-Y. Preliminary study of detecting neoplastic growths in vivo with real time calibrated autofluorescence imaging. Opt Express 2003;11:291-8. http://dx.doi.org/10.1364/OE.11.000291.
289. Wu TT, Cheung TH, Lo KW, Wen Y, Qu JY. Fluorescence detection of neoplastic growths in vivo. Diagnostic Optical Spectroscopy in Biomedicine II 2003;5141:58-63.
290. Wu TT, Cheung TH, Wen Y, Qu JY. n.d. http://dx.doi.org/10.1109/COS.2003.1278205.
291. Wu TT, Cheung T-H, Yim S-F, Qu JY. Optical imaging of cervical precancerous lesions based on active stereo vision and motion tracking. Opt Express 2008;16:11224-30. http://dx.doi.org/10.1364/OE.16.011224.
292. Wu TT, Cheung T-H, Yim S-F, Qu JY. Quantitative optical imaging of early cervical cancer: mechanisms, methods, and clinical study. Proc SPIE 2009;7187.
293. Wu T, Cheung T-H, Yim S-F, Qu JY. Clinical study of quantitative diagnosis of early cervical cancer based on the classification of acetowhitening kinetics. J Biomed Opt 2010;15. http://dx.doi.org/10.1117/1.3365940.
294. Wulan N, Rasool N, Belinson SE, Belinson JL, Wang C, Rong X, et al. Clevaland, OH: Imalux Corporation; 2008.
295. Wulan N, Rasool N, Belinson SE, Wang C, Rong X, Zhang W, et al. Study of the diagnostic efficacy of real-time optical coherence tomography as an adjunct to unaided visual inspection with acetic acid for the diagnosis of preinvasive and invasive neoplasia of the uterine cervix. Int J Gynecol Cancer 2010;20:422-7. http://dx.doi.org/10.1111/IGC.0b013e3181d09fbb.
296. Yang B, Pretorius RG, Belinson JL, Zhang X, Burchette R, Qiao Y-L. False negative colposcopy is associated with thinner cervical intraepithelial neoplasia 2 and 3. Gynecol Oncol 2008;110:32-6. http://dx.doi.org/10.1016/j.ygyno.2008.03.003.
297. Yu Y, Huang J, Zhang S, Restif C, Huang X, Metaxas D. 2011.
298. Zagaynova E, Gladkova N, Shakhov A, Terentyeva A, Snopova L, Shakhov N, et al. Optical coherence tomography: potentialities in clinical practice. Proc SPIE 2004;5474:103-14. http://dx.doi.org/10.1117/12.580771.
299. Zagaynova E, Gladkova N, Shakhova HM, Gelikonov G, Gelikonov V. Endoscopic OCT with forward-looking-probe: clinical studies in urology and gastroenterology. J. Biophotonics 2008;1:114-28. http://dx.doi.org/10.1002/jbio.200710017.
300. Zara JM, Lingley-Papadopoulos CA. Endoscopic OCT approaches toward cancer diagnosis. IEEE J Sel Top Quantum Electron 2008;14:70-81. http://dx.doi.org/10.1109/JSTQE.2007.912020.
301. Zertuche J, Pena A. A new colposcopic technique in uterine cervical cancer. Int J Gynaecol Obstet 2009;107:S605-6. http://dx.doi.org/10.1016/S0020-7292(09)62163-9.
302. Zhang S, Huang J, Metaxas D, Wang W, Huang X. Discriminative sparse representations for cervigram image segmentation n.d.
303. Zhao H, Liang J, Ma J, Yang Z. n.d.
304. Zhao F-H, Zhang W-H, Pan Q-J, Zhang X, Chen W, Liu B, et al. A study of cervical cancer screening algorithms. Chung Hua Chung Liu Tsa Chih 2010;32:420-4.

Appendix 1 Literature search strategies

Literature search strategies (PDF download)

Appendix 2 Data extraction tables

Data extraction tables (PDF download)

Appendix 3 Quality assessment

Quality assessment (PDF download)

Appendix 4 Table of excluded studies with rationale

Table of excluded studies with rationale (PDF download)

Appendix 5 Sensitivity analysis of the base case

Sensitivity analysis of the base case (PDF download)

Appendix 6 Sensitivity analyses of the secondary analysis quality-adjusted life-year decrement of treatment biopsy 0.13 (from 0.005)

Sensitivity analyses of the secondary analysis quality-adjusted life-year decrement of treatment biopsy 0.13 (from 0.005) (PDF download)

Appendix 7 Secondary analysis cost of treatment biopsy £2758 sensitivity analyses

Secondary analysis cost of treatment biopsy £2758 sensitivity analyses (PDF download)

Appendix 8 Protocol (submitted 22 September 2011)

Protocol (submitted 22 September 2011) (PDF download)

Glossary

Technical terms and abbreviations are used throughout this report. The meaning is usually clear from the context but a glossary is provided for the non-specialist reader.

Acetowhitening

Whitening effect following application of acetic acid to epithelial tissue, which is a sign of increased nuclear protein.

Adverse effect

An abnormal or harmful effect caused by, and attributable to, exposure to a medication or other intervention, which is indicated by some result such as death, a physical symptom or visible illness. An effect may be classed as adverse if it causes functional or anatomical damage, causes irreversible change in the homeostasis of the organism, or increases the susceptibility of the organism to other chemical or biological stress.

APX 100

A digital image analysing system for detecting cancerous and precancerous cervical tissue. It works by measuring the resistivity (via electrical impedance spectroscopy) of cervical epithelial cells.

Cervical intraepithelial neoplasia

A term describing abnormal changes in the squamous epithelial cells of the cervix. The disorder is graded according to its pathological progress, from CIN1 to CIN3.

Colposcope

A magnifying instrument designed to facilitate visual inspection of the cervix.

Correlation meeting

A meeting where the pathologists and colposcopists discuss the results and the management of patients who have clear colposcopic findings, but moderate or severe cytology results.

DySIS

A digital video colposcope using dynamic spectral imaging for detecting cancerous and precancerous cervical tissue. It works, following application of acetic acid, by mapping the acetowhitening of the epithelium of the cervix (the DySISmap). [Note: Subsequent to the production of this report, DySIS Medical informed the assessment group that the current terminology for the DySIS technology is ‘DySIS colposcopy’when referring to the DySISmap and colposcopy combined, and ‘DySISmap’when referring to the DySISmap alone (this was previously known as ‘DSI map’or ‘DSI colour-coded map’).]

Dyskaryosis

A term describing abnormality of the cell nucleus (but not the cytoplasm).

Electrical impedance spectroscopy

A form of spectroscopy that works by utilising electric current patterns.

Histology

An abbreviation of histopathology.

Histopathology

The microscopic study of tissue samples to enable diagnosis.

Human papillomavirus

A type of virus that can affect the skin and the moist membranes lining parts of the body. Some types of human papillomavirus (known as high-risk human papillomaviruses) can cause dyskaryosis in the cells of the cervix.

Liquid-based cytology

A method of preparing cervical samples for laboratory examination.

LuViva Advanced Cervical Scan

A digital image analysing system for detecting cancerous and precancerous cervical tissue. It works by detecting biochemical and morphological changes at the cellular level (using optical spectroscopy).

NHS Cervical Screening Programme

The programme set up in the UK aimed at detecting and treating early abnormalities which, if left untreated, could lead to cervical cancer.

Niris Imaging System

A digital image analysing system for detecting cancerous and precancerous cervical tissue. It works using optical coherence tomography to produce a two-dimensional image of the tissue.

Optical coherence tomography

A technique for creating two- or three-dimensional cross-sectional images of tissue using infrared light.

Pathologist

The individual responsible for examining and interpreting cell and/or tissue samples.

Quality of life

A concept incorporating all the factors that might impact on an individual's life, including factors such as the absence of disease or infirmity, as well as other factors that might affect the individual's physical, mental and social well-being.

Quality-adjusted life-year

An index of health gain by which survival duration is weighted or adjusted by the patient's quality of life during the survival period. Quality-adjusted life-years have the advantage of incorporating changes in both quantity (mortality) and quality (morbidity) of life.

See and treat

The removal of an abnormal area during colposcopy.

Spectroscopy

An analytical method for studying the structural and biochemical features of tissue, most commonly by utilising electromagnetic spectra readings.

Speculum

An instrument for opening a body cavity in order to allow visual inspection.

Statistical significance

An estimate of the probability of an association (effect) as large or larger than what is observed in a study occurring by chance, usually expressed as a p-value.

Threshold analysis

Amount of variation needed in the parameter values of a model to achieve a specified outcome. In the context of cost-effectiveness analysis in the UK NHS, this specified outcome is usually the cost-effectiveness threshold of £20,000–30,000 per additional QALY gained.

Transformation zone

An area of the cervix where nearly all precancerous and cancerous changes occur.

List of abbreviations

AGUS

atypical glandular cells of undetermined significance

AiC

academic in confidence

ASC-H

atypical squamous cells with possible high-grade squamous intraepithelial lesion

ATP

according to protocol

CE

Conformité Européenne

CI

confidence interval

CIN

cervical intraepithelial neoplasia

CRD

Centre for Reviews and Dissemination

DSI

dynamic spectral imaging

DySIS

dynamic spectral imaging system

EAG

External Assessment Group

GP

general practitioner

HPV

human papillomavirus

hrHPV

high-risk human papillomavirus

HRQoL

health-related quality of life

HSIL

high-grade squamous intraepithelial lesion

ICER

incremental cost-effectiveness ratio

ITT

intention to treat

LBC

liquid-based cytology

LLETZ

large-loop excision of the transformation zone

LR

likelihood ratio

lrHPV

low-risk human papillomavirus

LSIL

low-grade squamous intraepithelial lesion

NICE

National Institute for Health and Clinical Excellence

NIHR

National Institute for Health Research

NPV

negative predictive value

NR

not reported

OCT

optical coherence tomography

ONS

Office for National Statistics

PCM

pseudocolour map

PPV

positive predictive value

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

QALY

quality-adjusted life-year

SIGN

Scottish Intercollegiate Guidelines Network

STARD

STAndards for the Reporting of Diagnostic accuracy studies

TOMBOLA

Trial of Management of Borderline and Other Low-Grade Abnormal Smears