This article has Open Peer Review reports available.
5-Aminolevulinic-acid-based fluorescence spectroscopy and conventional colposcopy for in vivo detection of cervical pre-malignancy
© Vansevičiūtė et al.; licensee BioMed Central. 2015
Received: 11 November 2014
Accepted: 27 March 2015
Published: 17 April 2015
Sensitized fluorescence diagnostics are based on selective accumulation of photosensitizer in the tissue where carcinogenesis has started. The present study compared topical 5-aminolevulinic acid (5-ALA)-based fluorescence spectroscopy (FS) in vivo with conventional colposcopy for cervical intraepithelial neoplasia (CIN) detection.
We enrolled 48 patients who were referred for colposcopy because of high-grade changes in cervical cytology. Every inspected cervix was divided in to quadrants, and there were 174 quadrants included in the study. Each patient had a cytological smear, colposcopy, FS and histopathological analysis. For FS, 3% 5-ALA cream was used topically and after an average 135 min incubation, fluorescence spectra were recorded from the cervix in vivo. FS and colposcopy results were correlated with histopathology.
All spectra were evaluated by a ratio of the protoporphyrin IX fluorescence intensity at 634 nm and autofluorescence intensity at 510 nm. For proper grouping of low-risk and high-risk cases, a threshold of 3.87 was calculated. Data per quadrant showed that FS had higher sensitivity than colposcopy (71.7% vs 67.4%) but specificity was greater for colposcopy (86.6% vs 75.6%). Combination of the methods showed higher sensitivity (88.0% vs 67.4%) but reduced specificity (88.0% and 69.5%), but it had the highest number of correctly identified high-risk changes and the highest (79.3%) accuracy. Data for each patient showed FS sensitivity of 91.2%, which was greater than for colposcopy (88.2%). Higher overdiagnosis resulted in decreased specificity for fluorescence methodology—71.4% versus 78.6% for colposcopy. In both cases, accuracy was 85.4% and effectiveness was >80%, which means that both methods can be used to determine high-risk cervical intraepithelial neoplasia. The diagnostic sensitivity of 97.1% for this complementary diagnosis indicates that it could be the best choice for detection of high-risk changes.
5-ALA-based FS is an objective method, requiring short-term administration for appropriate fluorescence measurements. FS is a promising diagnostic tool with similar accuracy as colposcopy but with the potential advantage of providing objective results.
Worldwide, cervical cancer is the third most common cancer among women, with an age-standardized incidence rate of 15.3 per 100,000 and mortality rate of 7.8 per 100,000 . Cervical cancer mostly affects younger women aged 35–50 . In 2011, 452 new cases of cervical cancer were reported in Lithuania, which is one of the highest rates of morbidity for cervical cancer rate among the Baltic and Nordic countries [3,4].
Cervical intraepithelial neoplasia (CIN) is the potentially premalignant transformation and abnormal growth of epithelial cells on the surface of the cervix. A premalignant lesion is always at a risk of malignant transformation if stimulated by certain exogenous factors or conditions. The main factor directly related to CIN development is persistent human papillomavirus infection; mainly with high-risk types 16, 18, 31 and 33 [5,6]. If CIN is diagnosed at the appropriate time before cervical cancer manifestation, it may be cured and cervical cancer avoided [7,8].
The main disadvantages of contemporary CIN detection methods are high false-negative rates and low sensitivity of cytology and low specificity of colposcopy. Some studies showed that, in most cases, patients with high-grade smear results and high-grade impression on colposcopy have an acceptable overtreatment rate [9,10]. However conventional colposcopy demands long-term training and achieves an average ~48% (23–87%) specificity and has unsatisfactory accuracy even in trained hands [7,11-13]. The low positive predictive value seen with conventional colposcopy results in unwarranted surgical procedures and an additional burden on cervical cancer screening programs. These data suggest that there is a need for new diagnostic methods to improve or replace colposcopy for a clearer CIN diagnosis and more individual approach.
Recently, there has been increased interest in optical biopsy to determine pathological diagnoses in various organs. Optical biopsy refers to any technique that uses the interaction of light and tissue to provide information about tissue morphology without the need for excision [14-18]. Premalignant and malignant tissue differs from healthy tissue in its morphology and cell growth rate, which results in altered optical characteristics [16,19-21]. Most of the optical methods used in diagnostics are based on different types of spectroscopy: fluorescence, near infrared, Raman, diffuse reflectance, and similar techniques [22-25]. However, the most widely used techniques in clinical practice are based on fluorescence [26,27]. When light interacts with the molecules in the tissue they become excited and may re-emit light of different colors (fluorescence). The fluorescence of the tissue can be traced by sensitive spectrometers and provide characteristic spectra that reflect carcinogenicity of the tissue [22,23,27]. The acceptance and suitability of these methods clinically are determined by the diagnostic effectiveness, simplicity and low cost of the procedures. Moreover, they are noninvasive and can be repeated many times.
The potential of the fluorescence method to identify normal and pathological tissues of the uterine cervix was raised in 1994 [28,29]. The possibility was investigated of autofluorescence being used for detection of cervical neoplasia. Despite the demonstrated diagnostic possibilities, tissue spectra vary both among patients and within each individual patient. That complicates application of the method, and it is difficult to maintain the exact measurement conditions that could additionally influence the intensity and spectral variations . The developed algorithms partially resolve this problem but the lack of sufficient contrast between fluorescence of healthy and neoplastic tissue encourages the use of sensitized fluorescence.
One of the most widely used precursor exogenous molecules in medical applications is 5-aminolevulinic acid (5-ALA). 5-ALA is a precursor of the fluorescent endogenous fluorophore, protoporphyrin IX (PpIX); an excess of which is produced in altered, especially cancerous tissue. This results in accumulation of intracellular porphyrins, which increases tissue fluorescence in the red spectral region of cancerous tissue . Subsequent irradiation of the lesion with visible light matching the highest absorption of PpIX (~405 or ~630 nm) leads to red fluorescence emission from PpIX, peaking at 635 nm. The different accumulation of endogenous fluorophores in cancerous and normal tissue causes differences in the red fluorescence ratio between healthy and premalignant/malignant tissue, which makes detection and analysis by fluorescence light more applicable to discrimination between malignant and nonmalignant tissues . This intensity is governed by the biological object itself, as well as many other conditions, such as variations in different preparations, doses and forms of application of 5-ALA, as well as different light sources and incubation times.
The aim of this research was to compare the ability of conventional colposcopy and 5-ALA-based FS for detection of CIN. From the fluorescence measurements, we established low-risk and high-risk tissues for cancer development. We created a methodology to identify from the fluorescence spectra a universal independent threshold value for determining high-risk tissues for cancer development, eliminating possible measurement variations, depending on individual characteristics of the organism and technical errors.
The study was carried out at the National Cancer Institute in Lithuania between December 2012 and January 2014. The investigation protocol was approved by the Vilnius Regional Research Ethics Committee and written informed consent was obtained from all women. Inclusion criteria were nonpregnant women aged >18 years with suspected high-grade changes in cervical cytology, with no allergies or history of porphyria. Exclusion criteria were unsatisfactory or absent histopathological analysis, or inappropriate colposcopy or FS measurements.
Data of 18 cervical quadrants were excluded owing to absence of histopathological analysis (uncertain marking or absence of the quadrants after loop excision or biopsy ( cases), coagulation defects ( cases)) and inappropriate FS measurements (bleeding from the cervical tissue during the investigation ( cases), device failure during measurement [1 case]).
After colposcopy and FS, punch biopsy or loop electrosurgical excision of the cervix for histopathological analysis was performed. These tissue samples were diagnosed as inflammatory (chronic cervicitis) (CIN0), mild (CIN1), moderate (CIN2) or severe (CIN3) neoplastic. All cases were divided in two clinically significant groups: low risk for cancer development (CIN0-1) and high risk for cancer development (CIN2-3). The histopathological diagnosis was provided for every quadrant of the cervix. However, the diagnosis of separate cervical quadrant (per quadrant) and subsequent final diagnosis of the whole cervix (per patient) was concluded according to the highest degree of the neoplasia discovered histopathologically.
Using Origin Pro 8, we performed statistical analysis of the D values for all quadrants. To determine the normality of the distribution, we applied the Shapiro–Wilk test, when the significance level was 0.05. We identified that our data were not distributed normally, so for further analysis, we applied nonparametric tests. To inspect the raised hypotheses, we performed the Mann Whitney Wilcoxon two-independent-samples tests. The statistically significant difference was when P(H0 = 0) < 0.05. For proper grouping of LR and HR cases (D values), we calculated the threshold value. For this reason Youden’s index (J) was assessed. According to this criterion, the threshold value is when J is the greatest. Statistical analysis of the collected material included sensitivity, specificity, positive and negative predictive values for FS, colposcopy and combination of those methods, analyzing diagnosis per patient and per cervical quadrant. For evaluation of the method diagnostic efficiency, receiver operating characteristic curves (ROCs) were plotted. The diagnostic value was obtained by assessing area under the curve (SROC). Area under the curve was 0.9–1.0, the study could be evaluated as excellent; 0.8–0.9, very good; 0.7–0.8, good; 0.6–0.7, tolerable; and 0.5–0.6, deficient [39,40]. ROC analysis was performed using MedCalc.
FS data analysis
Histopathological diagnosis of analyzed cases
Total per patient
Total per cervical quadrant
n = 48
n = 174
The higher fluorescence intensity in the red spectrum region was detected from neoplastic tissue compared with healthy tissue. Normalization of fluorescence spectra to autofluorescence peaks helps to overcome varying measurement conditions and allow unambiguous evaluation of fluorescence spectra from different places. The spectrum measured on the skin did not have a peak specific for PpIX, which means that ALA cream used topically on the cervix causes no PpIX production in other parts of the body, showing no systemic accumulation and therefore causing no notable side effects.
Calculation of the threshold value
Feasibility of FS for high/low risk cervical neoplasia diagnosis
Comparison of fluorescence spectroscopy and colposcopy with histopathology (per quadrant)
Fluorescence spectroscopy + colposcopy
In determining the status of the cervical quadrants by colposcopy, 73 quadrants were high-risk and 101 were low-risk neoplastic. According to these results, the disease was missed in 30 quadrants, while only 11 cases were incorrectly diagnosed as high risk.
One of the potential applications of fluorescence diagnostics is as a complementary method to conventional techniques. High-risk neoplasia was considered if at least one method detected neoplastic changes. The data in Table 2 show that when colposcopy and fluorescence were combined, only 11 cases were missed, and combination of methods gave the best results. However, the 25 low-risk quadrants were mistakenly identified as high risk, which led to unnecessary surgical procedures.
Sensitivity, specificity, accuracy, positive and negative predictive values for fluorescence spectroscopy, colposcopy and combination of methods, analyzing cervical quadrant diagnosis
Colposcopy + fluorescence (combined)
Comparison of fluorescence spectroscopy and colposcopy with histopathology (per patient)
Fluorescence spectroscopy + colposcopy
Per patient analysis showed that FS failed to detect only three high-risk cases from 34, while during colposcopy, four cases were missed. However, colposcopy overdiagnosed three patients and FS four. These results show that the two methods are comparable. Combination of the methods increases the number of correctly diagnosed neoplasia (Table 4) and only one high-risk case was missed. However, overdiagnosis also increased, with six patients being mistakenly diagnosed with high-risk changes.
Full statistical analysis for the per patient results is presented in Table 5
Sensitivity, specificity, accuracy, positive and negative predictive values for fluorescence spectroscopy, colposcopy and combination of methods, analyzing per patient diagnosis
Colposcopy + fluorescence (combined)
Sensitivity of FS was 91.2%, which was greater than 88.2% for colposcopy. Greater overdiagnosis resulted in decreased sensitivity for FS—71.4% vs 78.6% for colposcopy. In both cases, the accuracy was 85.4%, and the effectiveness (SROC) was >80%, which means that the value in determining the high-risk CIN was good for both methods. On the contrary, the effectiveness of the combined methods was <80%. Nevertheless, the sensitivity of this combined diagnosis was 97.1%, indicating that it was best for detection of high-risk lesions.
We evaluated the use of 5-ALA-based FS for the detection of pathological areas in the cervix. Identification of the diseased areas was estimated by measuring the fluorescence of PpIX, and more quantitative evaluation could be done by calculating the ratio (R) between porphyrin fluorescence and tissue autofluorescence. The additional normalization of ratios (R) for the inspected areas of the tissue to the (Rhealthy) value of healthy tissue enabled us to calculate the diagnostic coefficients (D), and to obtain threshold values for low- and high-risk differentiation. The threshold value of 3.87 is the universal independent number that could be used for CIN diagnosis.
Examination of the cervical quadrants yields the possibility of specifying the localization of neoplastic regions. However, because they were observed during the study, the pathological areas could be small and missed during measurement. Therefore, for precise localization of all lesions, careful and fine scanning should be performed. In other cases, accurate diagnostics could only be performed per patient. The results of FS per quadrant in some cases were influenced by difficulties during conization, which resulted in incorrect margins of the quadrants and mixed correct results. However, during routine procedures, such situations cannot be fully avoided, therefore, such inaccuracies must be incorporated when calculating the sensitivity and specificity as a random error. Nevertheless, these effects do not influence per patient diagnosis.
Per patient diagnosis showed 85.4% accuracy of FS in identifying neoplastic changes, therefore, it has great diagnostic potential. However, to replace the conventional methods, the threshold value must be revised and, if necessary, corrected after data analysis from more patient cases.
Several studies show that the detection of porphyrin fluorescence can improve the identification of the cervical pre-malignancy. Only a few studies have been performed using FS with 5-ALA. Hillemanns et al.  showed that fluorescence imaging with 5-ALA after 60–90 minutes achieved similar sensitivity and specificity compared with colposcopy in detecting CIN: 94% and 51% versus 95% and 50%, respectively. However, the specificity was markedly improved by FS, achieving 75%. Sapoznikova et al.  reported that sensitized FS had a diagnostic efficiency of 79.5% for cervical neoplasia.
When comparing colposcopy with 5-ALA-based FS, on the one hand we have the experience of physicians and on the other hand, we have the objectivity of the diagnostic method, and one also should decide between overestimated and underestimated CIN diagnosis. To date, it seems that this should be a compromise because there is still no single method or combination of methods that can perfectly fulfill all the desired requirements.
Nevertheless, fluorescence diagnostics appears promising despite non- ideal results. Fluorescence is an objective method, requiring only a short training for appropriate fluorescence measurements, while colposcopy is highly dependent on physicians’ experience and might be more inaccurate when performed by less-trained hands. There is plenty of room for improving the precision and consistency of fluorescence measurement techniques. The additional burden for FS was because the analysis was performed on patients who already had high-grade cytology results. Altered cytology could determine greater production of PpIX, which results in elevated fluorescence intensity and finally provide false-positive results and decreased specificity. Moreover, the women who were diagnosed as having no evidence or low risk of neoplasia had significant signs of inflammation. The inflammation is not classified as malignancy, but it changes the optical properties of cervical tissue, therefore, the ability to differentiate inflammation from neoplastic changes should further increase the diagnostic value of FS.
A possible clinical application could be a combination of a porphyrin fluorescence image viewer and an in vivo spectrum analysis system, which could be used for optical biopsies. With this system, high-grade pre-cancer could be diagnosed and treated in one session [32-34,41].
It is possible to define further directions for development. FS is a powerful technique for detecting altered tissues, however, additional attention should be paid to differentiation/identification of the neoplastic changes and inflammation.
By standardization of instrumentation and proper diagnostic algorithms it was possible to obtain an independent threshold value for low- and high-risk differentiation, Dthreshold = 3.87. Analysis using this threshold showed the potential of FS for noninvasive identification of neoplastic changes at an early stage, and it could be used for differentiating high-risk changes. This method shows similar accuracy to currently used screening tests, but has the potential advantage of objective results. FS requires only a short time of training to achieve appropriate measurements, while colposcopy is highly dependent on physicians’ experience, which is usually considered a major factor in its success.
The authors would like to thank the following surgeons at the National Cancer Institute oncogynaecological department for the performed invasive procedures of this study: Drs. Rūta Čiurlienė, Birutė Intaitė, Kastytis Žilinskas, Eugenijus Drulia, Tomas Razumas and the entire operating room staff from this department. The authors would like to thank Dr. pathologist Jolita Rimienė and the entire staff of the „Pathology diagnostics“for providing histopathological analysis. The authors would like to thank National Cancer Institute administration for the covering the cost of study organization and article publication.
- Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. 2010;127:2893–917.View ArticlePubMedGoogle Scholar
- Arbyn M, Anttila A, Jordan J, Ronco G, Schenck U, Segnan N, et al. European guidelines for quality assurance in cervical cancer screening. Second edition—summary document. Ann Oncol. 2010;21(3):448–58.View ArticlePubMedPubMed CentralGoogle Scholar
- Lithuanian Cancer Registry Available from: http://www.nvi.lt/index.php?-1413089819.
- Arbyn M, Antoine J, Magi M, Smailyte G, Stengrevics A, Suteu O, et al. Trends in cervical cancer incidence and mortality in the Baltic countries. Bulgaria and Romania Int J Cancer. 2011;128:1899–907.View ArticlePubMedGoogle Scholar
- Wheeler CM, Hunt WC, Cuzick J, Langsfeld E, Robertson M, Castle PE. New Mexico HPV Pap registry steering committee the influence of type-specific human papillomavirus infections on the detection of cervical precancer and cancer: a population-based study of opportunistic cervical screening in the united states. Int J Cancer. 2014;135(3):624–34.View ArticlePubMedPubMed CentralGoogle Scholar
- Nair S, Pillai MR. Human papillomavirus and disease mechanisms: relevance to oral and cervical cancers. Oral Dis. 2005;11:350–9.View ArticlePubMedGoogle Scholar
- Singer A, Monaghan MJ, Quek SC. Lower Genital Tract Precancer. Colposcopy, pathology and treatment. 2nd ed. Oxford. Blacwell science; 2008.Google Scholar
- Tan JHJ, Wrede CDH. New technologies and advances in colposcopic assessment. Best Practice & Research Clinical Obstetrics and Gynaecology. 2011;25:667–77.View ArticlePubMedGoogle Scholar
- Bosgraaf RP, Mast PP, Struik-van der Zanden PH, Bulten J, Massuger LF, Bekkers RL. Overtreatment in a see-and-treat approach to cervical intraepithelial lesions. Obstet Gynecol. 2013;121(6):1209–16.View ArticlePubMedGoogle Scholar
- Aue-Aungkul A, Punyawatanasin S, Natprathan A, Srisomboon J, Kietpeerakool C. "See and treat" approach is appropriate in women with high-grade lesions on either cervical cytology or colposcopy. Asian Pac J Cancer Prev. 2011;12(7):1723–6.PubMedGoogle Scholar
- Palma PD, Rossi PG, Collina G, Buccoliero AM, Ghiringhello B, Lestani M, et al. The NTCC pathology group the risk of false-positive histology according to the reason for colposcopy referral in cervical cancer screening. Am J Clin Pathol. 2008;129:75–80.View ArticlePubMedGoogle Scholar
- Cremer ML, Peralta EI, Dheming SG, Jimenez ME, Davis-Dao CA, Alonzo TA, et al. Digital assessment of the reproductive tract versus colposcopy for directing biopsies in women with abnormal Pap smears. Journal of Lower Genital Tract Disease. 2010;14(1):5–10.View ArticlePubMedGoogle Scholar
- Guido RS, Jeronimo J, Schiffman M, Solomon D. ALTS Group The distribution of neoplasia arising on the cervix: results from the ALTS trial. Am J Obstet Gynecol. 2005;193(4):1331–7.View ArticlePubMedGoogle Scholar
- Wang TD, Van Dam J. Optical biopsy: a new frontier in endoscopic detection and diagnosis. Clin Gastroenterol Hepatol. 2004;2(9):744–53.View ArticlePubMedPubMed CentralGoogle Scholar
- Alfano RR. Advances in optical biopsy for cancer diagnosis. Technol Cancer Res Treat. 2011;10:101.View ArticlePubMedPubMed CentralGoogle Scholar
- Costas B. Review of biomedical optical imaging—a powerful, non-invasive, non-ionizing technology for improving in vivo diagnosis. Meas Sci Technol. 2009;20(10):104020.View ArticleGoogle Scholar
- Johansson A, Kromer K, Sroka R, Stepp H. Clinical optical diagnostics: Status and perspectives. Medical Laser Application. 2008;23(4):155–74.View ArticleGoogle Scholar
- Rotomskis R, Streckytė G, Bagdonas S, Venius J. Optinės diagnostikos technologijos (optinė biopsija) (Optical diagnostics technologies (optical biopsy)). Vilnius: Vilnius University press; 2008.Google Scholar
- Brown JQ, Vishwanath K, Palmer GM, Ramanujam N. Advances in quantitative UV-visible spectroscopy for clinical and pre-clinical application in cancer. Curr Opin Biotechnol. 2009;20(1):119–31.View ArticlePubMedPubMed CentralGoogle Scholar
- Chang VT, Cartwright PS, Bean SM, Palmer GM, Bentley RC, Ramanujam N. Quantitative physiology of the precancerous cervix in vivo through optical spectroscopy. Neoplasia. 2009;11(4):325–32.View ArticlePubMedPubMed CentralGoogle Scholar
- Rotomskis R, Streckytė G. Fluorescencinė navikų diagnostika. (Fluorescence diagnostics of tumors.). Medicina. 2004;40(12):1219–30.PubMedGoogle Scholar
- Quan L. Role of optical spectroscopy using endogenous contrasts in clinical cancer diagnosis. World J Clin Oncol. 2011;2(1):50–63.View ArticleGoogle Scholar
- 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.View ArticlePubMedGoogle Scholar
- Suhr MA, Hopper C, Jones L, George JG, Bown SG, MacRobert AJ. Optical biopsy systems for the diagnosis and monitoring of superficial cancer and precancer. Int J Oral Maxillofac Surg. 2000;29(6):453–7.View ArticlePubMedGoogle Scholar
- Jerjes WK, Upile T, Wong BJ, Betz CS, Sterenborg HJ, Witjes MJ, Veen van R, Biel MA, El-Naggar AK, Mosse CA, Olivo M, Richards-Kortum R. The future of medical diagnostics: review paper. Head Neck Oncol. 2011; 3:38.Google Scholar
- Sieroń A, Sieroń-Stołtny K, Kawczyk-Krupka A, Latos W, Kwiatek S, Straszak D, et al. The role of fluorescence diagnosis in clinical practice. Onco Targets Ther. 2013;6:977–82.PubMedPubMed CentralGoogle Scholar
- Moghissi K, Stringer MR, Dixon K. Fluorescence photodiagnosis in clinical practice. Photodiagnosis Photodyn Ther. 2008;5:235–7.View ArticlePubMedGoogle Scholar
- Ramanujam N, Mitchell MF, Mahadevan A, Thomsen S, Silva E, Richards-Kortum RR. In vivo diagnosis of cervical intraepithelial neoplasia using 337 nm laser induced fluorescence. Proc Natl Acad Sci. 1994;91:10193–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Ramanujam N, Mitchell MF, Mahadevan A, Thomsen S, Malpica A, Wright T, et al. Spectroscopic diagnosis of cervical intraepithelial neoplasia (CIN cpb using laser induced fluorescence spectra at multiple excitation wavelenghts). Lasers Surg Med. 1996;19:63–74.View ArticlePubMedGoogle Scholar
- Onizawa K, Okamura N, Saginoya H, Yoshida H. Characterization of autofluorescence in oral squamous cell carcinoma. Oral Oncol. 2003;39(2):150–6.View ArticlePubMedGoogle Scholar
- Zheng W, Soo KC, Sivanandan R, Olivo M. Detection of neoplasms in the oral cavity by digitized endoscopic imaging of 5-aminolevulinic acid-induced protoporphyrin IX fluorescence. Int J Oncol. 2002;21(4):763–8.PubMedGoogle Scholar
- Hillemanns P, Weingandt H, Baumgartner R, Diebold J, Xiang W, Stepp H. Photodetection of cervical intraepithelial neoplasia using 5-aminolevulinic acid-induced porphyrin fluorescence. Cancer. 2000;88(10):2275–82.View ArticlePubMedGoogle Scholar
- Collinet P, Jouve E, Regis C, Farine MO, Vinatier D, Cosson M, et al. Fluorescence diagnosis of cervical squamous intraepithelial lesions: A clinical feasibility study. Photodiagn Photodyn Ther. 2007;4(2):112–6.View ArticleGoogle Scholar
- 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(2):215–20.View ArticlePubMedGoogle Scholar
- Vansevičiūtė R, Žukovskaja O, Venius J. Comparison of 5-aminolevulinic acid based fluorescence diagnostics with histology in the detection of cervical neoplasia. Medical Physics in the Baltic States: proceedings of the 11th international conference on medical physics. 2013; 11:12-4.Google Scholar
- Brookner C, Follen M, Boiko I, Galvan J, Thomsen S, Malpica A, et al. Tissue slices autofluorescence patterns in fresh cervical tissue. Photochem Photobiol. 2000;71:730–6.View ArticlePubMedGoogle Scholar
- Drezek R, Brookner C, Pavlova I, Boiko I, Malpica A, Lotan R, et al. Autofluorescence microscopy of fresh cervical tissue sections reveals alterations in tissue biochemistry with dysplasia. Photochem Photobiol. 2001;73:636–41.View ArticlePubMedGoogle Scholar
- Richards-Kortum R, Drezek R, Basen-Enquist K, Cantor SB, Utzinger U, Brookner C, Follen M. Cervical dysplasia diagnosis with fluorescence spectroscopy. Handbook of biomedical fluorescence. CRC Press; 2003. p. 265-313.Google Scholar
- Florkowski CM. Sensitivity, specificity, receiver-operating characteristic (ROC) curves and likelihood ratios: communicating the performance of diagnostic tests. Clin Biochem Rev. 2008;29:83–7.Google Scholar
- Akobeng AK. Understanding diagnostic tests 3: receiver operating characteristic curves. Acta Pediatrica. 2007;644:7.Google Scholar
- 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 Phys. 2005;15(12):1664–73.Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.