PURPOSE: To determine the diagnostic efficacy of optical coherence tomography (OCT) to identify cervical intraepithelial neoplasia (CIN) grade 2 or higher by computer-aided diagnosis (CADx). METHODS: OCT has been investigated as a screening/diagnostic tool in the management of preinvasive and early invasive cancers of the uterine cervix. In this study, an automated algorithm was developed to extract OCT image features and identify CIN 2 or higher. First, the cervical epithelium was detected by a combined watershed and active contour method. Second, four features were calculated: The thickness of the epithelium and its standard deviation and the contrast between the epithelium and the stroma and its standard deviation. Finally, linear discriminant analysis was applied to classify images into two categories: Normal/inflammation/CIN 1 and CIN 2/CIN 3. The algorithm was applied to 152 images (74 patients) obtained from an international study. RESULTS: The numbers of normal/inflammatory/CIN 1/CIN 2/CIN 3 images are 74, 29, 14, 24, and 11, respectively. Tenfold cross-validation predicted the algorithm achieved a sensitivity of 51% (95% CI: 36%-67%) and a specificity of 92% (95% CI: 86%-96%) with an empirical two-category prior probability estimated from the data set. Receiver operating characteristic analysis yielded an area under the curve of 0.86. CONCLUSIONS: The diagnostic efficacy of CADx in OCT imaging to differentiate high-grade CIN from normal/low grade CIN is demonstrated. The high specificity of OCT with CADx suggests further investigation as an effective secondary screening tool when combined with a highly sensitive primary screening tool.
PURPOSE: To determine the diagnostic efficacy of optical coherence tomography (OCT) to identify cervical intraepithelial neoplasia (CIN) grade 2 or higher by computer-aided diagnosis (CADx). METHODS: OCT has been investigated as a screening/diagnostic tool in the management of preinvasive and early invasive cancers of the uterine cervix. In this study, an automated algorithm was developed to extract OCT image features and identify CIN 2 or higher. First, the cervical epithelium was detected by a combined watershed and active contour method. Second, four features were calculated: The thickness of the epithelium and its standard deviation and the contrast between the epithelium and the stroma and its standard deviation. Finally, linear discriminant analysis was applied to classify images into two categories: Normal/inflammation/CIN 1 and CIN 2/CIN 3. The algorithm was applied to 152 images (74 patients) obtained from an international study. RESULTS: The numbers of normal/inflammatory/CIN 1/CIN 2/CIN 3 images are 74, 29, 14, 24, and 11, respectively. Tenfold cross-validation predicted the algorithm achieved a sensitivity of 51% (95% CI: 36%-67%) and a specificity of 92% (95% CI: 86%-96%) with an empirical two-category prior probability estimated from the data set. Receiver operating characteristic analysis yielded an area under the curve of 0.86. CONCLUSIONS: The diagnostic efficacy of CADx in OCT imaging to differentiate high-grade CIN from normal/low grade CIN is demonstrated. The high specificity of OCT with CADx suggests further investigation as an effective secondary screening tool when combined with a highly sensitive primary screening tool.
Authors: Ilya V Turchin; Ekaterina A Sergeeva; Lev S Dolin; Vladislav A Kamensky; Natalia M Shakhova; Rebecca Richards-Kortum Journal: J Biomed Opt Date: 2005 Nov-Dec Impact factor: 3.170
Authors: Kirk W Gossage; Cynthia M Smith; Elizabeth M Kanter; Lida P Hariri; Alice L Stone; Jeffrey J Rodriguez; Stuart K Williams; Jennifer K Barton Journal: Phys Med Biol Date: 2006-03-01 Impact factor: 3.609
Authors: Xin Qi; Yinsheng Pan; Zhilin Hu; Wei Kang; Joseph E Willis; Kayode Olowe; Michael V Sivak; Andrew M Rollins Journal: J Biomed Opt Date: 2008 Sep-Oct Impact factor: 3.170
Authors: J Belinson; Y L Qiao; R Pretorius; W H Zhang; P Elson; L Li; Q J Pan; C Fischer; A Lorincz; D Zahniser Journal: Gynecol Oncol Date: 2001-11 Impact factor: 5.482
Authors: Chris J L M Meijer; Johannes Berkhof; Philip E Castle; Albertus T Hesselink; Eduardo L Franco; Guglielmo Ronco; Marc Arbyn; F Xavier Bosch; Jack Cuzick; Joakim Dillner; Daniëlle A M Heideman; Peter J F Snijders Journal: Int J Cancer Date: 2009-02-01 Impact factor: 7.396
Authors: Yu Gan; Wang Yao; Kristin M Myers; Joy Y Vink; Ronald J Wapner; Christine P Hendon Journal: Biomed Opt Express Date: 2015-03-03 Impact factor: 3.732