Accuracy of computer-aided diagnosis based on narrow-band imaging endocytoscopy for diagnosing colorectal lesions: comparison with experts

PurposeReal-time characterization of colorectal lesions during colonoscopy is important for reducing medical costs, given that the need for a pathological diagnosis can be omitted if the accuracy of the diagnostic modality is sufficiently high. However, it is sometimes difficult for community-based gastroenterologists to achieve the required level of diagnostic accuracy. In this regard, we developed a computer-aided diagnosis (CAD) system based on endocytoscopy (EC) to evaluate cellular, glandular, and vessel structure atypia in vivo. The purpose of this study was to compare the diagnostic ability and efficacy of this CAD system with the performances of human expert and trainee endoscopists.MethodsWe developed a CAD system based on EC with narrow-band imaging that allowed microvascular evaluation without dye (ECV-CAD). The CAD algorithm was programmed based on texture analysis and provided a two-class diagnosis of neoplastic or non-neoplastic, with probabilities. We validated the diagnostic ability of the ECV-CAD system using 173 randomly selected EC images (49 non-neoplasms, 124 neoplasms). The images were evaluated by the CAD and by four expert endoscopists and three trainees. The diagnostic accuracies for distinguishing between neoplasms and non-neoplasms were calculated.ResultsECV-CAD had higher overall diagnostic accuracy than trainees (87.8 vs 63.4%; $$P=0.01$$P=0.01), but similar to experts (87.8 vs 84.2%; $$P=0.76$$P=0.76). With regard to high-confidence cases, the overall accuracy of ECV-CAD was also higher than trainees (93.5 vs 71.7%; $$P<0.001$$P<0.001) and comparable to experts (93.5 vs 90.8%; $$P=0.38$$P=0.38).ConclusionsECV-CAD showed better diagnostic accuracy than trainee endoscopists and was comparable to that of experts. ECV-CAD could thus be a powerful decision-making tool for less-experienced endoscopists.

[1]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[2]  John Platt,et al.  Probabilistic Outputs for Support vector Machines and Comparisons to Regularized Likelihood Methods , 1999 .

[3]  Carl-Fredrik Westin,et al.  Tissue Classification Based on 3D Local Intensity Structures for Volume Rendering , 2000, IEEE Trans. Vis. Comput. Graph..

[4]  S. Kudo,et al.  Pit pattern in colorectal neoplasia: endoscopic magnifying view. , 2004, Endoscopy.

[5]  Chih-Jen Lin,et al.  Probability Estimates for Multi-class Classification by Pairwise Coupling , 2003, J. Mach. Learn. Res..

[6]  S. Kudo,et al.  Development of Virtual Histology and Virtual Biopsy Using Laser-scanning Confocal Microscopy , 2003, Scandinavian journal of gastroenterology. Supplement.

[7]  Charles J. Lightdale,et al.  The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon: November 30 to December 1, 2002. , 2003, Gastrointestinal endoscopy.

[8]  H Inoue.  Development of virtual biopsy using laser-scanning confocal microscopy , 2003 .

[9]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[10]  A. Polglase,et al.  Confocal laser endoscopy for diagnosing intraepithelial neoplasias and colorectal cancer in vivo. , 2004, Gastroenterology.

[11]  Haruhiro Inoue,et al.  In vivo observation of living cancer cells in the esophagus, stomach, and colon using catheter-type contact endoscope, "Endo-Cytoscopy system". , 2004, Gastrointestinal endoscopy clinics of North America.

[12]  Eiji Hidaka,et al.  Real-time in vivo virtual histology of colorectal lesions when using the endocytoscopy system. , 2006, Gastrointestinal endoscopy.

[13]  Noriko Suzuki,et al.  Optical diagnosis of small colorectal polyps at routine colonoscopy (Detect InSpect ChAracterise Resect and Discard; DISCARD trial): a prospective cohort study. , 2009, The Lancet. Oncology.

[14]  Douglas K Rex,et al.  Estimation of Impact of American College of Radiology Recommendations on CT Colonography Reporting for Resection of High-Risk Adenoma Findings , 2009, The American Journal of Gastroenterology.

[15]  Thomas Schultz,et al.  Topological Features in 2D Symmetric Higher‐Order Tensor Fields , 2011, Comput. Graph. Forum.

[16]  Endoscopic Innovations,et al.  The American Society for Gastrointestinal Endoscopy PIVI (Preservation and Incorporation of Valuable Endoscopic Innovations) on real-time endoscopic assessment of the histology of diminutive colorectal polyps , 2011 .

[17]  H Inoue,et al.  Diagnosis of colorectal lesions with a novel endocytoscopic classification - a pilot study. , 2011, Endoscopy.

[18]  C. Trautwein,et al.  Computer-based classification of small colorectal polyps by using narrow-band imaging with optical magnification. , 2011, Gastrointestinal endoscopy.

[19]  K. Chayama,et al.  Computer-aided system for predicting the histology of colorectal tumors by using narrow-band imaging magnifying colonoscopy (with video). , 2012, Gastrointestinal endoscopy.

[20]  J. GarcíaSánchez.  [Colonoscopic polypectomy and long-term prevention of colorectal cancer deaths]. , 2012 .

[21]  Kazufumi Kaneda,et al.  Computer-Aided Colorectal Tumor Classification in NBI Endoscopy Using CNN Features , 2016, ArXiv.

[22]  M. Desai,et al.  Real-time optical biopsy of colon polyps with narrow band imaging in community practice does not yet meet key thresholds for clinical decisions. , 2013, Gastroenterology.

[23]  Mariska M G Leeflang,et al.  Diagnostic performance of narrowed spectrum endoscopy, autofluorescence imaging, and confocal laser endomicroscopy for optical diagnosis of colonic polyps: a meta-analysis. , 2013, The Lancet. Oncology.

[24]  Yuichi Mori,et al.  Endocytoscopy can provide additional diagnostic ability to magnifying chromoendoscopy for colorectal neoplasms , 2014, Journal of gastroenterology and hepatology.

[25]  Yuichi Mori,et al.  Efficiency of endocytoscopy in differentiating types of serrated polyps. , 2014, Gastrointestinal endoscopy.

[26]  Helmut Neumann,et al.  Advanced colonoscopic imaging using endocytoscopy , 2015, Digestive endoscopy : official journal of the Japan Gastroenterological Endoscopy Society.

[27]  Vani Konda,et al.  ASGE Technology Committee systematic review and meta-analysis assessing the ASGE PIVI thresholds for adopting real-time endoscopic assessment of the histology of diminutive colorectal polyps. , 2015, Gastrointestinal endoscopy.

[28]  Yuichi Mori,et al.  Endocytoscopic narrow-band imaging efficiency for evaluation of inflammatory activity in ulcerative colitis. , 2015, World journal of gastroenterology.

[29]  S. Kudo,et al.  Novel computer-aided diagnostic system for colorectal lesions by using endocytoscopy (with videos). , 2015, Gastrointestinal endoscopy.

[30]  Noriya Uedo,et al.  An alternative option for “resect and discard” strategy, using magnifying narrow-band imaging: a prospective “proof-of-principle” study , 2015, Journal of Gastroenterology.

[31]  Yuichi Mori,et al.  Endocytoscopic microvasculature evaluation is a reliable new diagnostic method for colorectal lesions (with video). , 2015, Gastrointestinal endoscopy.

[32]  Yuichi Mori,et al.  Diagnosis of sessile serrated adenomas/polyps using endocytoscopy (with videos) , 2016, Digestive endoscopy : official journal of the Japan Gastroenterological Endoscopy Society.

[33]  K. Mori,et al.  Characterization of Colorectal Lesions Using a Computer-Aided Diagnostic System for Narrow-Band Imaging Endocytoscopy. , 2016, Gastroenterology.

[34]  K. Chayama,et al.  Computer-aided diagnosis of colorectal polyp histology by using a real-time image recognition system and narrow-band imaging magnifying colonoscopy. , 2016, Gastrointestinal endoscopy.

[35]  Swati G. Patel,et al.  Real-Time Characterization of Diminutive Colorectal Polyp Histology Using Narrow-Band Imaging: Implications for the Resect and Discard Strategy. , 2016, Gastroenterology.

[36]  Ahmedin Jemal,et al.  Global patterns and trends in colorectal cancer incidence and mortality , 2016, Gut.

[37]  Helen Close,et al.  Narrow band imaging optical diagnosis of small colorectal polyps in routine clinical practice: the Detect Inspect Characterise Resect and Discard 2 (DISCARD 2) study , 2016, Gut.