Quest for the best endoscopic imaging modality for computer-assisted colonic polyp staging
|
摘要
BACKGROUND It was shown in previous studies that high definition endoscopy, high magnification endoscopy and image enhancement technologies, such as chromoendoscopy and digital chromoendoscopy [narrow-band imaging(NBI), iScan] facilitate the detection and classification of colonic polyps during endoscopic sessions. However, there are no comprehensive studies so far that analyze which endoscopic imaging modalities facilitate the automated classification of colonic polyps. In this work, we investigate the impact of endoscopic imaging modalities on the results of computer-assisted diagnosis systems for colonic polyp staging.AIM To assess which endoscopic imaging modalities are best suited for the computerassisted staging of colonic polyps.METHODS In our experiments, we apply twelve state-of-the-art feature extraction methods for the classification of colonic polyps to five endoscopic image databases of colonic lesions. For this purpose, we employ a specifically designed experimental setup to avoid biases in the outcomes caused by differing numbers of images per image database. The image databases were obtained using different imaging modalities. Two databases were obtained by high-definition endoscopy in combination with i-Scan technology(one with chromoendoscopy and one without chromoendoscopy). Three databases were obtained by highmagnification endoscopy(two databases using narrow band imaging and one using chromoendoscopy). The lesions are categorized into non-neoplastic and neoplastic according to the histological diagnosis.RESULTS Generally, it is feature-dependent which imaging modalities achieve high results and which do not. For the high-definition image databases, we achieved overall classification rates of up to 79.2% with chromoendoscopy and 88.9% without chromoendoscopy. In the case of the database obtained by high-magnification chromoendoscopy, the classification rates were up to 81.4%. For the combination of high-magnification endoscopy with NBI, results of up to 97.4% for one database and up to 84% for the other were achieved. Non-neoplastic lesions were classified more accurately in general than non-neoplastic lesions. It was shown that the image recording conditions highly affect the performance of automated diagnosis systems and partly contribute to a stronger effect on the staging results than the used imaging modality.CONCLUSION Chromoendoscopy has a negative impact on the results of the methods. NBI is better suited than chromoendoscopy. High-definition and high-magnification endoscopy are equally suited.
BACKGROUND It was shown in previous studies that high definition endoscopy, high magnification endoscopy and image enhancement technologies, such as chromoendoscopy and digital chromoendoscopy [narrow-band imaging(NBI), iScan] facilitate the detection and classification of colonic polyps during endoscopic sessions. However, there are no comprehensive studies so far that analyze which endoscopic imaging modalities facilitate the automated classification of colonic polyps. In this work, we investigate the impact of endoscopic imaging modalities on the results of computer-assisted diagnosis systems for colonic polyp staging.AIM To assess which endoscopic imaging modalities are best suited for the computerassisted staging of colonic polyps.METHODS In our experiments, we apply twelve state-of-the-art feature extraction methods for the classification of colonic polyps to five endoscopic image databases of colonic lesions. For this purpose, we employ a specifically designed experimental setup to avoid biases in the outcomes caused by differing numbers of images per image database. The image databases were obtained using different imaging modalities. Two databases were obtained by high-definition endoscopy in combination with i-Scan technology(one with chromoendoscopy and one without chromoendoscopy). Three databases were obtained by highmagnification endoscopy(two databases using narrow band imaging and one using chromoendoscopy). The lesions are categorized into non-neoplastic and neoplastic according to the histological diagnosis.RESULTS Generally, it is feature-dependent which imaging modalities achieve high results and which do not. For the high-definition image databases, we achieved overall classification rates of up to 79.2% with chromoendoscopy and 88.9% without chromoendoscopy. In the case of the database obtained by high-magnification chromoendoscopy, the classification rates were up to 81.4%. For the combination of high-magnification endoscopy with NBI, results of up to 97.4% for one database and up to 84% for the other were achieved. Non-neoplastic lesions were classified more accurately in general than non-neoplastic lesions. It was shown that the image recording conditions highly affect the performance of automated diagnosis systems and partly contribute to a stronger effect on the staging results than the used imaging modality.CONCLUSION Chromoendoscopy has a negative impact on the results of the methods. NBI is better suited than chromoendoscopy. High-definition and high-magnification endoscopy are equally suited.
BACKGROUND It was shown in previous studies that high definition endoscopy, high magnification endoscopy and image enhancement technologies, such as chromoendoscopy and digital chromoendoscopy [narrow-band imaging(NBI), iScan] facilitate the detection and classification of colonic polyps during endoscopic sessions. However, there are no comprehensive studies so far that analyze which endoscopic imaging modalities facilitate the automated classification of colonic polyps. In this work, we investigate the impact of endoscopic imaging modalities on the results of computer-assisted diagnosis systems for colonic polyp staging.AIM To assess which endoscopic imaging modalities are best suited for the computerassisted staging of colonic polyps.METHODS In our experiments, we apply twelve state-of-the-art feature extraction methods for the classification of colonic polyps to five endoscopic image databases of colonic lesions. For this purpose, we employ a specifically designed experimental setup to avoid biases in the outcomes caused by differing numbers of images per image database. The image databases were obtained using different imaging modalities. Two databases were obtained by high-definition endoscopy in combination with i-Scan technology(one with chromoendoscopy and one without chromoendoscopy). Three databases were obtained by highmagnification endoscopy(two databases using narrow band imaging and one using chromoendoscopy). The lesions are categorized into non-neoplastic and neoplastic according to the histological diagnosis.RESULTS Generally, it is feature-dependent which imaging modalities achieve high results and which do not. For the high-definition image databases, we achieved overall classification rates of up to 79.2% with chromoendoscopy and 88.9% without chromoendoscopy. In the case of the database obtained by high-magnification chromoendoscopy, the classification rates were up to 81.4%. For the combination of high-magnification endoscopy with NBI, results of up to 97.4% for one database and up to 84% for the other were achieved. Non-neoplastic lesions were classified more accurately in general than non-neoplastic lesions. It was shown that the image recording conditions highly affect the performance of automated diagnosis systems and partly contribute to a stronger effect on the staging results than the used imaging modality.CONCLUSION Chromoendoscopy has a negative impact on the results of the methods. NBI is better suited than chromoendoscopy. High-definition and high-magnification endoscopy are equally suited.
引文
1 Berr F,Oyama T,Ponchon T,Yahagi N.Early Neoplasias of the Gastrointestinal Tract:Endoscopic Diagnosis and Therapeutic Decisions.Boston:Springer 2014;[DOI:10.1007/978-1-4614-8292-5]
2 Rex DK,Kahi C,O'Brien M,Levin TR,Pohl H,Rastogi A,Burgart L,Imperiale T,Ladabaum U,Cohen J,Lieberman DA.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.Gastrointest Endosc 2011;73:419-422[PMID:21353837 DOI:10.1016/j.gie.2011.01.023]
3 Patel SG,Schoenfeld P,Kim HM,Ward EK,Bansal A,Kim Y,Hosford L,Myers A,Foster S,Craft J,Shopinski S,Wilson RH,Ahnen DJ,Rastogi A,Wani S.Real-Time Characterization of Diminutive Colorectal Polyp Histology Using Narrow-Band Imaging:Implications for the Resect and Discard Strategy.Gastroenterology 2016;150:406-418[PMID:26522260 DOI:10.1053/j.gastro.2015.10.042]
4 ASGE Technology Committee;Manfredi MA,Abu Dayyeh BK,Bhat YM,Chauhan SS,Gottlieb KT,Hwang JH,Komanduri S,Konda V,Lo SK,Maple JT,Murad FM,Siddiqui UD,Wallace MB,Banerjee S.Electronic chromoendoscopy.Gastrointest Endosc 2015;81:249-261[PMID:25484330 DOI:10.1016/j.gie.2014.06.020]
5 Hoffman A,Kagel C,Goetz M,Tresch A,Mudter J,Biesterfeld S,Galle PR,Neurath MF,Kiesslich R.Recognition and characterization of small colonic neoplasia with high-definition colonoscopy using i-Scan is as precise as chromoendoscopy.Dig Liver Dis 2010;42:45-50[PMID:19473893 DOI:10.1016/j.dld.2009.04.005]
6 Bouwens MW,de Ridder R,Masclee AA,Driessen A,Riedl RG,Winkens B,Sanduleanu S.Optical diagnosis of colorectal polyps using high-definition i-scan:an educational experience.World JGastroenterol 2013;19:4334-4343[PMID:23885144 DOI:10.3748/wjg.v19.i27.4334]
7 Tischendorf JJ,Wasmuth HE,Koch A,Hecker H,Trautwein C,Winograd R.Value of magnifying chromoendoscopy and narrow band imaging(NBI)in classifying colorectal polyps:a prospective controlled study.Endoscopy 2007;39:1092-1096[PMID:18072061 DOI:10.1055/s-2007-966781]
8 Chang CC,Hsieh CR,Lou HY,Fang CL,Tiong C,Wang JJ,Wei IV,Wu SC,Chen JN,Wang YH.Comparative study of conventional colonoscopy,magnifying chromoendoscopy,and magnifying narrowband imaging systems in the differential diagnosis of small colonic polyps between trainee and experienced endoscopist.Int J Colorectal Dis 2009;24:1413-1419[PMID:19603174 DOI:10.1007/s00384-009-0760-9]
9 ASGE Technology Committee.High-definition and high-magnification endoscopes.Gastrointest Endosc2014;80:919-927[PMID:25442091 DOI:10.1016/j.gie.2014.06.019]
10 Rameshshanker R,Wilson A.Electronic Imaging in Colonoscopy:Clinical Applications and Future Prospects.Curr Treat Options Gastroenterol 2016;14:140-151[PMID:26923476 DOI:10.1007/s11938-016-0075-1]
11 H?fner M,Uhl A,Wimmer G,Medical Computer Vision.A novel shape feature descriptor for the classification of polyps in HD colonoscopy.Medical Computer Vision.Large Data in Medical Imaging.Cham:Springer 2014;205-213[DOI:10.1007/978-3-319-05530-5_20]
12 H?fner M,Tamaki T,Tanaka S,Uhl A,Wimmer G,Yoshida S.Local fractal dimension based approaches for colonic polyp classification.Med Image Anal 2015;26:92-107[PMID:26385078 DOI:10.1016/j.media.2015.08.007]
13 Ribeiro E,Uhl A,Wimmer G,H?fner M.Exploring Deep Learning and Transfer Learning for Colonic Polyp Classification.Comput Math Methods Med 2016;2016:6584725[PMID:27847543 DOI:10.1155/2016/6584725]
14 H?fner M,Liedlgruber M,Uhl A,Vécsei A,Wrba F.Delaunay triangulation-based pit density estimation for the classification of polyps in high-magnification chromo-colonoscopy.Comput Methods Programs Biomed 2012;107:565-581[PMID:22325257 DOI:10.1016/j.cmpb.2011.12.012]
15 H?fner M,Liedlgruber M,Uhl A,Vécsei A,Wrba F.Color treatment in endoscopic image classification using multi-scale local color vector patterns.Med Image Anal 2012;16:75-86[PMID:21624846 DOI:10.1016/j.media.2011.05.006]
16 H?fner M,Kwitt R,Uhl A,Gangl A,Wrba F,Ve′csei A.Computer-assisted pit-pattern classification in different wavelet domains for supporting dignity assessment of colonic polyps.Pattern Recognition 2008;42:1180-1191[DOI:10.1016/j.patcog.2008.07.012]
17 H?fner M,Liedlgruber M,Uhl A.Colonic polyp classification in high-definition video using complex wavelet-packets.Bildverarbeitung für die Medizin 2015;365-370[DOI:10.1007/978-3-662-46224-9_63]
18 Wimmer G,Tamaki T,Tischendorf JJ,H?fner M,Yoshida S,Tanaka S,Uhl A.Directional wavelet based features for colonic polyp classification.Med Image Anal 2016;31:16-36[PMID:26948110 DOI:10.1016/j.media.2016.02.001]
19 Tamaki T,Yoshimuta J,Kawakami M,Raytchev B,Kaneda K,Yoshida S,Takemura Y,Onji K,Miyaki R,Tanaka S.Computer-aided colorectal tumor classification in NBI endoscopy using local features.Med Image Anal 2013;17:78-100[PMID:23085199 DOI:10.1016/j.media.2012.08.003]
20 Kominami Y,Yoshida S,Tanaka S,Sanomura Y,Hirakawa T,Raytchev B,Tamaki T,Koide T,Kaneda K,Chayama K.Computer-aided diagnosis of colorectal polyp histology by using a real-time image recognition system and narrow-band imaging magnifying colonoscopy.Gastrointest Endosc 2016;83:643-649[PMID:26264431 DOI:10.1016/j.gie.2015.08.004]
21 Gross S,Palm S,Tischendorf JJW,Behrens A,Trautwein C,Aach T.Automated classification of colon polyps in endoscopic image data.SPIE Proceedings Medical Imaging 2012:Computer-Aided Diagnosis,83150W.[DOI:10.1117/12.911177]
22 Kodashima S,Fujishiro M.Novel image-enhanced endoscopy with i-scan technology.World JGastroenterol 2010;16:1043-1049[PMID:20205272 DOI:10.3748/wjg.v16.i9.1043]
23 Wimmer G,Gadermayr M,Kwitt R,H?fner M,Merhof D,Uhl A.Evaluation of i-scan virtual chromoendoscopy and traditional chromoendoscopy for the automated diagnosis of colonic polyp.Proceedings of the 3rd International Workshop on Computer-Assisted and Robotic Endoscopy(CARE).Cham:Springer 2016;59-71[DOI:10.1007/978-3-319-54057-3_6]
24 Wimmer G,Ve′csei A,Uhl A.Cnn transfer learning for the automated diagnosis of celiac disease.Proceedings of the Sixth International Conference on Image Processing Theory,Tools and Applications(IPTA)2016;1-6[DOI:10.1109/IPTA.2016.7821020]
25 H?fner M,Uhl A,Wimmer G.Shape and size adapted local fractal dimension for the classification of polyps in HD colonoscopy 2014;Oct 27-30;Paris,France.IEEE,2014[DOI:10.1109/ICIP.2014.7025466]
26 Liao S,Zhu X,Lei Z,Zhang L,Li SZ.Learning multi-scale block local binary patterns for face recognition.In:Proceedings of the 2007;Springer,2007:828-837[DOI:10.1007/978-3-540-74549-5_87]
27 Fan RE,Chang KW,Hsieh CJ,Wang XR,Lin CJ.LIBLINEAR:A library for large linear classification.JMach Learn Res 2008;9:1871-1874
2 Rex DK,Kahi C,O'Brien M,Levin TR,Pohl H,Rastogi A,Burgart L,Imperiale T,Ladabaum U,Cohen J,Lieberman DA.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.Gastrointest Endosc 2011;73:419-422[PMID:21353837 DOI:10.1016/j.gie.2011.01.023]
3 Patel SG,Schoenfeld P,Kim HM,Ward EK,Bansal A,Kim Y,Hosford L,Myers A,Foster S,Craft J,Shopinski S,Wilson RH,Ahnen DJ,Rastogi A,Wani S.Real-Time Characterization of Diminutive Colorectal Polyp Histology Using Narrow-Band Imaging:Implications for the Resect and Discard Strategy.Gastroenterology 2016;150:406-418[PMID:26522260 DOI:10.1053/j.gastro.2015.10.042]
4 ASGE Technology Committee;Manfredi MA,Abu Dayyeh BK,Bhat YM,Chauhan SS,Gottlieb KT,Hwang JH,Komanduri S,Konda V,Lo SK,Maple JT,Murad FM,Siddiqui UD,Wallace MB,Banerjee S.Electronic chromoendoscopy.Gastrointest Endosc 2015;81:249-261[PMID:25484330 DOI:10.1016/j.gie.2014.06.020]
5 Hoffman A,Kagel C,Goetz M,Tresch A,Mudter J,Biesterfeld S,Galle PR,Neurath MF,Kiesslich R.Recognition and characterization of small colonic neoplasia with high-definition colonoscopy using i-Scan is as precise as chromoendoscopy.Dig Liver Dis 2010;42:45-50[PMID:19473893 DOI:10.1016/j.dld.2009.04.005]
6 Bouwens MW,de Ridder R,Masclee AA,Driessen A,Riedl RG,Winkens B,Sanduleanu S.Optical diagnosis of colorectal polyps using high-definition i-scan:an educational experience.World JGastroenterol 2013;19:4334-4343[PMID:23885144 DOI:10.3748/wjg.v19.i27.4334]
7 Tischendorf JJ,Wasmuth HE,Koch A,Hecker H,Trautwein C,Winograd R.Value of magnifying chromoendoscopy and narrow band imaging(NBI)in classifying colorectal polyps:a prospective controlled study.Endoscopy 2007;39:1092-1096[PMID:18072061 DOI:10.1055/s-2007-966781]
8 Chang CC,Hsieh CR,Lou HY,Fang CL,Tiong C,Wang JJ,Wei IV,Wu SC,Chen JN,Wang YH.Comparative study of conventional colonoscopy,magnifying chromoendoscopy,and magnifying narrowband imaging systems in the differential diagnosis of small colonic polyps between trainee and experienced endoscopist.Int J Colorectal Dis 2009;24:1413-1419[PMID:19603174 DOI:10.1007/s00384-009-0760-9]
9 ASGE Technology Committee.High-definition and high-magnification endoscopes.Gastrointest Endosc2014;80:919-927[PMID:25442091 DOI:10.1016/j.gie.2014.06.019]
10 Rameshshanker R,Wilson A.Electronic Imaging in Colonoscopy:Clinical Applications and Future Prospects.Curr Treat Options Gastroenterol 2016;14:140-151[PMID:26923476 DOI:10.1007/s11938-016-0075-1]
11 H?fner M,Uhl A,Wimmer G,Medical Computer Vision.A novel shape feature descriptor for the classification of polyps in HD colonoscopy.Medical Computer Vision.Large Data in Medical Imaging.Cham:Springer 2014;205-213[DOI:10.1007/978-3-319-05530-5_20]
12 H?fner M,Tamaki T,Tanaka S,Uhl A,Wimmer G,Yoshida S.Local fractal dimension based approaches for colonic polyp classification.Med Image Anal 2015;26:92-107[PMID:26385078 DOI:10.1016/j.media.2015.08.007]
13 Ribeiro E,Uhl A,Wimmer G,H?fner M.Exploring Deep Learning and Transfer Learning for Colonic Polyp Classification.Comput Math Methods Med 2016;2016:6584725[PMID:27847543 DOI:10.1155/2016/6584725]
14 H?fner M,Liedlgruber M,Uhl A,Vécsei A,Wrba F.Delaunay triangulation-based pit density estimation for the classification of polyps in high-magnification chromo-colonoscopy.Comput Methods Programs Biomed 2012;107:565-581[PMID:22325257 DOI:10.1016/j.cmpb.2011.12.012]
15 H?fner M,Liedlgruber M,Uhl A,Vécsei A,Wrba F.Color treatment in endoscopic image classification using multi-scale local color vector patterns.Med Image Anal 2012;16:75-86[PMID:21624846 DOI:10.1016/j.media.2011.05.006]
16 H?fner M,Kwitt R,Uhl A,Gangl A,Wrba F,Ve′csei A.Computer-assisted pit-pattern classification in different wavelet domains for supporting dignity assessment of colonic polyps.Pattern Recognition 2008;42:1180-1191[DOI:10.1016/j.patcog.2008.07.012]
17 H?fner M,Liedlgruber M,Uhl A.Colonic polyp classification in high-definition video using complex wavelet-packets.Bildverarbeitung für die Medizin 2015;365-370[DOI:10.1007/978-3-662-46224-9_63]
18 Wimmer G,Tamaki T,Tischendorf JJ,H?fner M,Yoshida S,Tanaka S,Uhl A.Directional wavelet based features for colonic polyp classification.Med Image Anal 2016;31:16-36[PMID:26948110 DOI:10.1016/j.media.2016.02.001]
19 Tamaki T,Yoshimuta J,Kawakami M,Raytchev B,Kaneda K,Yoshida S,Takemura Y,Onji K,Miyaki R,Tanaka S.Computer-aided colorectal tumor classification in NBI endoscopy using local features.Med Image Anal 2013;17:78-100[PMID:23085199 DOI:10.1016/j.media.2012.08.003]
20 Kominami Y,Yoshida S,Tanaka S,Sanomura Y,Hirakawa T,Raytchev B,Tamaki T,Koide T,Kaneda K,Chayama K.Computer-aided diagnosis of colorectal polyp histology by using a real-time image recognition system and narrow-band imaging magnifying colonoscopy.Gastrointest Endosc 2016;83:643-649[PMID:26264431 DOI:10.1016/j.gie.2015.08.004]
21 Gross S,Palm S,Tischendorf JJW,Behrens A,Trautwein C,Aach T.Automated classification of colon polyps in endoscopic image data.SPIE Proceedings Medical Imaging 2012:Computer-Aided Diagnosis,83150W.[DOI:10.1117/12.911177]
22 Kodashima S,Fujishiro M.Novel image-enhanced endoscopy with i-scan technology.World JGastroenterol 2010;16:1043-1049[PMID:20205272 DOI:10.3748/wjg.v16.i9.1043]
23 Wimmer G,Gadermayr M,Kwitt R,H?fner M,Merhof D,Uhl A.Evaluation of i-scan virtual chromoendoscopy and traditional chromoendoscopy for the automated diagnosis of colonic polyp.Proceedings of the 3rd International Workshop on Computer-Assisted and Robotic Endoscopy(CARE).Cham:Springer 2016;59-71[DOI:10.1007/978-3-319-54057-3_6]
24 Wimmer G,Ve′csei A,Uhl A.Cnn transfer learning for the automated diagnosis of celiac disease.Proceedings of the Sixth International Conference on Image Processing Theory,Tools and Applications(IPTA)2016;1-6[DOI:10.1109/IPTA.2016.7821020]
25 H?fner M,Uhl A,Wimmer G.Shape and size adapted local fractal dimension for the classification of polyps in HD colonoscopy 2014;Oct 27-30;Paris,France.IEEE,2014[DOI:10.1109/ICIP.2014.7025466]
26 Liao S,Zhu X,Lei Z,Zhang L,Li SZ.Learning multi-scale block local binary patterns for face recognition.In:Proceedings of the 2007;Springer,2007:828-837[DOI:10.1007/978-3-540-74549-5_87]
27 Fan RE,Chang KW,Hsieh CJ,Wang XR,Lin CJ.LIBLINEAR:A library for large linear classification.JMach Learn Res 2008;9:1871-1874