SVM-MRF segmentation of colorectal NBI endoscopic images

In this paper we investigate a method for segmentation of colorectal Narrow Band Imaging (NBI) endoscopic images with Support Vector Machine (SVM) and Markov Random Field (MRF). SVM classifiers recognize each square patch of an NBI image and output posterior probabilities that represent how likely the given patch falls into a certain label. To prevent the spatial inconsistency between adjacent patches and encourage segmented regions to have smoother shapes, MRF is introduced by using the posterior outputs of SVMs as a unary term of MRF energy function. Segmentation results of 1191 NBI images are evaluated in experiments in which SVMs were trained with 480 trimmed NBI images and the MRF energy was minimized by an α - β swap Graph Cut.

[1]  Shinji Tanaka,et al.  Narrow-band imaging magnification predicts the histology and invasion depth of colorectal tumors. , 2009, Gastrointestinal endoscopy.

[2]  Foster Provost,et al.  On Applied Research in Machine Learning , 1998 .

[3]  Sebastian Gross,et al.  Chan-Vese-Segmentation of Polyps in Colonoscopic Image Data , 2011 .

[4]  Fernando Vilariño,et al.  A Region Segmentation Method for Colonoscopy Images Using a Model of Polyp Appearance , 2011, IbPRIA.

[5]  Til Aach,et al.  Active contours for localizing polyps in colonoscopic NBI image data , 2011, Medical Imaging.

[6]  Gabriele Moser,et al.  Combining Support Vector Machines and Markov Random Fields in an Integrated Framework for Contextual Image Classification , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[7]  B. S. Manjunath,et al.  A semantic representation for image retrieval , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[8]  Gabriele Moser,et al.  Contextual remote-sensing image classification by support vector machines and Markov random fields , 2010, 2010 IEEE International Geoscience and Remote Sensing Symposium.

[9]  Olga Veksler,et al.  Fast Approximate Energy Minimization via Graph Cuts , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

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

[11]  Stefano Soatto,et al.  Localizing Objects with Smart Dictionaries , 2008, ECCV.

[12]  Ron Kohavi,et al.  Guest Editors' Introduction: On Applied Research in Machine Learning , 1998, Machine Learning.

[13]  Stefano Soatto,et al.  Class segmentation and object localization with superpixel neighborhoods , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[14]  Min Zhang,et al.  A prior feature SVM-MRF based method for mouse brain segmentation , 2012, NeuroImage.

[15]  Guilherme Hoefel.  Learning a two-stage SVM/CRF sequence classifier , 2008, CIKM '08.

[16]  Til Aach,et al.  Polyp Segmentation in NBI Colonoscopy , 2009, Bildverarbeitung für die Medizin.