Stable polyp-scene classification via subsampling and residual learning from an imbalanced large dataset

Healthcare Technology Letters

article

Hayato Itoh¹  Holger Roth¹  Masahiro Oda¹  Masashi Misawa²  Yuichi Mori²  Shin-Ei Kudo²  Kensaku Mori¹ 

[1] Graduate School of Informatics, Nagoya University;Digestive Disease Center, Showa University Northern Yokohama Hospital;Information Technology Center, Nagoya University;Research Center for Medical Bigdata, National Institute of Informatics

关键词: feature extraction;    image classification;    learning (artificial intelligence);    cancer;    biological organs;    computerised tomography;    endoscopes;    medical image processing;    convolutional neural nets;    polyp-detection dataset;    stable polyp-scene classification method;    false positive detection;    high-performance CAD system;    nonpolyp scenes;    colonoscopic video dataset;    unstable polyp detection;    subsampling;    residual learning;    imbalanced large dataset;    computer-assisted diagnosis system;    three-dimensional convolutional neural network;    3D CNN;   

DOI  :  10.1049/htl.2019.0079

学科分类：肠胃与肝脏病学

来源: Wiley

【 摘 要 】

This Letter presents a stable polyp-scene classification method with low false positive (FP) detection. Precise automated polyp detection during colonoscopies is essential for preventing colon-cancer deaths. There is, therefore, a demand for a computer-assisted diagnosis (CAD) system for colonoscopies to assist colonoscopists. A high-performance CAD system with spatiotemporal feature extraction via a three-dimensional convolutional neural network (3D CNN) with a limited dataset achieved about 80% detection accuracy in actual colonoscopic videos. Consequently, further improvement of a 3D CNN with larger training data is feasible. However, the ratio between polyp and non-polyp scenes is quite imbalanced in a large colonoscopic video dataset. This imbalance leads to unstable polyp detection. To circumvent this, the authors propose an efficient and balanced learning technique for deep residual learning. The authors’ method randomly selects a subset of non-polyp scenes whose number is the same number of still images of polyp scenes at the beginning of each epoch of learning. Furthermore, they introduce post-processing for stable polyp-scene classification. This post-processing reduces the FPs that occur in the practical application of polyp-scene classification. They evaluate several residual networks with a large polyp-detection dataset consisting of 1027 colonoscopic videos. In the scene-level evaluation, their proposed method achieves stable polyp-scene classification with 0.86 sensitivity and 0.97 specificity.

【 授权许可】

CC BY|CC BY-ND|CC BY-NC|CC BY-NC-ND