Journal Article

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

Itoh, Hayato, Oda, Masahiro, Misawa, Masashi, Mori, Yuichi, Roth, Holger, Kudo, Shin-Ei, Mori, Kensaku

2019

Overview

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.

• Publisher: The Institution of Engineering and Technology,John Wiley & Sons, Inc,Wiley
• Subject: 3D CNN / Architecture / biological organs / cancer / Classification / colonoscopic video dataset / Colonoscopy / computer-assisted diagnosis system / computerised tomography / convolutional neural nets / Datasets / Deep learning / endoscopes / false positive detection / feature extraction / high-performance cad system / image classification / imbalanced large dataset / learning (artificial intelligence) / Machine learning / medical image processing / nonpolyp scenes / polyp-detection dataset / Polyps / residual learning / Special Issue: Papers from the 13th Workshop on Augmented Environments for Computer Assisted Interventions / stable polyp-scene classification method / subsampling / three-dimensional convolutional neural network / Tumors / unstable polyp detection