Robust endocytoscopic image classification based on higher-order symmetric tensor analysis and multi-scale topological s
- PDF / 1,855,350 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 74 Downloads / 160 Views
ORIGINAL ARTICLE
Robust endocytoscopic image classification based on higher-order symmetric tensor analysis and multi-scale topological statistics Hayato Itoh1 · Yukitaka Nimura2 · Yuichi Mori3 · Masashi Misawa3 · Shin-Ei Kudo3 · Kinichi Hotta4 · Kazuo Ohtsuka5 · Shoichi Saito6 · Yutaka Saito7 · Hiroaki Ikematsu8 · Yuichiro Hayashi1 · Masahiro Oda1 · Kensaku Mori1 Received: 22 January 2020 / Accepted: 2 September 2020 © CARS 2020
Abstract Purpose An endocytoscope is a new type of endoscope that enables users to perform conventional endoscopic observation and ultramagnified observation at the cell level. Although endocytoscopy is expected to improve the cost-effectiveness of colonoscopy, endocytoscopic image diagnosis requires much knowledge and high-level experience for physicians. To circumvent this difficulty, we developed a robust endocytoscopic (EC) image classification method for the construction of a computer-aided diagnosis (CAD) system, since real-time CAD can resolve accuracy issues and reduce interobserver variability. Method We propose a novel feature extraction method by introducing higher-order symmetric tensor analysis to the computation of multi-scale topological statistics on an image, and we integrate this feature extraction with EC image classification. We experimentally evaluate the classification accuracy of our proposed method by comparing it with three deep learning methods. We conducted this comparison by using our large-scale multi-hospital dataset of about 55,000 images of over 3800 patients. Results Our proposed method achieved an average 90% classification accuracy for all the images in four hospitals even though the best deep learning method achieved 95% classification accuracy for images in only one hospital. In the case with a rejection option, the proposed method achieved expert-level accurate classification. These results demonstrate the robustness of our proposed method against pit pattern variations, including differences of colours, contrasts, shapes, and hospitals. Conclusions We developed a robust EC image classification method with novel feature extraction. This method is useful for the construction of a practical CAD system, since it has sufficient generalisation ability. Keywords Endocytoscopy · CAD · Pathological pattern classification · Machine learning · Texture analysis
Introduction
B
Hayato Itoh [email protected]
1
Graduate School of Informatics, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan
2
Information Strategy Office, Information and Communications, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan
3
4
Digestive Disease Center, Showa University Northern Yokohama Hospital, Chigasaki-chuo 35-1, Tsuduki-ku, Yokohama 224-8503, Japan Division of Endoscopy, Shizuoka Cancer Center, Shimonagakubo 1007, Nagaizumi-cho, Sunto-gun, Shizuoka 411-8777, Japan
The early detection of colorectal cancer is a critical problem because the survival rate for the cancer particularly depends 5
Department of Gastroenterology and Hepat
Data Loading...
