HayatoItoh

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

HayatoItoh

Invited Talk by Dr. Hayato Itoh

Monday 10 September 2018, IFL, 10am

Towards Automated Colonoscopy Diagnosis: Binary Polyp-Size Classification via Unsupervised Depth Learning

Dr. Hayato Itoh, Mori Lab, Nagoya, Japan

Abstract: In colon cancer screening, polyp size estimation using only colonoscopy images or videos is difficult even for expert physicians although the size information of polyps is important for diagnosis. To construct a fully automated computer-aided diagnosis (CAD) pipeline, a robust and precise polyp size estimation method is highly desired. However, the size estimation problem of a three-dimensional object from a single two-dimensional image is ill-posed due to the lack of three-dimensional spatial information. To circumvent this challenge, we formulate a relaxed form of size estimation as a binary-size classification problem and solve it by a new deep neural network architecture. This relaxed form of size estimation is defined as a two-category classification: under and over a certain polyp dimension criterion that would provoke different clinical treatments (resecting the polyp or not). Our proposed deep learning architecture estimates the depth map from an input colonoscopic RGB image using unsupervised deep learning, and integrates RGB with the computed depth information to produce a four-channel RGB-D imagery data, that is subsequently encoded as a pipeline to extract deep RGB-D image features and facilitate the size classification into two categories: under and over 10mm polyps. We collect a large dataset of colonoscopic videos of totally over 16 hours is constructed for the evaluation of our proposed method. Using this dataset, we evaluate the accuracies of both polyp detection and binary polyp-size estimation since detection is a prerequisite step of a fully automated CAD system. The experimental results show that our proposed method achieves 79.2% accuracy for binary polyp-size classification. We also combine the extraction features by our deep learning architecture and classification of short video clips using a long short-term memory (LSTM) network. Polyp detection (if the video clip contains a polyp or not) shows 88.8% sensitivity when employing the spatio-temporal image feature extraction and classification.

Biography: Hayato Itoh is a currently a postdoctoral researcher at the Graduate School of Informatics, Nagoya University, Japan. He received M.E. and PhD?. in Engineering from Chiba University in 2012 and 2017, respectively. During his PhD? course, he had great interests about theoretical aspect of image pattern recognition and computer vision, such that tensorial structure of images and topological property of image manifolds. After the graduation of PhD? course, he moved to Nagoya University and started his research in medical image processing. His research interests range from signal processing to machine learning for computer vision, pattern recognition, and applications of them to medical image processing. He serves one of the AMED (Japan Agency for Medical Research and Development) project for developing an automated diagnosis system for colonoscopy: detection, size estimation and pathological diagnosis of colorectal polyps.

WebEventForm
Title: Invited Talk by Dr. Hayato Itoh
Date: 10 September 2018
Location: IFL
Abstract: In colon cancer screening, polyp size estimation using only colonoscopy images or videos is difficult even for expert physicians although the size information of polyps is important for diagnosis. To construct a fully automated computer-aided diagnosis (CAD) pipeline, a robust and precise polyp size estimation method is highly desired. However, the size estimation problem of a three-dimensional object from a single two-dimensional image is ill-posed due to the lack of three-dimensional spatial information. To circumvent this challenge, we formulate a relaxed form of size estimation as a binary-size classification problem and solve it by a new deep neural network architecture. This relaxed form of size estimation is defined as a two-category classification: under and over a certain polyp dimension criterion that would provoke different clinical treatments (resecting the polyp or not). Our proposed deep learning architecture estimates the depth map from an input colonoscopic RGB image using unsupervised deep learning, and integrates RGB with the computed depth information to produce a four-channel RGB-D imagery data, that is subsequently encoded as a pipeline to extract deep RGB-D image features and facilitate the size classification into two categories: under and over 10mm polyps. We collect a large dataset of colonoscopic videos of totally over 16 hours is constructed for the evaluation of our proposed method. Using this dataset, we evaluate the accuracies of both polyp detection and binary polyp-size estimation since detection is a prerequisite step of a fully automated CAD system. The experimental results show that our proposed method achieves 79.2% accuracy for binary polyp-size classification. We also combine the extraction features by our deep learning architecture and classification of short video clips using a long short-term memory (LSTM) network. Polyp detection (if the video clip contains a polyp or not) shows 88.8% sensitivity when employing the spatio-temporal image feature extraction and classification.
Imageurl: index_new.png
Type: News
Videourl:  
Conferencelink:  


Edit | Attach | Refresh | Diffs | More | Revision r1.1 - 03 Sep 2018 - 13:59 - GuillaumeZahnd

Lehrstuhl für Computer Aided Medical Procedures & Augmented Reality    rss.gif