Tanaka Toshihisa

Establishing AI-based Diagnosis Support for Epilepsy With Electroencephalographic Big Data of Multiple Facilities

Research Director

Tanaka Toshihisa

Collaborator

Hidenori Sugano

Associate Professor Department of Neurosurgery Juntendo University

Vision

We establish a platform fusing a multi-facility database of electrophysiological signals, the knowledge of experts, and signal processing and machine learning techniques to achieve the most advanced neurophysiological investigation so that people all over the world can equally have the best diagnosis of “brain disease.”

Outline

The current challenge is the unavailability of sufficient number of specialists that can analyze the needed at present complex neurophysiological (electrophysiological) investigations and provide adequate management to a broad range of conditions, such as epilepsy, in a large patient population. For that purpose, we attempted to create a model integrating specialists’ knowledge and decision-making in the form of AI modules in a platform that can be shared.

Hiroki Matsutani

Development and Social Implementation of On-Device Learning Technology

Research Director

Hiroki Matsutani

Collaborator

Tamao Okamoto	Manager Product Analysis Center Panasonic Corporation
Masaaki Kondo	Professor Faculty of Science and Technology Keio University
Yasuhiko Shioda	Senior Consultant Solutions Division 1 Fixstars Corporation
Takahiro Nishiyama	Manager Circuit Technology Development Div. ROHM CO.,Ltd.
Takemasa Miyoshi	Team Leader Center for Computational Science RIKEN

Vision

Based on an on-device learning algorithm, associated technologies, and their chip integration, we will extend an application range of edge AI to sensor devices, provide a self-adaptive intelligence which is resistant to environmental changes for them, and promote safe and reliable industries and their automation toward a highly optimized social system.

Outline

We will provide an on-device learning algorithm, federation learning algorithm, associated technologies, and their chip integration, so that a lot of IoT devices can acquire a self-adaptive intelligence which is resistant to environmental changes, and support a maintenance-free operation of low-end edge AI which is intrinsically diverse and embedded. The concept of the proposed on-device learning technology will be demonstrated in various application fields including smart industry, facility monitoring, meteorological sensors, consumer devices, and sensor tags.

Ken’ichi Morooka

Computer-Aided Cancer Screening System by 3D AI for Identifying Cells.

Research Director

Ken’ichi Morooka

Collaborator

Eiji Ohno	Visiting researcher Research Center for Life and Health Sciences Kyoto Tachibana University
Hajime Nagahara	Professor Institute for Datability Science Osaka University
Hideki Hashimoto	General Manager R&D Planning Division Proassist, Ltd

Vision

We automate cytological cancer screening, which has depended on visual observation until now, by 3D shape measurement of cells, cell database development, and machine　learning techniques for the diagnosis. Through this automation, we provide a high-speed and high-precise cancer screening system, by which everyone in the world can receive a high-quality cancer diagnosis anywhere.

Outline

Cancer screening using human tissue samples is useful for early detection of cancers with its less invasive natures. Our project is to develop an automatic AI-based system of cancer screening using 3D information about cells obtained from multi-focus digital images with gigapixels. Our system contributes to the drastic improvement of the accuracy and efficiency of the cancer screening compared with existing commercial screening system using 2D cell images.

Link
Project website