Jump to Main Contents

Home > Division of Medical AI Research and Development

Division of Medical AI Research and Development



In our division, we try to elucidate the molecular mechanism of how abnormal modification causes human tumorigenesis, in particular focusing on protein methylation, and apply our findings for development of novel anti-cancer drugs. In addition, we introduced artificial intelligence (AI) technologies into a medical study during the early stages of the development of medical AI in Japan and developed an integrated cancer medical system using AI technology with the intention of applying it to Precision Medicine. We now play a central role in the medical AI research field in Japan.

Past and Current Research Activities

1. Development of a next-generation ChIP-seq method with the intention of applying it to Precision Medicine.

In order to establish the platform to conduct multi-omics analysis using machine learning and deep learning, we tried to develop a next-generation ChIP-seq method. Consequently, we succeeded in developing a system that enables the ChIP-seq analysis of formalin-fixed, paraffin-embedded (FFPE) samples effectively. On the basis of this system, we successfully identified not only the status of histone modifications (H3K27 acetylation, H3K4 tri-methylation) but also the identification of the binding sites of transcription factor CCCTC-binding Factor (CTCF). This is the first case to succeed in identifying the transcription factor binding sites by ChIP-seq analysis using the FFPE sample. In this system, we introduced the human general-purpose robot LabDroid, which enabled acquisition of highly robust data (Figure 1).


2. Development of highly precise image diagnosis devices and image diagnosis methods by AI technology.

As for the endoscope image analysis, we developed a system that immediately detects colorectal cancer and ulcerative colon polyps, a precursor to cancer, during an endoscopic examination using artificial intelligence (AI). It automatically detects colorectal cancer and polyps from images and videos taken during an endoscopic examination of the colon, and aids in discovery of lesions by endoscopists. It improves polyp detection, which was an issue during such exams, and increases the detection rate. In this manner, it greatly contributes to the prevention and early detection of colorectal cancer. In addition, we built an annotation platform designed from the viewpoint of the clinician, and designed an integrated cancer medical care database. The results have already been installed in our partner company to achieve their social implementation

3. Construction of multi-omics analysis systems using artificial intelligence technologies

In this study, we started developing an integrated lung cancer database including whole exome sequencing data, transcriptome analysis data, histone modification data, DNA methylation data and radiation image data, and developed an analysis algorithm.


Figure1: Automation system of next-generation ChIP-seq with the bio experiment automation robot