Healthcare Big Data Analysis Using Concept Encoder

「As of the end of December 2025, there were 23 DDAIF-related patents worldwide.」

「In this experiment, using DDAIF, we successfully extracted 17 candidate target genes from approximately 20,000 human genes in just 2 days. Conventional approaches often require more than 2 years to search for such target molecules.」

「To meet these customer needs, in November 2023 we entered into a partnership agreement with Axcelead Drug Discovery Partners, which has platform functions required for drug discovery.」

「In FRONTEO's DD-AIF business, combining AI specialized in hypothesis generation with the knowledge of drug discovery researchers has enabled high-throughput generation of high-quality hypotheses, making it possible to produce evidence-based hypotheses rapidly, in large numbers and with high precision.」

「During the first-quarter cumulative consolidated period, our drug discovery researchers presented research at the annual scientific meeting of the Japanese Society of Toxicology using FRONTEO's proprietary Drug Discovery Best Known Method, or DD-BKM.」

「The functionality of DD-AIF is also progressing steadily, including the expansion of diseases subject to evaluation of target genes and molecules analyzed by KIBIT liGALILEO from 1,500 to 12,000.」

「We began collaborating with Springer Nature, a global academic publisher, in December 2022 to launch a new service for analyzing academic papers using Amanogawa. In the medical digital transformation field, adoption of Mekiki, an AI system for automated classification and review, began at major pharmaceutical companies. We also advanced expansion in the Medical Intelligence field by adding Medical Insight, a new feature to WordATLAS, an AI system supporting information strategy, that visualizes topics of interest to physicians and prescribing and treatment trends.」

「In the life sciences AI field, within the Medical Intelligence domain, we worked to expand the drug discovery business through 3 applications: Amanogawa, which streamlines academic paper searches; Cascade Eye, which visualizes relationships among molecules and genes; and liGALILEO, which scores complex evaluation metrics, including the scientific validity of target molecules.」

「The project was completed during the first-quarter cumulative consolidated period, earlier than initially expected, owing to improved functionality in the drug discovery application, and we began considering the next stage. In the drug discovery business, we sequentially upgraded all 3 applications: Amanogawa, which streamlines academic paper searches; Cascade Eye, which visualizes relationships among molecules and genes; and liGALILEO, which scores complex evaluation metrics, including the scientific validity of target molecules.」

「During the quarter, we began working with Axcelead Inc. on the joint venture Druggable Target 1000, which uses AI to analyze the potential of target molecules in drug development. We position as our AI drug mining business the discovery of drug discovery seeds in an entirely new way from among several million compounds in Japan.」

「In the Medical Intelligence field, we announced on 22 June 2021 that Cascade Eye, which supports the discovery of candidate compounds for drug discovery, had received a decision to grant a patent in Japan. We will further promote market penetration of this drug discovery support solution, whose technological capabilities and uniqueness have been officially recognized.」

「In the Medical Intelligence field, adoption of the Amanogawa AI system for searching academic papers expanded beyond pharmaceutical companies to academia. By March 2021, it had been introduced at several universities, including the University of Tokyo, Tokushima University and Tokyo Women's Medical University. On 29 March 2021, we also announced a business partnership with MOLCURE to promote digital transformation in drug discovery.」

「In the drug discovery field, we launched a joint venture with Takeda Pharmaceutical Company's T-CiRA Discovery to collaborate on the discovery and development of new medicines for rare diseases using our proprietary AI drug discovery technology.」

「In the drug discovery field, we entered into a license agreement with Chugai Pharmaceutical Co., Ltd. in May 2020 for a drug discovery support AI system. The system is currently being used in the drug discovery processes of several major pharmaceutical companies.」

「In March 2020, we entered into a license agreement with Takeda Pharmaceutical Company Limited for a drug discovery support AI system, marking a major step toward business expansion.」

「In particular, in the life sciences AI field, we announced a joint research project with Takeda Pharmaceutical Company Limited and Iwate Medical University to study the diagnosis and treatment of Parkinson's disease using AI and genomics. Analysis using the Concept Encoder artificial intelligence (AI) engine revealed important fundamental information related to the diagnosis of Parkinson's disease, and we will continue advancing the research toward personalized medicine tailored to each patient.」

「The system includes a database of 14 million papers on diseases and genes and 1.7 million publicly available datasets, which are essential for drug discovery research. CE has been pre-trained on this data, enabling immediate searches and analysis.」

「By applying CE's vectorization technology, it becomes possible to train CE on literature information from papers and patents related to disease-associated genes, drug discovery target molecules and candidate compounds, together with numerical data such as gene expression analyses obtained from public databases and our own research, and to perform an integrated analysis of their relevance to working hypotheses in drug discovery research.」

「CE was developed to effectively analyze and use healthcare-related big data containing large volumes of free-form text data based on evidence. Through joint analysis with the numerical data accumulated across the healthcare field, including gene expression information, vital signs and various test results, we will continue contributing to the development of the healthcare industry as a whole.」

「During the first quarter, we obtained a patent for our Group's second proprietary artificial intelligence (AI) engine, Concept Encoder.」

「In addition to language, this enables analysis incorporating numerical data such as gene expression information, vital signs and various test results, promoting the use of big data in the healthcare sector.」