Fujitsu and Tokai National Higher Education and Research System utilize AI to accelerate clinical research and tackle ‘drug loss’ in Japan

Fujitsu Limited and Tokai National Higher Education and Research System (THERS) have successfully completed field trials using generative AI to process clinical data for selecting clinical trial participants. The initiative targets the issue of ‘drug loss’ in Japan, caused by stringent local approval requirements preventing the availability of drugs used elsewhere. Patient data in clinical settings is categorized into structured (e.g., vital signs, lab results) and unstructured (e.g., doctors’ notes) data. Traditionally, structuring unstructured clinical data has been a time-consuming manual process. The trials, involving about 1,800 patient records from breast surgery procedures, achieved a 90% accuracy in structuring unstructured data. This structured data facilitated the screening of clinical trial candidates, identifying 42 potential participants, 27 of whom were deemed eligible, potentially reducing patient selection time by a third. This development allows for quicker decision-making and enhanced patient access to suitable clinical trials. Future plans involve expanding the scope of diseases, facilities, and improving the accuracy of the approach to promote its use in real clinical trials. Fujitsu and THERS will connect the trial results with Paradigm Health, Inc.’s clinical trial platform to expedite the use of real-world data, supporting collaboration with medical institutions and pharmaceutical companies to enhance clinical trial planning and execution and mitigate drug loss in Japan. THERS aims to attract international joint clinical trials to the Japanese Tokai region, improving patient care through operational efficiency and utilizing a clinical research environment. Fujitsu will expand its Healthy Living Platform on May 30, incorporating a function to promote the structuring and use of medical data through its AI service Fujitsu Kozuchi. This function will integrate with the enterprise large language model (LLM) Takane in the future to support data analysis in clinical research and patient selection efficiency. Fujitsu is committed to enhancing clinical research performance and medical care development in Japan under its Fujitsu Uvance business model focused on addressing societal issues.