Yamazaki Masaki, Tomikawa Emi, Okada Miyoko, Kajikawa Satoru, Terayama Yui, Kumabe Shino, Sakairi Tetsuya, Inomata Akira, Matsumoto Izumi, Sato Gen, Suzuki Mutsumi
AI Pathology Task Force, Non-clinical Evaluation Expert Committee, Drug Evaluation Committee, Japan Pharmaceuticals Manufacturers Association (JPMA), 2-3-11 Nihombashi-honcho, Chuo-ku, Tokyo 103-0023, Japan.
Chugai Pharmaceutical Co., Ltd., Chugai Life Science Park Yokohama, 216 Totsuka-cho, Totsuka-ku, Yokohama, Kanagawa 244-8602, Japan.
J Toxicol Pathol. 2025 Jul;38(3):205-211. doi: 10.1293/tox.2024-0099. Epub 2025 Mar 11.
In recent years, the development of Artificial Intelligence (AI) technology has led to the introduction and use of AI-based histopathological evaluation (AI pathology) by various companies and organizations. The AI Pathology Task Force of the Non-clinical Evaluation Expert Committee within the Drug Evaluation Committee of the Japan Pharmaceutical Manufacturers Association (JPMA) recognizes the importance of understanding the current use and needs surrounding AI pathology in Japan. This includes its role in non-clinical research fields, such as toxicity evaluation, drug efficacy evaluation, and basic research. In addition, assessing needs and challenges related to pathology image databases is essential. Between October and November 2023, with the cooperation of the Japanese Society of Toxicologic Pathology (JSTP), we conducted a questionnaire survey on non-clinical pathology image databases to explore these issues among JPMA-affiliated and JSTP-affiliated organizations. The questionnaire survey consisted of three items: (1) implementation and utilization of whole slide images, (2) use of AI pathology in non-clinical research fields, and (3) needs and feasibility of establishing a precompetitive pathology image database (repository) and AI pathology in the non-clinical pathology field. This report summarizes the survey results and serves as a foundation for guiding future directions in the use of AI pathology in non-clinical studies in Japan.
近年来，人工智能（AI）技术的发展促使各公司和组织引入并使用基于人工智能的组织病理学评估（人工智能病理学）。日本制药商协会（JPMA）药物评估委员会非临床评估专家委员会的人工智能病理学特别工作组认识到了解日本人工智能病理学的当前应用情况和需求的重要性。这包括其在非临床研究领域的作用，如毒性评估、药物疗效评估和基础研究。此外，评估与病理图像数据库相关的需求和挑战至关重要。在2023年10月至11月期间，我们在日本毒理病理学会（JSTP）的合作下，针对非临床病理图像数据库开展了问卷调查，以在JPMA附属组织和JSTP附属组织中探讨这些问题。该问卷调查包括三个项目：（1）全切片图像的实施与利用；（2）人工智能病理学在非临床研究领域的应用；（3）在非临床病理学领域建立竞争前病理图像数据库（储存库）和人工智能病理学的需求与可行性。本报告总结了调查结果，并为指导日本非临床研究中人工智能病理学的未来发展方向奠定基础。
