Characteristics of potentially serious incidental findings (PSIF) in radiological imagings. Epidemiological analysis.

2023 Fiscal Year Final Research Report

• Project/Area Number: 20K18887
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 58010:Medical management and medical sociology-related
• Research Institution: Saiseikai Research Institute of Health Care and Welfare (2021-2023); Yokohama City University (2020)
• Principal Investigator: TORIMOTO IZDUMI 社会福祉法人恩賜財団済生会(済生会保健・医療・福祉総合研究所研究部門), 済生会保健・医療・福祉総合研究所研究部門, 客員研究員 (30757464)
• Project Period (FY): 2020-04-01 – 2024-03-31
• Keywords: 医療安全 / 偶発的所見 / 人工知能プログラム

Outline of Final Research Achievements

This study introduced a system used within the target medical institutions to address various medical safety issues related to insufficient confirmation of diagnostic imaging reports, and presented considerations and issues, while also attempting to implement artificial intelligence that will help detect life-threatening incidental findings in tests where no image reading was requested.
As there was a particularly high rate of cases in which thrombi were found incidentally (12.6%), efforts were made to create artificial intelligence that will make it possible to identify thrombi and help assist diagnostic imaging physicians in their image reading work.

Free Research Field

医療安全

Academic Significance and Societal Importance of the Research Achievements

画像診断報告書の確認不足に関する医療安全上の問題解決策は、各医療機関の規模と人的パワー（画像診断医の数、患者さんの総数など）に応じて様々な戦略に基づき実施されていることが各々の診療機関に係る学会発表により確認されており、2021年の研究発表も参考戦略の一つとして公地された。
偶発的所見として見つかる血栓については、医療行為としての血栓防止策に加えて、血栓の発見および評価を可能ならしめる人工知能が存在すれば、より安全で安心な医療サービスの一助となると考えられる。