Artificial intelligent system to propose a proper trial desgin for medical devices based on regulatory science

• Project/Area Number: 16K16417
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Medical engineering assessment
• Research Institution: Kyushu University
• Principal Investigator: Funakoshi Kouta 九州大学, 大学病院, 学術研究員 (60536853)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Project Status: Completed (Fiscal Year 2018)
• Budget Amount: ¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000); Fiscal Year 2018: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000); Fiscal Year 2017: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000); Fiscal Year 2016: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
• Keywords: レギュラトリーサイエンス / 試験デザイン / 医療機器 / 審査報告書 / 規制科学 / 人工知能

Outline of Final Research Achievements

There are 124 new medical devices application review reports. Ninety-eight medical devices reports were analyzed, after expulsion of 7 diagnostic devices, 6 accessary devices, 13 only clinical evaluation report. 161 trials were reffered in the review reoports. When more than 2 trials were reviewed, one trial was defined as pivotal trial. A database with pairs of one medical device one pivotal clinical trial.
There were enough sample size for machine learning, I performed a descriptive epidemiological analysis. Median time for approvals were; 378.5 for RCT, 496 for Historical Control, 364 for Objective Peformance Criterion / Performance Goal, 513.5 for before-and-after trial, 301 for descriptive statistics.

Academic Significance and Societal Importance of the Research Achievements

審査報告書は医薬品医療機器総合機構において公開されているものの、PDFでの公開であり、横断的な解析が困難であった。著者は、約10年間に作成された全ての審査報告書を、試験デザインの違いに注目してデータベース化した。試験デザインのみではなく、外資・内資の違い、分野の違い、被験者数の違いもデータベース化し、医療機器のレギュラトリーサイエンス研究がこのデータベースにより可能となる。