Development of machine-learning based resuscitation strategy

2023 Fiscal Year Final Research Report

• Project/Area Number: 22K21143
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 0908:Society medicine, nursing, and related fields
• Research Institution: Kyoto University
• Principal Investigator: Okada Yohei 京都大学, 医学研究科, 特定研究員 (00966955)
• Project Period (FY): 2022-08-31 – 2024-03-31
• Keywords: 心停止 / 蘇生 / 機械学習

Outline of Final Research Achievements

The results of this study showed that resuscitation strategies using membrane artificial lungs potentially improve the prognosis (survival rate and good neurological outcomes) of out-of-hospital cardiac arrest patients in the region by using machine learning analysis of data from out-of-hospital cardiac arrest patients in Osaka and Singapore. In addition, an analysis using advanced statistical methods (time-dependent propensity score matching) based on a nationwide database of out-of-hospital cardiac arrests in Japan showed that resuscitation strategies using extracorporeal membrane oxygenation could potentially improve the prognosis of out-of-hospital cardiac arrest patients. He also published a review article summarizing the Japanese emergency resuscitation system and resuscitation statistics database.

Free Research Field

救急

Academic Significance and Societal Importance of the Research Achievements

本研究の結果は機械学習などの解析手法を用いて、従来では救命困難であった難治性の院外心停止患者の蘇生戦略を検証した研究である。本研究を通じて、膜型人工肺を用いた蘇生戦略がその予後改善に寄与する可能性について示した。本研究結果は膜型人工肺を用いた蘇生戦略を推進するエビデンスとなり、さらなる院外心停止患者の予後改善につながることことが期待される。