Development of Real-time AI for Predicting Lesion Formation and Steam-pop during High-frequency Ablation

2023 Fiscal Year Final Research Report

• Project/Area Number: 22K16068
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 53020:Cardiology-related
• Research Institution: Tokyo Medical and Dental University
• Principal Investigator: Masateru Takigawa 東京医科歯科大学, 大学院医歯学総合研究科, 寄附講座講師 (40760062)
• Project Period (FY): 2022-04-01 – 2024-03-31
• Keywords: 不整脈 / catheter ablation / radiofrequency / 焼灼巣 / 機械学習

Outline of Final Research Achievements

RF applications were performed in ex-vivo environment with several configurations. Machine learning was performed using the data showing the relation between active indices, passive indices, and detailed lesion metrics, in order to create the model predicting the lesion characteristics and steam-pops. Patent was applied with this development （TOKUGAN2023-119252；TOKUGAN2023-110489）. This model showed significantly more accurate predictive accuracy. Since the first model was developed based on the ex-vivo model, we are now working to develop the novel model based on the in-vivo data. During this project, several abstracts were presented in a scientific session, and the several manuscript was published.

Free Research Field

不整脈

Academic Significance and Societal Importance of the Research Achievements

本研究の結果、通電中に変動するパラメータより自動的に、焼灼巣の形状が予測されることにより、より効率的で安全な高周波アブレーション治療が可能になる可能性がある。