Artificial intelligence-based analysis of chest X-ray to predict hemodynamic parameters

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K17559
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 53020:Cardiology-related
• Research Institution: Mie University
• Principal Investigator: Toba Shuhei 三重大学, 医学部附属病院, 助教 (20806111)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: 人工知能 / 胸部X線写真 / 先天性心疾患 / 血行動態

Outline of Final Research Achievements

We developed artificial intelligence to predict pulmonary to systemic flow ratio, an important hemodynamic index for patients with congenital heart disease, from chest radiographs, and applied for a patent. In addition, we have developed an artificial intelligence to predict hemodynamics from chest radiographs in the field of adult cardiology, and an artificial intelligence that automatically classifies pediatric 12-lead electrocardiograms. The results were presented in a journal and at national and international conferences.

Free Research Field

先天性心疾患

Academic Significance and Societal Importance of the Research Achievements

本研究は、人工知能により胸部X線写真から血行動態（カテーテル検査結果）を予測できることを世界で初めて示し、それが小児・成人を問わず様々な血行動態指標に応用可能であることを示した。より低侵襲かつ簡便な胸部X線写真から血行動態を正確に予測できれば、小児・成人循環器診療において、正確な血行動態評価を頻回に行うことができるようになり、より優れた医療の実現につながる。