• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2023 Fiscal Year Final Research Report

The Construction of A Prognostic Model for Heart Failure in Deep Learning

Research Project

  • PDF
Project/Area Number 21K10287
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 58010:Medical management and medical sociology-related
Research InstitutionToho University

Principal Investigator

NAKAMURA KEIJIRO  東邦大学, 医学部, 講師 (20366181)

Co-Investigator(Kenkyū-buntansha) 朱 欣  会津大学, コンピュータ理工学部, 上級准教授 (70448645)
Project Period (FY) 2021-04-01 – 2024-03-31
Keywords心不全 / 予後因子 / 人工知能
Outline of Final Research Achievements

Artificial intelligence (AI) is expected to able to find regularities in complex data and analyze them to find new findings. Real World Data consists of heterogeneous patient groups, which is sometimes difficult to perform multiple regression analysis due to data bias and confounding factors.1273 heart failure patients were analyzed to predict the outcome from DPC database. The network-based model showed better prediction performance than a deep feed-forward neural network-based model and Cox proportional hazard model in identifying patients with high risk of mortality. AI model demonstrated an improved prognostic prediction performance in consideration of temporal information from time-varying covariates that could assist clinical decision-making. In this study, AI model in heart failure found that significant risk and protective factors of mortality were specific to risk levels and causes of mortality, highlighting the demand in an individual-specific clinical strategy.

Free Research Field

循環器 心不全 人工知能

Academic Significance and Societal Importance of the Research Achievements

人工知能(AI)を用いたリアルワールドデータ分析が広く行われるようになっている。人工知能は多層のニューラルネットワークを使用して複雑なデータの中から規則性や特徴を見つけだし、新たな発見をする能力が期待されている。従来の統計手法による解析は、心不全のようなheterogeneousで構成されている集団に対しては、患者背景のばらつきやデータの偏りによって高い精度での解析に困難であったが、本研究においてAIによる予後因子の解析をすることで従来では描出できなかった因子を高い精度で描出することが可能であった。本研究の結果は、今後の心不全のデータベース研究および心不全診療において高い臨床的意義が得られた。

URL: 

Published: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi