Application of machine learning algorithms to predict heat-sensitive angina (HSA) attacks: a multicentric observational cohort study - Takeaways - MDSpire

Utilization of machine learning techniques to forecast episodes of heat-sensitive angina (HSA): findings from a multicenter observational cohort study

  • By

  • Jincheng Wang

  • Conghui Zhou

  • Yue Zhao

  • Jingqing Hu

  • July 21, 2026

Share

  • 1

    The study involved 1,246 individuals with stable angina from 43 clinical centers in China to develop a machine learning model for heat-sensitive angina.

  • 2

    A random forest model was developed using 14 predictors, demonstrating strong predictive performance during both training and external validation.

  • 3

    The model's performance was assessed using metrics such as AUC, sensitivity, specificity, and F1 score, indicating its robustness.

  • 4

    Exploratory analysis suggested that patients identified as high risk for heat-sensitive angina experienced increased angina frequency in summer.

  • 5

    The study highlights the need for further validation of the random forest model before it can be routinely applied in clinical settings.

Original Source(s)

Application of machine learning algorithms to predict heat-sensitive angina (HSA) attacks: a multicentric observational cohort study

  • by Jincheng Wang, Conghui Zhou, Yue Zhao, Jingqing Hu

  • July 21, 2026

Related Content

Plasma levels of soluble Flt-1 and Tie-2 correlate with neovascularization in human carotid atherosclerotic plaques – a pilot study

Efficacy and safety of pulsed field ablation and very high-power short-duration ablation for atrial fibrillation: a meta-analysis

Effects of the active parts of Yangxin Tongmai Formula on myocardial injury and spatial metabolic remodeling in rats with myocardial ischemia-reperfusion injury