Related Experiment Video
Updated: Apr 22, 2026
07:51
Hydra, a Computer-Based Platform for Aiding Clinicians in Cardiovascular Analysis and Diagnosis
Published on: September 26, 2018
7.1K
Telemedicine adoption in cardiology: Determinants and predictors identified using Bayesian Model Averaging and
Pascal Petit1, Jonathan Nübel2,3, Marie Josephine Walter2,3
1Institute of Engineering, Univ. Grenoble Alpes, CNRS, Grenoble INP, LIG, SANGRIA, Grenoble, France.
PLOS Digital Health
|April 20, 2026
Summary
Telemedicine (TM) use in cardiology is influenced by knowledge and attitudes. Machine learning accurately predicts which healthcare professionals (HCPs) will adopt TM, aiding targeted implementation.
Area of Science:
- Cardiology
- Digital Health
- Health Informatics
Background:
- Telemedicine (TM) adoption is growing in cardiology.
- Understanding factors influencing healthcare professionals' (HCPs) use of TM is crucial for effective implementation.
Purpose of the Study:
- To identify key determinants and predictors of TM use among German HCPs treating cardiology patients.
- To evaluate the predictive performance of Machine Learning (ML) models for TM adoption.
Main Methods:
- Secondary analysis of German cross-sectional survey data from 112 HCPs.
- Bayesian Model Averaging (BMA) for identifying TM determinants.
- Extreme Gradient Boosting (XGBoost) ML algorithm for predictive modeling and predictor importance analysis using Shapley additive explanations.
Main Results:
- 57% of HCPs reported using TM.
- BMA identified 12 determinants, including TM knowledge, being a cardiologist, female gender, and perceived suitability for heart failure management.
- XGBoost model demonstrated strong predictive performance (AUROC: 0.88), with key predictors including TM knowledge, cardiologist role, female gender, patient volume, and perceived suitability for specific conditions.
Conclusions:
- HCPs' knowledge and attitudes are significant drivers of TM adoption in cardiology.
- ML models can accurately predict TM use among HCPs, facilitating targeted interventions for safer and more effective TM implementation.
