What does this research mean for the field?

Machine learning can predict physical activity decline in cardiac patients with 70.8% accuracy and 91.7% sensitivity based on baseline digital health data. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

The research aims to determine if machine learning can predict declines in physical activity after cardiac interventions using digital health data.

January 14, 2026European Heart Journal - Digital HealthOpen Access

Prediction of physical activity decline post-cardiac intervention: a machine learning approach

View Full Paper

Ask AI

Bookmark

Key result

Machine learning predicts physical activity decline in cardiac patients with ~92% sensitivity using baseline digital data.

Why the study?

Physical activity after cardiac interventions predicts re-events and prognosis, but identifying patients at risk of physical activity decline remains challenging.

Can machine learning applied to multimodal digital health data predict which patients will experience a decline in physical activity following a cardiac intervention?

Comparison

Baseline multimodal digital health predictors of sustained physical activity decline

Design

Prospective cohort prediction model development and validation study

Follow-up

12 months

Authors

FBF J C Van BlerckRadboud University Nijmegen VEV A A Van EsRadboud University Nijmegen WGW F GoevaertsEindhoven University of Technology

Discussion

Member takes

Implication

Should not yet change practice; leaves open machine learning for risk stratification in post-intervention cardiac cohorts.

Key Points

The research aims to determine if machine learning can predict declines in physical activity after cardiac interventions using digital health data.
Followed 80 patients post-cardiac intervention for 12 months
Collected baseline multimodal data, including activity, sleep, and demographics
Utilized a tree-bagging ensemble classifier with Bayesian optimization for prediction
Evaluated model performance with five-fold cross-validation and a held-out test set
Defined physical activity decline as a reduction of ≥1,000 steps/day from baseline
Final model achieved 70.8% accuracy in predicting physical activity decline
Sensitivity was high at 91.7%
Area under the curve (AUC) was 0.833
Age, circadian rhythm amplitude, and fitness were key predictors of decline
Patients with low baseline activity and disrupted rhythms were at highest risk

Structured PICO

Can machine learning applied to multimodal digital health data predict which patients will experience a decline in physical activity following a cardiac intervention?

Population

80 patients (mean age 73, 89% male) who underwent cardiac interventions including coronary revascularisation (PCI or bypass surgery), radiofrequency catheter ablation or electrophysiology study, or transcatheter aortic valve implantation or valve surgery.

Intervention

Machine learning model (tree-bagging ensemble classifier) using multimodal digital health data (physical activity, sleep, heart rate, circadian rhythm metrics, lifestyle behaviour, clinical, psychosocial, and demographical characteristics) collected at baseline.

Outcome

Future physical activity decline, defined as a sustained reduction of ≥1,000 steps/day from baseline during the follow-up period of one year.surrogate

Machine learning applied to baseline digital health and clinical data can accurately identify patients at high risk for physical activity decline after cardiac interventions, potentially enabling targeted early interventions.

Cite This Study

Blerck et al. (2026) studied this question. Machine learning achieved 70.8% accuracy and 91.7% sensitivity in predicting physical activity decline in cardiac patients based on baseline digital health data.

synapsesocial.com/papers/69671985c0d1e3cfbfce8eb5 https://doi.org/10.1093/ehjdh/ztaf143.061

Topics

STEMI management Percutaneous coronary intervention

View Full Paper

Ask AI

Bookmark

January 14, 2026European Heart Journal - Digital HealthOpen Access

Prediction of physical activity decline post-cardiac intervention: a machine learning approach

View Full Paper

Ask AI

Bookmark

Key result

Machine learning predicts physical activity decline in cardiac patients with ~92% sensitivity using baseline digital data.

Why the study?

Physical activity after cardiac interventions predicts re-events and prognosis, but identifying patients at risk of physical activity decline remains challenging.

Can machine learning applied to multimodal digital health data predict which patients will experience a decline in physical activity following a cardiac intervention?

Comparison

Baseline multimodal digital health predictors of sustained physical activity decline

Design

Prospective cohort prediction model development and validation study

Follow-up

12 months

Authors

FBF J C Van BlerckRadboud University Nijmegen VEV A A Van EsRadboud University Nijmegen WGW F GoevaertsEindhoven University of Technology

Discussion

Member takes

Implication

Should not yet change practice; leaves open machine learning for risk stratification in post-intervention cardiac cohorts.

Key Points

The research aims to determine if machine learning can predict declines in physical activity after cardiac interventions using digital health data.
Followed 80 patients post-cardiac intervention for 12 months
Collected baseline multimodal data, including activity, sleep, and demographics
Utilized a tree-bagging ensemble classifier with Bayesian optimization for prediction
Evaluated model performance with five-fold cross-validation and a held-out test set
Defined physical activity decline as a reduction of ≥1,000 steps/day from baseline
Final model achieved 70.8% accuracy in predicting physical activity decline
Sensitivity was high at 91.7%
Area under the curve (AUC) was 0.833
Age, circadian rhythm amplitude, and fitness were key predictors of decline
Patients with low baseline activity and disrupted rhythms were at highest risk

Structured PICO

Can machine learning applied to multimodal digital health data predict which patients will experience a decline in physical activity following a cardiac intervention?

Population

Intervention

Outcome

Future physical activity decline, defined as a sustained reduction of ≥1,000 steps/day from baseline during the follow-up period of one year.surrogate

Cite This Study

synapsesocial.com/papers/69671985c0d1e3cfbfce8eb5 https://doi.org/10.1093/ehjdh/ztaf143.061

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

Topics

STEMI management Percutaneous coronary intervention

View Full Paper

Ask AI

Bookmark