PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
July 10, 2024Open Access

Machine-Learning based Prediction Models for Healthcare Outcomes in Patients Participating in Cardiac Rehabilitation: A Systematic Review

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Machine-learning techniques are increasingly used to predict healthcare outcomes in cardiac rehabilitation, but a critical appraisal of existing ML-based prognosis predictive models and identification of key research gaps were needed.

Can machine-learning based prediction models accurately predict healthcare outcomes in patients participating in cardiac rehabilitation?

Population

22 ML-based clinical models from 7 studies across multiple phases of CR

Design

Systematic review

Authors

XTXiarepati TieliwaerdiKMKathryn ManaloAAAbulikemu Abuduweili

Discussion

Loading...

Member takes

Overview

ML models warrant caution for clinical use absent validation; leaves open their adoption pending external testing.

Structured PICO

Can machine-learning based prediction models accurately predict healthcare outcomes in patients participating in cardiac rehabilitation?

P
Population
Patients participating in cardiac rehabilitation (CVD patients) from 7 studies evaluating 22 ML-based clinical models, with cohort sizes ranging from 41 to 2280 patients.
I
Intervention
Machine-learning based prediction models
O
Outcome
Prediction of healthcare outcomes including patient intention to initiate cardiac rehabilitation, graduation from outpatient cardiac rehabilitation, and interval physiological and psychological response to cardiac rehabilitation

While machine learning models show good predictive capabilities for cardiac rehabilitation outcomes, their readiness for clinical implementation is questionable due to a lack of external validation and a high risk of bias.

Limitations

  • None of the models underwent calibration or external validation
  • Most studies raised concerns for bias
  • Most models were developed using smaller patient cohorts

Cite This Study

Tieliwaerdi et al. (2024) studied this question.

synapsesocial.com/papers/6a1d303333e2df9c962f1604https://doi.org/10.1101/2024.07.09.24310007
View Full Paper
Ask AI
Bookmark
Share