Log in to MyACC
Machine Learning-Based Prediction of Adverse Events After ACS
Quick Takes
- PRAISE risk score uses the advantages of artificial intelligence and machine-based learning to risk stratify patients with ACS.
- The PRAISE score (includes 16 clinical variables, five therapeutic variables, two angiographic variables, and two procedural variables) showed good discrimation power to predict both ischemic and bleeding events 1 year after ACS.
- Whether the score has equal applicability in women and minorities and whether tailoring therapy to personalize care based on the PRAISE score improves outcomes remains to be determined.
Study Questions:
Can a machine learning-based model predict adverse clinical events after acute coronary syndrome (ACS)?
Methods:
Different machine learning models for the prediction of 1-year post-discharge all-cause death, myocardial infarction (MI), and major bleeding (defined as Bleeding Academic Research Consortium type 3 or 5) were trained on a cohort of 19,826 adult patients with ACS (split into a training cohort [80%] and internal validation cohort [20%]) from the BleeMACS and RENAMI registries, which included patients across several continents. Twenty-five clinical features routinely assessed at discharge were used to inform the models. The best-performing model for each study outcome (the PRAISE score) was tested in an external validation cohort of 3,444 patients with ACS pooled from a randomized controlled trial and three prospective registries. Model performance was assessed according to a range of learning metrics including area under the receiver operating characteristic curve (AUC).
Results:
The PRAISE score showed an AUC of 0.82 (95% confidence interval [CI], 0.78–0.85) in the internal validation cohort and 0.92 (95% CI, 0.90–0.93) in the external validation cohort for 1-year all-cause death; an AUC of 0.74 (95% CI, 0.70–0.78) in the internal validation cohort and 0.81 (95% CI, 0.76–0.85) in the external validation cohort for 1-year MI; and an AUC of 0.70 (95% CI, 0.66–0.75) in the internal validation cohort and 0.86 (95% CI, 0.82–0.89) in the external validation cohort for 1-year major bleeding.
Conclusions:
A machine learning-based approach for the identification of predictors of events after an ACS is feasible and effective. The PRAISE score showed accurate discriminative capabilities for the prediction of all-cause death, MI, and major bleeding, and might be useful to guide clinical decision making.
Perspective:
The PRAISE risk score is a next-generation prediction tool, which avails the advantages of artificial intelligence and machine-based learning to stratify patient risk after ACS. This mammoth undertaking included derivation and validation of the PRAISE score from a predominantly male cohort of patients to assess 1-year risk of death, MI, and major bleeding after ACS. The PRAISE score (includes 16 clinical variables, five therapeutic variables, two angiographic variables, and two procedural variables) showed good discrimation power to predict all three clinical outcomes and provides reliable prediction of both ischemic and bleeding events. Whether the score has equal applicability in women and minorities and whether tailoring therapy to personalize care based on the PRAISE score improves outcomes remains to be determined.
Clinical Topics: Acute Coronary Syndromes, Cardiovascular Care Team, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Prevention, Interventions and ACS, Interventions and Imaging, Angiography, Nuclear Imaging
Keywords: Acute Coronary Syndrome, Artificial Intelligence, Coronary Angiography, Hemorrhage, Myocardial Infarction, Myocardial Ischemia, Patient Discharge, Risk Factors, Secondary Prevention
< Back to Listings
