Key result
Machine learning using STS variables predicts one-year IMR repair durability with ~0.75 AUC.
Why the study?
To assess the potential predictive contribution of standardized Society of Thoracic Surgeons Database clinical variables for repair durability in ischemic mitral regurgitation patients using machine learning.
Can machine learning models using STS database variables predict 1-year repair durability (recurrence of ≥3+ IMR or death) in patients undergoing surgical revascularization and mitral valve repair for ischemic mitral regurgitation?
Observational (n=224)
No
Can machine learning models using STS database variables predict 1-year repair durability (recurrence of ≥3+ IMR or death) in patients undergoing surgical revascularization and mitral valve repair for ischemic mitral regurgitation?
Effect estimate: AUC 0.72-0.75
Machine learning models using readily available STS database variables can predict 1-year repair durability in patients with ischemic mitral regurgitation with promising accuracy.
No takes yet. Share an insight, caveat, or question.
Hypothesis-generating for STS-based ML durability prediction in ischemic MR repair; requires prospective validation before clinical adoption.
Kachroo et al. (2021) conducted an observational in Ischemic mitral regurgitation (n=224). Machine learning models using STS Database variables was evaluated on Recurrence (≥3+ IMR) or death versus nonrecurrence (<3+ IMR) (AUC 0.72-0.75). Machine learning models using STS Database variables predicted 1-year repair durability (recurrence of ≥3+ IMR or death) in ischemic mitral regurgitation patients with an AUC of 0.72-0.75.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: