Key result
Machine learning using STS variables predicts significant ischemic mitral regurgitation with an AUC of 0.80.
Why the study?
Machine learning models have potential to identify non-intuitive and previously unrecognized relationships between standardized clinical variables and pathophysiological conditions, motivating evaluation of STS Database variables to identify clinically significant IMR in CABG patients.
Can machine learning models using standard STS database variables predict the presence of clinically significant ischemic mitral regurgitation in patients undergoing CABG?
Observational (n=7,005)
No
Can machine learning models using standard STS database variables predict the presence of clinically significant ischemic mitral regurgitation in patients undergoing CABG?
Effect estimate: AUC 0.80
Machine learning models using readily available STS database variables can predict the presence of clinically significant ischemic mitral regurgitation with good accuracy (AUC up to 0.80) in patients undergoing surgical revascularization.
No takes yet. Share an insight, caveat, or question.
May aid preoperative identification of ischemic MR in CABG patients using routine data; hypothesis-generating pending validation and outcome trials.
MacGregor et al. (2021) conducted an observational in Clinically significant ischemic mitral regurgitation (IMR) (n=7,005). Machine learning models (RF, SVM, LR, DNN) was evaluated on Presence of clinically significant ischemic mitral regurgitation (AUC 0.80). Machine learning models using Society of Thoracic Surgeons database variables predicted the presence of clinically significant ischemic mitral regurgitation with an AUC up to 0.80.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: