A prediction model for Myocardial injury after non-cardiac surgery (MINS) was developed by researchers from South Korea using machine learning techniques with the extreme gradient boosting algorithm. The study analyzed 6811 patients who underwent non-cardiac surgery between January 2010 and June 2019.
The top 12 variables affecting MINS were preoperative cardiac troponin level, inotropic drug infusion, operation duration, emergency status, operation type, age, high-risk surgery, body mass index, chronic kidney disease, coronary artery disease, blood transfusion, and current alcohol use. Two prediction models were developed. The main model used all 12 variables, while the light model used the top 6 variables. These are available online. https://sjshin.shinyapps.io/mins_occur_prediction/
The prediction models, available online, have an accuracy of 0.97 with an AUC of 0.78 (main) and 0.77 (light) and require further verification.
Oh, A.R., Park, J., Shin, S.J. et al. Prediction model for myocardial injury after non-cardiac surgery using machine learning. Sci Rep 13, 1475 (2023). https://doi.org/10.1038/s41598-022-26617-w
Medical News Observer 
Leave a Reply