Prediction of Blood Lactate Levels in Children After Cardiac Surgery Using Machine Learning Technique

Computer Engineering

Elevated blood lactate levels following cardiac surgery in children are often indicative of inadequate tissue perfusion and can signal the onset of critical post-operative complications. Early identification of patients at risk through predictive modeling can significantly improve clinical decision-making and outcomes. This study explores the application of machine learning algorithms to predict blood lactate levels in pediatric patients after cardiac surgery. By analyzing pre-operative, intra-operative, and early post-operative clinical data, various models are trained to estimate lactate concentrations and identify high-risk cases. Techniques such as regression and classification algorithms, including Random Forest, Support Vector Machines, and Gradient Boosting, are evaluated for performance. The model's accuracy is validated using standard metrics like Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and ROC-AUC where applicable. Results demonstrate that machine learning can serve as a reliable tool in predicting post-operative lactate levels, offering a non-invasive approach to support timely interventions in pediatric cardiac care

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.