Key result
Machine learning genomic profiling identifies AMI patients with ~89% accuracy using three gene markers.
Why the study?
Can the XGBoost method accurately classify AMI patients and identify relevant gene markers based on genomic profiling?
Can the XGBoost method accurately classify AMI patients and identify relevant gene markers based on genomic profiling?
Answers are grounded in this paper and cite their sources.
The XGBoost machine learning algorithm can accurately classify AMI patients and identified calponin 2, ribosomal protein S11, and myotropin as potential biomarkers.
One disease that can cause death is Acute Myocardial Infarction (AMI). AMI, also known as a heart attack, is a condition that causes permanent damage to heart muscle tissue due to prolonged ischemia or lack of blood flow that occurs due to blockage of the epicardial coronary arteries and results in blood clots and limiting blood supply to the myocardium. During the years the young AMI patients are increasing. One of the ways to diagnose early is providing information of biomarkers related to this disease by implementing the bioinformatics data analysis. The research was conducted to look at the genomic profile of patients suffering from AMI based on without recurrent events and normal control, using the XGBoost method, due to its scalability and efficiency. Based on the grid search of tuning hyperparameters, the XGBoost method gives a classification accuracy of 88.89%, AUC 90 and kappa 0.7805. These results indicate that the XGBoost method can classify patients suffering from AMI well. This research has identified three genes that contribute the most to classifying AMI patients, namely calponin 2, ribosomal protein S11 and myotropin. Based on the heatmap visualization, information was obtained that the three genes are class markers without recurrent events.
No takes yet. Share an insight, caveat, or question.
Fajriyah et al. (2024) studied Acute Myocardial Infarction (AMI). Genomic profiling using Extreme Gradient Boosting (XGBoost) vs. Normal controls was evaluated on Classification accuracy (AUC 90). Genomic profiling using the XGBoost method classified AMI patients without recurrent events from normal controls with 88.89% accuracy and an AUC of 90, identifying three key gene markers.
Synapse has enriched 10 related papers. Consider them for comparative context: