Predicting Acute Liver Failure in Obese Population: A Machine Learning Approach

The aim of this research is to examine how obesity influences organ disease development, with a special focus on the development of the severe organ disease Acute Liver Failure (ALF). Well known risk factors for various diseases, including liver related conditions, are obesity, especially when defined by excessive visceral fat. Although the association between obesity and liver disease has been widely studied, the impact of obesity on progression to ALF is poorly understood. For these studies, we employ a variety of machine learning models, including KNearest Neighbors (KNN) and Logistic Regression, Support Vector Machine (SVM) to forecast its likelihood of ALF in obese individuals for a range of demographic and health related features including waist circumference, blood pressure and body mass index (BMI). Several performance metrics, e.g., precision, accuracy, ROC-AUC, F1 score, recall were employed for analysis. A comparison of Logistic Regression (LR), SVM, and KNN methods for detecting Acute Liver Failure (ALF) in obese patients forms the basis of this study. SVM stands out as the top performer among the tested models because it delivers 99.83% accuracy and 99.67% precision and 99.77% recall alongside a 99.72% F1 score and 99.99% ROC-AUC. Among the tested models Logistic Regression demonstrated 98.55% accuracy whereas KNN achieved the lowest performance at 92.42%.