Predictive factors affecting hepatitis patients survival results via application of the machine learning methods.

Hepatitis, caused by viruses A-E, can silently progress to liver damage, cirrhosis, or cancer. Chronic B and C increase failure risk. Machine learning models help predict hepatitis risks using patient data, symptoms, and history. This study used Decision Tree Classification (DTC) and Extreme Gradient Boosting Classification (XGBC) with three optimizers Rhizotomy Optimization Algorithm (ROA), Gold Rush Optimizer (GRO), and Motion-Encoded Electric Charged Particles Optimization Algorithm (MEPO) to enhance accuracy. Among hybrids, DTRO achieved the highest accuracy (0.991), outperforming DTC. XGRO followed with 0.991, and DTME with 0.954. DTRO emerged as the most reliable model for predicting hepatitis survival.