The increasing significance of predictive modeling in healthcare, especially in forecasting patient survival across various diseases. As medical data become more abundant and complex, statistical approaches and machine learning algorithms offer promising tools to analyze and extract meaningful insights for personalized treatment and improved patient outcomes. This study aims to explore the performance of various machine learning models in survival prediction, considering the diverse clinical scenarios presented by diseases such as End-Stage Renal Disease (ESRD) and diabetes. This study evaluated the performance of Artificial Neural Networks (ANN), Kaplan–Meier Estimation, and Cox proportional hazard models on datasets comprising patients with End-Stage Renal Disease (ESRD) undergoing dialysis and diabetes. The study compares three approaches—Kaplan–Meier Survival, Cox Model, and ANN models—in predicting survival rates for End-Stage Renal Disease (ESRD) and diabetes. For ESRD, the ANN model achieved the highest accuracy (80%), precision (76%), recall (72%), specificity (85%), and F1-score (74%). Similarly, for diabetes, the Cox Model demonstrated high performance with accuracy (83%), precision (84%), recall (69%), specificity (92%), and F1-score (76%). These findings highlight the effectiveness of machine learning models in predicting survival outcomes for specific diseases. The findings reveal that the ANN algorithm excels in predicting the survival of dialysis patients with the highest specificity, accurately identifying those not at risk of death. Conversely, the Cox model proved most effective for predicting survival outcomes in diabetes patients. These results underscore the importance of selecting an appropriate model based on the specific disease characteristics.