The research reports a complete web-based platform developed for predicting multiple diseases like diabetes, under-nutrition anemia, thalassemia, and thrombocyte disorders with the help of machine learning technologies. Unlike older systems that dealt with individual disease prediction, our method relies on an integrated framework that incorporates various disease-specific datasets for more complex analyses to be performed and timely actions to be taken. For each type of disease, we apply a set of different classification schemes: Support Vector Machine, Random Forest, Logistic Regression, and Gradient Boosting. After evaluation, these algorithms are compared and the best model for each disease is selected. The system chooses the best predictor for each disease category based on the performance metrics. The platform is centered on being user-friendly while ensuring that it can be used effectively in clinical practice. Healthcare professionals and patients can enter the required values through an interface which subsequently provides the patients with diagnostic insights in no time.