MEDIPREDICT: Heart and Cataract disease prediction using SVM and CNN

This extensive manuscript delineates a system that predicts cardiovascular and cataract-related conditions through the application of Support Vector Machine (SVM) and Convolutional Neural Network (CNN) methodologies. The system comprises two discrete predictive models corresponding to each medical condition. The attributes provided by users encompass blood pressure (bp), chronological age, sex, lipid levels, fasting blood glucose (FBS), resting electrocardiogram (restECG) measurements, the maximum heart rate attained (THALACH), exercise-induced angina (EXANG), ST segment depression resulting from exercise relative to rest (oldpeak), the slope of the peak exercise ST segment (slope), the quantity of major coronary arteries visualized via fluoroscopy (ca), and the classification of thalassemia (Thal), all instrumental in forecasting cardiovascular disease. Conversely, in the context of cataract diagnosis, thermal imaging of the ocular region serves as the input data, subsequently analyzed utilizing CNN techniques. Initially, thermal images undergo conversion into grayscale and binary formats to facilitate the extraction of pertinent features. The architectural framework, modalities for input, methodologies for feature extraction, as well as the Support Vector Machine (SVM) and Convolutional Neural Network (CNN) models employed in the predictive analysis are meticulously elucidated within this study. Furthermore, it examines the potential of the system to facilitate the early identification and management of cardiac and cataract-related ailments, both of which are likely to enhance patient prognoses.