Predictive Model for Early Detection of Cardiovascular, Renal and Neurodegenerative Disorder Using Machine Learning Approaches

This investigation is concerned with the difficulties of quickly diagnosing diseases while working with various disease-specific datasets. It addresses these issues in the context of cardiovascular illnesses, neurological problems, and severe renal ailments by applying approaches like dimensionality reduction, particularly using Principal Component Analysis (PCA). The study investigates how several machine learning methods, such as Random Forest, Naive Bayes, Support Vector Machine (SVM), K Nearest Neighbours (KNN), and Logistic Regression, perform when used after PCA. The study contrasts the efficacy of models with and without PCA to show the advantages of dimensionality reduction techniques. Notably, KNN predicts Chronic Kidney Disease with a flawless 100% accuracy whereas Logistic Regression predicts Cardiovascular Disease with an 85% accuracy. Additionally, when evaluated on the Alzheimer's Neuroimaging Initiatives (ADNI) cohort, the suggested machine learning algorithms demonstrate a remarkable 97.82% accuracy. This demonstrates the efficiency of Logistic Regression in predicting cardiac illness and backs up the applicability of KNN for identifying kidney-related problems.