A stroke, also known as a brain attack, is a life-threatening condition that happens when blood flow to the brain is disrupted either through a blockage in an artery or bleeding in the brain that results in a neurological impairment. This study provides a comprehensive analysis of stroke prediction by using machine learning models on demographic health data and CNN models on CT scanned images. Age, smoking status, hypertension, and gender were identified as strong predictors of stroke risks. XGBoost’s superior recall and F1-score surpassed Random Forest, highlighting its sensitivity to true positive cases. The CNN model achieved 96Future works include expanding the dataset to improve diversity, integrating with Electronic Health Record (EHR) systems, which are digital platforms for storing and managing patients medical information and refining the model through continuous feedback to improve the model prediction across diverse population.