2024, Vol. 9, Issue 5, Part C

One of the dangerous medical conditions that can occur from a decrease or blockage of blood flow to a part of the brain is a stroke, which occurs when brain cells are deprived of oxygen and nourishment. This might result in death, disability, or brain damage if untreated. For long-term issues to be minimized and survival rates to rise, prompt medical intervention and early diagnosis are crucial. This work aims to investigate using fuzzy logic and elastic Net regularization to enhance stroke prediction accuracy. By utilizing fuzzy logic to manage uncertainty in risk factors, Elastic Net addresses multicollinearity among predictors. We evaluate the proposed models on a comprehensive demographic and clinical variables dataset. The results demonstrate improved predictive accuracy compared to traditional methods, highlighting the potential of these advanced techniques to provide valuable insights for stroke risk prediction and inform clinical decision-making.