Precision in Stroke Detection - Unleashing CNN, Bilstm and Genetic Algorithm

Cerebrovascular diseases, such as stroke, are a leading cause of disability worldwide. However, they are both preventable and treatable. A stroke occurs when the blood supply to the brain is disrupted, resulting in damage to brain cells. Early detection of strokes and rapid intervention can significantly improve survival rates and reduce long-term disability. In recent years, machine learning methods have gained considerable attention as tools for stroke detection. This study aims to predict the frequency of strokes and improve strategies for their prevention. The goal is to develop a stroke detection system using CT images of the brain, combined with a genetic algorithm and a Bidirectional Long Short-Term Memory (BiLSTM) network, to detect strokes at an early stage. For image classification, a genetic algorithm based on neural networks was employed to select the most relevant features. The research utilized feature selection algorithms that not only enhanced detection accuracy but also reduced computational requirements by optimizing feature selection. Validated on an image-based dataset, the system demonstrated high precision and recall.