Ischemic Stroke Prediction with B-LSTM based on ECG Signals

Ischemic Stroke is one of the dangerous diseases that are rapidly increasing in the modern era due to the constant variations in ecological aspects, comprising diet, lifestyles, and working environment. Rendering a new statement from the World Health organization (WHO), Ischemic Stroke was the instant major illness, after heart disorder that heads toward death. For both patient and medical personnel, one of the most important requirements was the early detection of ischemic stroke illnesses. Deep learning methods are one of the blooming techniques in modern healthcare disease diagnosis applications, which provide an accurate and instant diagnosis for the majority of critical diseases. Stroke-related diseases need immediate attention from medical professionals to avoid critical conditions for the patients. This research work proposed the deep learning framework approach for the detection of abnormalities in the Electro-cardio graphic abnormalities related to Ischemic Stroke disease. Heart-related diseases are one of the important risk factors for stroke disease. ECG plays an important part in the detection of numerous stroke risk factors. To predict stroke-related disorders using ECG data and other characteristics, this study presented a detection approach centered on Bi-direction Long Short-Term Memory (BLSTM) algorithm. The empirical analysis demonstrates that the deep learning approaches yield good results with 92% accuracy. Receiving operating characteristic (ROC) curves have also produced encouraging outcomes. This research established the model’s suitability for the early identification of heart structural anomalies, which are the primary cause of disorders like Ischemic Stroke.