Advanced Computational Techniques In Stroke Prediction: Analyzing Cardiac And Vascular Bio-Signals

Stroke is a potentially fatal or disabling condition, making it crucial to promptly identify the early warning signs. Ischemic and hemorrhagic strokes are clinically differentiated, necessitating immediate management with thrombolytic or coagulant medication. Initially, it is imperative to identify the prompt indicators of a stroke, which may differ across individuals and necessitate the involvement of a specialist and admission to a hospital. Previous research has primarily focused on post-stroke treatment rather than stroke prevention. Recent research has shown a growing dependence on image processing, specifically MRI and CT scans, to identify and forecast prognostic indicators in individuals who have suffered from strokes. Identifying such processes in real time poses difficulties, and assessing these processes incurs costs and time constraints. This work introduces a multi-modal bio-alert system that utilizes photoplethysmography (PPG) and electrocardiogram (ECG) data to diagnose and evaluate cognitive stroke prognostic indicators in older individuals. We have created a stroke prediction gadget that uses a combination of regression algorithms to accurately estimate the occurrence of stroke infections in real-time, even when the user is walking outdoors. Elderly individuals can conveniently transport bio-signal sensing devices due to the meticulous apparatus that captures bio-indicators while walking at a sample rate of 1,000 Hz, either from the three conductors of the ECG or the forefinger for PPG. The performance and efficiency of the accurate prediction were proven through the real-time study of stroke patients in the senior age group.