Design Analysis of Stroke Risk Prediction Model Employing the Hybrid Structure Implementation of Deep Transfer Learning System

Stroke has been given the highest priority among all widespread diseases which effectuate mortality or interminable debility in aging individuals over the globe. The study aims to progress the conventionally developed stroke risk design and predict the risk level based on the dataset collected from various sources concerning weighing factors. Moreover, the study considers employing artificial intelligence attributed to the comparative analysis of risk estimated with the probability of occurring in 10 years. A framework of Deep learning can outperform all the conventional models even with the dependency on big data. Because of the stringent confidentiality safety policy in healthcare systems, the data related to stroke is typically delivered among numerous hospitals in insignificant parts. It is a proven state-of-the-art technology for modeling stroke risk prediction. Nevertheless, the instances of data with positive and negative values undergo life-threatening and imbalanced cases. Therefore, a novel technique of Hybrid Deep Transfer Learning-based Stroke Risk Prediction arrangement for the potential to develop the configuration of multiple associated sources (such as chronic diseases, external stroke data, such as diabetes and hypertension). Extensive investigation has been conducted for the proposed framework in both synthetic and real-world circumstances. Finally, outcomes demonstrate the ability of real-world implementation, including several clinics facilitated with 5G/B5G substructures.