Deep Learning Enabled Intelligent Healthcare Management System in Smart Cities Environment

Abstract

In recent times, cities are getting smart and can be managed effectively through diverse architectures and services. Smart cities have the ability to support smart medical systems that can infiltrate distinct events (i.e., smart hospitals, smart homes, and community health centres) and scenarios (e.g., rehabilitation, abnormal behavior monitoring, clinical decision-making, disease prevention and diagnosis postmarking surveillance and prescription recommendation). The integration of Artificial Intelligence (AI) with recent technologies, for instance medical screening gadgets, are significant enough to deliver maximum performance and improved management services to handle chronic diseases. With latest developments in digital data collection, AI techniques can be employed for clinical decision making process. On the other hand, Cardiovascular Disease (CVD) is one of the major illnesses that increase the mortality rate across the globe. Generally, wearables can be employed in healthcare systems that instigate the development of CVD detection and classification. With this motivation, the current study develops an Artificial Intelligence Enabled Decision Support System for CVD Disease Detection and Classification in e-healthcare environment, abbreviated as AIDSS-CDDC technique. The proposed AIDSS-CDDC model enables the Internet of Things (IoT) devices for healthcare data collection. Then, the collected data is saved in cloud server for examination. Followed by, training and testing processes are executed to determine the patient’s health condition. To accomplish this, the presented AIDSS-CDDC model employs data pre-processing and Improved Sine Cosine Optimization based Feature Selection (ISCO-FS) technique. In addition, Adam optimizer with Autoencoder Gated Recurrent Unit (AE-GRU) model is employed for detection and classification of CVD. The experimental results highlight that the proposed AIDSS-CDDC model is a promising performer compared to other existing models.