Year: 2024
Type: conference-paper
Abstract: Diabetes is a chronic illness that requires continual control. Due to the rising prevalence of diabetes, innovative technological solutions are needed to control healthcare costs and predict diabetes onset. This research presents a machine learning system that combines supervised and unsupervised learning to monitor illness progression and deliver real-time projections to patients. We will compare our method to others by evaluating public diabetic datasets. This study lays the groundwork for future research on machine learning and diabetes care. 5G's rapid development of Internet of Things technology has affected healthcare, notably remote patient monitoring. Lack of common frameworks, inefficient network resource optimisation, energy waste, and data security concerns hinder 5G-IoT application in smart healthcare. A four-part system architecture is proposed by the authors to overcome these concerns. First up: a Cisco Packet Tracer-certified smart healthcare system. IoT and 5G are used. Module 2 introduces the network-efficient Fast Library for Automated Machine Learning (FLAML). The third module collaborates with the Proximity-based Service to increase D2D energy efficiency using an Optimal Enumeration Algorithm (OEA). The fourth module discusses malware risks in 5G-IoT healthcare applications that use CNNs for detection and classification. CNN models outperform competitors in power efficiency and accuracy, laying the groundwork for healthcare infrastructure protection.
Authors: Venkata Achuta Rao S, Yaakulya Sabbani, GNV Vibhav Reddy, V. Sambasiva Rao, Dinesh Goyal, Pankaj Dadheech
Language: English
Topic: Internet of Things and AI
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4413915571