Citation

Subramanian, Suresh and Sungheetha, Akey and R., Rajesh Sharma and Suresh, Bavani and Kavandiran, Indumathy (2026) AI-Enhanced Adaptive Stream Processing for Smart Healthcare Blood Flow Monitoring Systems: A Novel Neuromorphic-Driven Approach for Real-Time Cardiovascular Assessment. Procedia Computer Science, 282. pp. 773-782. ISSN 18770509

Text 12.pdf - Published Version Restricted to Repository staff only Download (412kB)

Abstract

Cardiovascular diseases represent the leading cause of mortality globally, necessitating advanced real-time monitoring solutions for early detection and intervention. Traditional blood flow monitoring systems face significant challenges in processing highvelocity physiological data streams while maintaining diagnostic accuracy and energy efficiency. This research addresses critical limitations in existing healthcare monitoring frameworks by proposing a novel neuromorphic-driven adaptive stream processing architecture for real-time cardiovascular assessment. The proposed methodology integrates spiking neural networks with adaptive data fusion mechanisms to enable continuous blood flow analysis with minimal computational overhead. Our framework achieves 94.7 percent accuracy in detecting abnormal cardiovascular patterns while reducing processing latency by 67.3 percent compared to conventional approaches. The system demonstrates energy consumption reduction of 58.2 percent through neuromorphic computing paradigms while maintaining real-time response capabilities of 12.4 milliseconds for critical event detection. Experimental validation using synthetic cardiovascular datasets derived from standardized medical databases reveals superior performance across multiple metrics including sensitivity of 92.8 percent, specificity of 95.3 percent, and F1-score of 93.6 percent. The novelty lies in the synergistic combination of neuromorphic processing with adaptive stream fusion algorithms, enabling scalable deployment in resource-constrained medical IoT environments while ensuring clinical-grade diagnostic reliability for smart healthcare applications.

Item Type:	Article
Uncontrolled Keywords:	Cardiovascular assessment, Medical IoT
Subjects:	R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	02 Sep 2026 08:44
Last Modified:	02 Sep 2026 08:45
URII:	http://shdl.mmu.edu.my/id/eprint/16551

Downloads

Edit (login required)