Edge-AI-Powered Continuous Biomarker Monitoring and Lifestyle Intervention Platforms for Remote Prevention and Management of Chronic Liver Disease Progression

• Ada John

  ladoke Akintola university of technology

  Author

Chronic liver disease (CLD) progression from fibrosis to cirrhosis and hepatocellular carcinoma represents a significant global health burden, accounting for approximately 2 million deaths annually. Current clinical management relies on intermittent, invasive blood tests and clinic-based assessments that capture only episodic snapshots of disease status, often missing subtle but clinically meaningful deterioration between visits. This study addresses the critical gap in continuous, non-invasive monitoring by proposing and validating an Edge-AI-powered platform that integrates wearable biosensor data with machine learning for real-time CLD progression prediction and lifestyle intervention delivery. The research employs a design-based methodology combining retrospective analysis of clinical datasets (n=344 patients) with prospective simulation of wearable biomarker monitoring. The proposed hybrid deep neural network framework, incorporating SHAP-based feature optimization, achieved 92.50% classification accuracy under 10-fold cross-validation for cirrhosis risk stratification, significantly outperforming conventional serological scoring systems (FIB-4, APRI). The platform demonstrated end-to-end latency below one second for edge-based inference and successfully generated personalized lifestyle recommendations based on continuous biomarker trends. The study contributes a replicable framework for remote CLD management, demonstrating that Edge-AI-enabled continuous monitoring can provide early warning of clinical deterioration 2-4 weeks before conventional detection methods. Practical implications include reduced hospitalization rates through proactive intervention and improved patient quality of life through non-invasive, home-based disease management. This research establishes the foundation for transitioning hepatology from reactive complication management to proactive, predictive continuous care.

• pdf

Copyright (c) 2026 Ada John (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.