Human Gastrointestinal Gas as a Non-Invasive Biomarker: A Wearable IoT and Edge AI Framework for Real-Time Health Monitoring

This paper presents a novel non-invasive framework for real-time monitoring of gastrointestinal health using analysis of human intestinal gas emissions. The proposed system integrates wearable gas sensors with edge artificial intelligence (AI) to detect key biomarkers such as methane (CH₄), hydrogen (H₂), and hydrogen sulfide (H₂S). These biomarkers are strongly correlated with digestive disorders including irritable bowel syndrome (IBS), small intestinal bacterial overgrowth (SIBO), and gastrointestinal infections. The architecture combines flexible sensor arrays, embedded processing, and cloud-based analytics for continuous monitoring and anomaly detection. Additionally, this work explores the feasibility of micro-scale energy harvesting from methane emissions to support ultra-low-power IoT systems. Experimental simulations demonstrate promising classification accuracy and real-time responsiveness, highlighting the potential of gas-based diagnostics in next-generation digital healthcare systems.