Multimodal Wearable Biosensing Meets Multidomain AI: A Pathway to Decentralized Healthcare.

Area of Science:

• Biomedical Engineering
• Artificial Intelligence
• Digital Health

Background:

• Wearable biosensors offer continuous health monitoring, while AI can fuse diverse data streams for personalized analytics.
• Current sensor-AI systems often lack integration with longitudinal clinical data and medical knowledge, limiting their reasoning capabilities.
• Decentralized healthcare models require robust systems that bridge real-world data with clinical decision-making.

Purpose of the Study:

• To propose a framework uniting multimodal wearable biosensing with multidomain AI for decentralized healthcare.
• To review advancements in wearable sensing and AI for improved patient profiling, diagnostics, and risk prediction.
• To outline AI pipelines integrating biosensor data with electronic health records and medical knowledge for evidence-based decision support.

Main Methods:

• Review of recent progress in multimodal wearable biosensing technologies.
• Summary of multisensory fusion techniques for enhanced patient profiling and risk prediction.
• Description of AI pipelines integrating biosensor data, electronic health records, and medical knowledge graphs.

Main Results:

• Multisensory fusion improves patient profiling, diagnostic accuracy, and early risk prediction.
• AI pipelines integrating diverse data sources support evidence-grounded clinical decision support.
• The proposed framework enables a shift towards AI-enabled decentralized healthcare, moving insights from clinic to daily life.

Conclusions:

• Integrating multimodal biosensing with multidomain AI offers a pathway to decentralized healthcare.
• Addressing challenges in data quality, privacy, and generalization is crucial for real-world implementation.
• Emerging AI techniques like continual learning and retrieval-augmented reasoning hold promise for future advancements.