BioGAP-Ultra: A Modular Edge-AI Platform for Wearable Multimodal Biosignal Acquisition and Processing

Area of Science:

• Wearable technology
• Biomedical engineering
• Artificial intelligence in healthcare

Background:

• Growing demand for continuous physiological monitoring and human-machine interaction in real-world settings.
• Need for flexible, low-power wearable platforms with on-device intelligence.
• Limitations of existing wearable biosensing platforms in terms of storage, connectivity, and signal modality.

Purpose of the Study:

• To present BioGAP-Ultra, an advanced multimodal biosensing platform.
• To enable synchronized acquisition of diverse electrophysiological and hemodynamic signals.
• To achieve state-of-the-art energy efficiency for embedded AI processing on wearable devices.

Main Methods:

• Developed BioGAP-Ultra, an extension of the BioGAP design with increased storage, improved wireless connectivity, and enhanced signal modalities.
• Integrated BioGAP-Ultra into various wearable form factors (headband, sleeve, chestband).
• Deployed on-device AI applications for photoplethysmography (PPG) and electromyography (EMG) signal analysis.

Main Results:

• BioGAP-Ultra supports synchronized acquisition of EEG, EMG, ECG, and PPG signals.
• Demonstrated low power consumption for continuous monitoring and AI processing across different form factors (e.g., 9.3 mW for ECG-PPG chestband).
• Achieved high accuracy for on-device AI applications, such as EMG-based motion phase classification (79.9% ± 5.7% at 23.6 mW).

Conclusions:

• BioGAP-Ultra is a versatile and energy-efficient multimodal biosensing platform for advanced wearable applications.
• The platform enables on-device AI processing for real-time physiological monitoring and human-machine interaction.
• Open-source release of hardware and software facilitates further research and development in wearable biosensing.