Real-Time Diagnostic Integrity Meets Efficiency: A Novel Platform-Agnostic Architecture for Physiological Signal Compression

IEEE-EMBS International Conference on Body Sensor Networks (IEEE BSN 2024)
Head-based signals such as EEG, EMG, EOG, and ECG collected by wearable systems will play a pivotal role in clinicaldiagnosis, monitoring, and treatment of important brain disorder diseases. However, the real-time transmission of thesignificant corpus physiological signals over extended periods consumes substantial power and time, limiting the viability of battery-dependent physiological monitoring wearables.This paper presents a novel deep-learning framework employing a variational autoencoder (VAE) for physiologicalsignal compression to reduce wearables’ computational complexity and energy consumption. Our approach achieves animpressive compression ratio of 1:293 specifically for spectrogram data, surpassing state-of-the-art compression techniques such as JPEG2000, H.264, Direct Cosine Transform(DCT), and Huffman Encoding, which do not excel in handling physiological signals. We validate the efficacy of thecompressed algorithms using collected physiological signalsfrom real patients in the Hospital and deploy the solutionon commonly used embedded AI chips (i.e., ARM Cortex V8and Jetson Nano). The proposed framework achieves a 91%seizure detection accuracy using XGBoost, confirming theapproach’s reliability, practicality, and scalability. arxiv link