Biosignal Analysis & Neurological Applications
Advancing automated analysis of biosignals including EEG, sleep monitoring, and neurological pattern recognition through deep learning and signal processing.
This research area focuses on developing AI systems that can automatically analyze and interpret complex biosignals, with particular emphasis on neurological applications such as seizure detection, sleep staging, and brain activity pattern recognition. The work combines advanced signal processing techniques with deep learning to create systems that can match or exceed expert-level performance in clinical signal interpretation.
Key innovations include transformer architectures for cross-data biosignal learning, expert-level automated EEG classification systems, and deep neural networks for sleep scoring. The research addresses critical challenges in neurological monitoring including signal variability across patients and devices, real-time processing requirements, artifact detection and removal, and the need for interpretable predictions in critical care settings.
The developed systems have direct clinical applications in intensive care monitoring, epilepsy management, and sleep medicine.
BIOT: Biosignal Transformer for Cross-data Learning in the Wild (NeurIPS 2024)
Development of Expert-level Classification of Seizures and Rhythmic Patterns during EEG Interpretation (Neurology, 2023)
Expert-level Sleep Scoring with Deep Neural Networks (JAMIA, 2018)
Interested in This Research Area?
We welcome collaborations with researchers, clinicians, and industry partners working in biosignal analysis & neurological applications. Our lab is always looking for motivated students and postdocs to join our team.