BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系 - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://ece.hku.hk
X-WR-CALDESC:Events for Department of Electrical and Computer Engineering (HKUECE) 電機與計算機工程系
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:20250101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Hong_Kong:20260410T100000
DTEND;TZID=Asia/Hong_Kong:20260410T110000
DTSTAMP:20260511T035832
CREATED:20260402T035329Z
LAST-MODIFIED:20260402T035329Z
UID:115545-1775815200-1775818800@ece.hku.hk
SUMMARY:RPG Seminar – Efficient AI for Neural Signal Decoding in Healthcare Applications
DESCRIPTION:Zoom Link:\nhttps://hku.zoom.us/j/93628966280 \nAbstract\nRecent advances in artificial intelligence have brought new momentum to healthcare driven by neural signal decoding. This seminar presents three lines of our research on AI-based neural signal analysis across diverse application domains\, including wearable EEG-based epileptic seizure detection\, emotion recognition in SSVEP-based brain–computer interfaces (BCIs)\, and multimodal neuroimaging-based tinnitus diagnosis. First\, for real-time epileptic seizure detection on resource-constrained wearable devices\, we propose a multi-scale LBP-based hyperdimensional computing framework that captures seizure-related temporal dynamics with a compact model size\, strong few-shot learning capability\, and improved interpretability. Second\, to enhance emotion-aware interaction in SSVEP-based BCIs\, we develop a valence-arousal disentangled representation learning method that separates core emotional factors\, extracts global affective features\, and improves cross-subject generalization. Third\, for objective tinnitus diagnosis\, we introduce TinnitusLLM\, a multimodal large language model that integrates EEG and fMRI through neuro-inspired positional encoding\, multimodal autoregressive pretraining\, and subject-invariant cross-modal fine-tuning. Across these studies\, our common goal is to build efficient\, interpretable\, and clinically meaningful learning frameworks for robust neural decoding in real-world healthcare and human–machine interaction scenarios. \nSpeaker\nMr Yipeng Du\nDepartment of Electrical and Computer Engineering\nThe University of Hong Kong \nBiography of the Speaker\nYipeng Du is currently pursuing the Ph.D. degree in the Department of Electrical and Electronic Engineering at The University of Hong Kong. He received the B.E. degree in Communication Engineering from the University of Science and Technology Beijing and the M.E. degree in Signal and Information Processing from Peking University. His research focuses on developing deep learning methods for neural signal processing in healthcare\, particularly for disease diagnosis\, monitoring\, and brain–computer interface applications. \nOrganiser\nProf Edith C.H. Ngai \nDepartment of Electrical and Computer Engineering\, The University of Hong Kong \nAll are welcome.
URL:https://ece.hku.hk/events/20260410/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://ece.hku.hk/wp-content/uploads/2024/11/rpg-seminar.jpg
END:VEVENT
END:VCALENDAR