|
Now I am a master of science at Hong Kong University of Science and Technology(HKUST). In September 2026, I will pursue a PhD in Peking University, supervised by Prof. Shenda Hong. Previously, I worked with Shenda Hong, PhD, in national institute of health data science at Peking University and worked as a research intern in Department of Neurology, Harvard Medical School , working with M Brandon Westover, MD, PhD. Email: nickljlee0306[AT]gmail.com CV / Google Scholar / GitHub |
|
|
My research focuses on the intersection of artificial intelligence for healthcare, computational clinical research, and wearable devices, with a particular interest in multimodal learning for healthcare and physiological signal analysis. The goal of my research is to investigate how machine learning frameworks can be used to aid, understand and integrate multimodal clinical medical data.
|
|
|
|
Jun Li, Aaron D Aguirre, Valdery Moura Junior, Jiarui Jin, Che Liu, Lanhai Zhong, Chenxi Sun, Gari Clifford, M Brandon Westover, Shenda Hong NEJM AI, 2025 Paper 
This study introduces ECGFounder, a general-purpose electrocardiogram foundation model pretrained on over 10 million recordings to enhance cardiovascular disease assessment. By employing a unified framework for both single-lead and multilead data, the model achieves state-of-the-art performance and robust generalization across diverse downstream clinical diagnostic tasks. |
|
Jun Li, Daniel M Goldenholz, Moritz Alkofer, Chenxi Sun, Fabio A Nascimento, Jonathan J Halford, Brian C Dean, Mattia Galanti, Aaron F Struck, Adam S Greenblatt, Alice D Lam, Aline Herlopian, Chinasa Nwankwo, Dan Weber, Douglas Maus, Hiba A Haider, Ioannis Karakis, Ji Yeoun Yoo, Marcus C Ng, Olga Selioutski, Olga Taraschenko, Gamaleldin Osman, Roohi Katyal, Sarah E Schmitt, Selim Benbadis, Sydney S Cash, William O Tatum, Zubeda Sheikh, Wan Yee Kong, Grace Bayas, Niels Turley, Shenda Hong, M Brandon Westover, Jin Jing NEJM AI, 2025 Paper 
This study presents an AI system--SpikeNet2 that achieves expert-level accuracy in detecting epileptiform discharges within EEG recordings. By operating across both short and long timescales, the model effectively identifies specific epilepsy markers and predicts long-term seizure risk to assist in clinical diagnosis. |
|
Jun Li*, Che Liu*, Sibo Cheng, Rossella Arcucci, Shenda Hong Medical Image in Deep Learning (MIDL), 2023,(Oral Presentaion) Paper 
This study introduces a cross-modal training framework that aligns electrocardiogram (ECG) signals with natural language reports using a frozen language model to achieve zero-shot classification. By leveraging the semantic knowledge of a pre-trained language model, the proposed method enables the identification of heart diseases without requiring any annotated samples for new diagnostic categories. |
|
Jun Li, Yuzhu Guo IEEE Journal of Biomedical and Health Informatics (JBHI), 2023 Paper 
This study introduces a novel EEG-based method combining dynamic mode decomposition and analytic common spatial patterns (DMD-ACSP) to accurately detect and predict freezing of gait in Parkinson's disease. By extracting dynamic spatiotemporal coherent modes of brain activity, the approach significantly improves predictive performance and provides valuable insights for personalized clinical interventions. |
|
|
|
|
 |
Video An Intelligent Piano Education System based on the multi-modal deep learning. Computer vision, Music AI and photoelectric sensors are used to improve the quality of piano teaching. Committed to promoting piano education to children who are not rich but love piano. |
|
|
|
Clinical Data Animation Center, Harvard Medical School, Beth Israel Deaconess Medical Center
Summer Intern • Jul 2023 - Oct 2023 Advisor: Prof. M.Brandon Westover |
 |
|
Hong Lab, Peking University
Research Intern • Nov 2022 - Present Advisor: Prof. Shenda Hong |
 |
|
|
Updated at Apr. 2026
Thanks Jon Barron for this amazing template
|