Health Informatics · AI for Healthcare
MSc in Health Informatics
Karolinska Institutet & Stockholm University
Dept. of Learning, Informatics, Management and Ethics (LIME)
Researcher at the intersection of machine learning, clinical AI, and epidemiology. My work focuses on intelligent systems for fall risk prediction, multimodal sensor fusion, and population health analytics using large-scale biomedical databases.
About
I am a researcher working at the intersection of health informatics, machine learning, and epidemiology. My primary focus is on developing intelligent systems for predictive health management — particularly fall risk assessment and geriatric care — through multimodal deep learning and wearable sensor fusion.
With a clinical background in preventive medicine from Sun Yat-sen University and ongoing graduate training in health informatics at Karolinska Institutet & Stockholm University, I bridge the gap between data-driven AI and real-world clinical applications.
My published work includes transformer architectures for skeleton-based fall prediction, multimodal IMU-image fusion networks, global disease burden analysis using GBD data, and graph neural networks for epidemiological nowcasting. I have served as Principal Investigator on a nationally-funded research project on gait-based risk identification in the elderly.
Education
Sep 2025 — Jun 2027 (Expected)
Master of Science in Health Informatics
Dept. of Learning, Informatics, Management and Ethics (LIME) · Joint Master's Programme · 120 ECTS credits
Sep 2020 — Jun 2025
Bachelor of Medicine · Preventive Medicine
School of Public Health · Shenzhen Campus
GPA 3.4 / 4.0Publications
MIEF-Net: Multimodal Image-Enhanced Fusion Network for Intelligent Fall Risk Prediction
Neural Networks, vol. 195, article 108260, Oct. 30, 2025.
MSS-Former: Multi-Scale Skeletal Transformer for Intelligent Fall Risk Prediction in Older Adults
IEEE Internet of Things Journal, vol. 11, pp. 33040–33052, Jun. 28, 2024.
Assessing and projecting the global burden of thyroid cancer, 1990–2030: Analysis of the Global Burden of Disease Study
Journal of Global Health, vol. 14, article 04090, Apr. 5, 2024.
MIEFP-Net: A Multimodal Image-Enhanced Network for Fall Prediction Using IMU Data
The 15th ACM Conference on Bioinformatics, Computational Biology, and Health Informatics, 2024. (Poster)
Identifying sensors-based parameters associated with fall risk in community-dwelling older adults: an investigation and interpretation of discriminatory parameters
BMC Geriatrics, published Feb. 1, 2024.
Fig. IMU sensor placement at L4-ASIS and 6-axis signal output (AccX/Y/Z, EulerX/Y/Z) used for gait parameter extraction in community-dwelling older adults.
Enhancing City-Level Influenza Nowcasting on Island Terrain with Graph Neural Networks: Spatial Feature Insights
Intelligent Systems Conference. Cham: Springer Nature Switzerland, August 2024.
An advanced integrated sensor-based method for fall risk assessment in rehabilitation setting
IEEE Sensors Journal (Accepted, 2023 IF: 4.3)
Fig. 3-Metre Timed Up-and-Go (3MTUG) rehabilitation assessment protocol with integrated IMU sensors for fall risk evaluation.
Research
Mar 2022 — Present
Sun Yat-sen University School of Public Health (Shenzhen) · A.P. Yang Zhao Research Group
PI & Team MemberDeveloped predictive models for personalized health management and elderly care using machine learning and deep learning. Concentrated on sensor-based health monitoring — integrating motion recognition technologies, spectral analysis (FFT/GAF), and spatial-temporal skeletal transformers (MSS-Former) — for early detection and risk stratification of geriatric fall risk. This work produced 5 peer-reviewed publications including in Neural Networks (IF 6.3) and IEEE IoT Journal (IF 10.6).
Dec 2022 — Dec 2023 | Funded Jan 2024 — Dec 2024
Innovation & Entrepreneurship Training Program 2023 · Shenzhen Medical Research Fund (A2301041)
Project Leader / Principal InvestigatorLed the development of a real-time gait risk identification model for elderly individuals using inertial measurement units (IMU). Applied Gramian Angular Field (GAF) image encoding and multi-branch ResNet architecture to convert 6-axis IMU signals into RGB feature images for fall prediction. The innovation project received the highest school review rating of "Excellent". The subsequent national grant (¥50,000, SMRF) extended this work into an intelligent intervention system combining large deep learning models with deep imaging data.
Jul 2023 — Present
Sun Yat-sen University School of Public Health (Shenzhen) · Prof. Yiqiang Zhan Research Group
Team MemberInvestigating epidemiological characteristics and survival outcomes of neurodegenerative diseases using advanced statistical methodologies. Applied Mendelian randomization techniques to examine causal relationships in genetic epidemiology; leveraged large-scale databases (UK Biobank, NHANES, GBD) for comprehensive data mining. Published global disease burden analysis on thyroid cancer in Journal of Global Health (IF 7.66). Two additional manuscripts on diabetes and neurodegenerative disease burden are currently under review.
Dec 2021 — Nov 2022
Innovation & Entrepreneurship Training Program 2022 · Sun Yat-sen University
Team MemberEstablished an early predictive model for Alzheimer's Disease risk using meta-analyses of modifiable risk factors. Identified key early-stage AD risk factors and trained six machine learning classifiers (Logistic Regression, SVM, Random Forest, XGBoost, etc.) achieving high predictive accuracy. The project was awarded "Excellent" rating for innovation and practical utility.
Skills
Awards & Funding