Research Highlights

(A) Integration of Genetic Data, Electronic Health Records, and Behavioral Environmental Exposures in Precision Health Research

Objective: Through deep integrative analysis of datasets from the Taiwan Biobank, Taiwan Precision Medicine Initiative, and National Health Insurance Research Database, we aim to establish comprehensive Disease Gene Knowledge Databases and Disease Risk Assessment Systems, enhancing our understanding of disease etiologies and individual risk in disease development, enabling more effective prevention strategies.

(B) Integration Research and Product Implementation of Smart Health and Precision Health

Objective: By combining medical imaging, electronic health records, and multi-OMICS data, we strive to develop automated tools for sample segmentation, lesion annotation, disease diagnosis, and risk assessment systems. We aim to improve disease classification, prevention, and treatment response across various medical conditions by employing deep learning, machine learning, and statistical learning techniques.

(C) Generative AI in Smart Health

Objective: We are developing generative AI tools, including large language models and Generative Adversarial Networks, to handle complex medical reports and electronic health records, thereby expediting processes in clinical applications and enhancing efficiency in healthcare delivery.

Note. Image was generated by Image Creator.

(D) Infectious Disease Virus Research  

Objective: Our research involves extensive analysis of global virus data to track and comprehend virus variations, classifications, and evolution. Through the integration of datasets from the UK Biobank and the Global Influenza Data Sharing Initiative with Our World in Data, we aim to elucidate the risk and causal relationship between host genes and disease susceptibility/severity.

Publication:

Huang, Y.-J., Chen, C.-h. and Yang, H.-C.* (2024/05). AI-enhanced integration of genetic and medical imaging data for risk assessment of type 2 diabetes. Nature Communications 15, 4230. [SCI] (https://rdcu.be/dIiU4)

Yang, H.-C.*, Wang, J.-H., Yang, C.-T., Lin, Y.-C., Chen, P.-W., Liao, H.-C., Chen, C.-h. and Liao, J. C.* (2022/09). Subtyping of major SARS-CoV-2 variants reveals different transmission dynamics based on 10 million genomes. PNAS Nexus 1, pgac181. (https://doi.org/10.1093/pnasnexus/pgac181)

Yang, H.-C.*, Chen, C.-h., Wang, J.-H., Liao, H.-C., Yang, C.-T., Chen, C.-W., Lin, Y.-C., Kao, C.-H., Lu, M.-Y. J. and Liao, J. C.* (2020/12). Analysis of genomic distributions of SARS-CoV-2 reveals a dominant strain type with strong allelic associations. Proceedings of the National Academy of Sciences of the United States of America 117, 48. [SCI] (https://doi.org/10.1073/pnas.2007840117)