COURSE DESCRIPTION
Advances in sensing and computing technologies now generate complex biomedical datasets across laboratory research, clinical environments, and decentralized home-care settings. These data are nonlinear, nonstationary, and often uncertain, which makes rigorous scientific foundations essential for the next generation of medical AI. This summer school focuses on the science behind data science: how AI models are formulated, analyzed, and quantitatively validated.
In this three-week program, we emphasize the central role of mathematical modeling and interpretability of algorithms used in medical AI. Participants will learn how to build models that encode physiological mechanisms, phenomenological behaviors, and system dynamics, and how these models guide algorithm design and statistical inference. We highlight how proper modeling contributes to interpretability, uncertainty quantification, and reliable clinical insights, and implement such highlight on hands-on practice through real biomedical case studies. Participants will also be introduced to the cycle of research, from idea to analysis to presentation.
Core topics include:
-
How to perform research.
-
Mathematical modeling of physiological and biomedical systems, from mechanistic formulations to data-driven structures, together with statistical quantification and uncertainty analysis.
-
Modern time-frequency analysis for nonstationary signals.
-
Unsupervised manifold learning for understanding intrinsic geometric structure in high-dimensional data.
-
Hands-on practice on real world biomedical signals.
The school is designed for PhD, master’s, and advanced undergraduate students interested in data science, medical AI, and interdisciplinary research at the interface of mathematics, statistics, engineering, and medicine.
ORGANZIERS
- Steven Altschuler, Systems pharmacology, University of California, San Francisco
- Hau-Tieng Wu, Mathematics, New York University/Academia Sinica
- Lani Wu, Systems pharmacology, University of California, San Francisco
- Shih-Hsien Yu, Mathematics, Academia Sinica
LIST OF INSTRUCTORS(tentative)
- Steven Altschuler, Systems pharmacology, University of California, San Francisco
- Stefan Borik, Electrical Engineering, University of Zilina
- Gi-Ren Liu, Mathematics, National Cheng Kung University
- Yi-Wen Liu, Electrical Engineering, National Tsing Hua University
- Govind Menon, Mathematics, Brown University
- Gal Mishne, Data Science, University of California, San Diego
- Hau-Tieng Wu, Mathematics, New York University/Academia Sinica
- Lani Wu, Systems pharmacology, University of California, San Francisco
- Junho Yang, Statistics, Academia Sinica
PREREQUISITES
Advanced senior & PhD level students.
- For students from math, stat, EE, CS: linear algebra, multivariable calculus and probability.
- For students from biomedicine: linear algebra and basic calculus
- All: Matlab or Python.
GENERAL INFORMATION
Time: July 1st (Wednesday)-July 21 (Tuesday): 3 weeks.
Location: IOM, Academia Sinica in the Astromathematics building
MORE INFORMATION
Email: hautiengw@gmail.com
CONTACT
For administrative supports, please contact
Email: conference@gmail.math.sinica.edu.tw
*Notice: Please include [Summer School 2026] in the subject line of your email, thank you!
