Program
Program
Coming soon.
Short Courses
The 2026 ICSA China Conference will offer three optional short courses to be held on Friday 6/26. You may review the information for each short course below.
AM Short Course
C03. Statistical and Algorithmic Foundations of Diffusion Models
Time: 6/26 Friday 8:30AM-12:30PM (Half Day)
Abstract:
Diffusion generative models have emerged as a cornerstone of modern generative AI, delivering state-of-the-art performance across a wide range of data generation tasks. At their core, diffusion models seek to gradually transform pure noise into new data samples that emulate a target data distribution, accomplished by learning to reverse a forward stochastic process that progressively converts data into Gaussian noise. Despite their empirical successes, the statistical and algorithmic foundations of diffusion models remain far from mature. This lack of fundamental understanding limits their broader adoption, especially in applications that demand interpretability and reproducibility.
This short course provides a timely introduction to diffusion models and presents recent progress toward understanding their striking effectiveness, with an emphasis on core principles and statistical insights. We will examine the fundamental mechanisms of score-based diffusion models; characterize the statistical limits of learning score functions; analyze the convergence behavior of diffusion-based samplers; explore how these models adapt to unknown low-dimensional data structures; discuss conditional generation via diffusion guidance; and highlight ideas for accelerating inference through higher-order approximations. Throughout this short course, we will connect theoretical advances to practical applications, illustrating how fundamental insights can inform effective algorithm design.
Lecturer 1: Yuxin Chen
Professor, Department of Statistics and Data Science
Wharton School at the University of Pennsylvania
Lecturer 2: Gen Li
Assistant Professor, Department of Statistics
Chinese University of Hong Kong
Lecturer 3: Yuting Wei
Associate Professor, Department of Statistics and Data Science
Wharton School at the University of Pennsylvania
PM Short Course
C01. Applied Meta-Analysis Using R
Time: 6/26 Friday 1:30PM-5:30PM (Half Day)
Abstract:
Meta-analysis integrates evidence from diverse studies to support more reliable and efficient inference. As the cost of medical and public health research continues to rise, many clinical studies are conducted with relatively small sample sizes, limiting their statistical power to detect clinically meaningful effects. This often leads to inconsistent or even conflicting findings across studies. By synthesizing effect estimates from multiple investigations using meta-analytic techniques, researchers can effectively increase the overall sample size, improve statistical power, and generate more precise and generalizable conclusions. Given its substantial impact on evidence-based practice, methodological choices in meta-analysis have received increasing attention, and the development of new methods and software has become a rapidly expanding area of research.
This half-day short course provides a comprehensive overview of meta-analysis, covering both the theoretical foundations of common meta-analytic models and their practical implementation in the widely used, freely available software R. Drawing on real-world examples from medical and public health research, the course will guide participants through step-by-step analyses using appropriate R packages and functions, offering both conceptual understanding and hands-on experience. Prior knowledge of R is helpful but not required.
Lecturer: Yan Ma
Professor and Chair, Department of Biostatistics and Health Data Science
University of Pittsburgh
Full Day Short Course
C02. Bayesian Adaptive Designs for Oncology Clinical Trials
Time: 6/26 Friday 8:30AM-4:30PM (Full Day)
Abstract:
In this short course, we will delve into Bayesian clinical trial designs and their implementation, with a focus on early phase trials. We will first examine phase I dose finding and optimization trial designs. Our focus will be on model-assisted designs, which offer simplicity, flexibility, and excellent operating characteristics. We will also highlight state-of-the-art designs that support this effort. We will use real-world trial examples to illustrate these novel designs using freely available software.
Moving on to phase II trial design, we will introduce the Bayesian phase II design and demonstrate its practical application. Additionally, we will cover biomarker-based designs, such as enrichment and marker-stratified designs. We will explain the fundamental principles of Bayesian monitoring and decision-making, with a particular focus on sharing practical experience with the BOP2 design and its applications.
Furthermore, master protocol designs have developed rapidly and have been widely adopted in recent years. We will discuss the application of Bayesian approaches from the perspective of adaptive elements in master protocol designs, with particular emphasis on basket trials and platform trials. The focus of this short course is to bridge the gap between theoretical understanding and practical application. By the end of the course, attendees will have a solid understanding of how to implement Bayesian clinical trial designs in their own research.
Lecturer 1: Yong Zang
Associate Professor, Department of Biostatistics and Health Data Science
Co-Director of Clinical Research for the Biostatistics and Data Management Core, IU Simon Comprehensive Cancer Center
Indiana University School of Medicine
Lecturer 2: Fangrong Yan
Professor and Doctoral Supervisor, Director of the Department of Biostatistics
Director of the Center for Biostatistics and Computational Pharmacy
China Pharmaceutical University
Lecturer 3: Wenyun Yang
Ph.D. candidate
Department of Biostatistics
China Pharmaceutical University