Scientific Sessions

Session 1: The Architecture of Causal Evidence: Translating Targeted Learning into Real-World Data Reality

Schedule: Thursday, October 22, 2026, 9:50 am - 11:30 am

As the volume of Real-World Data (RWD) expands, the pharmaceutical industry faces a critical bottleneck: transforming unstructured data into rigorous, actionable clinical evidence.

This session explores the end-to-end architecture required to achieve this, bridging cutting-edge causal inference methodology with modern data infrastructure and pharmaceutical execution.

Attendees will gain a comprehensive understanding of how advanced statistical frameworks - specifically Targeted Learning and Target Trial Emulation - are moving beyond academic theory to directly impact clinical differentiation and improve the Probability of Success (PoS) in drug development.

Session 2: High-Dimensional Statistical Learning for Biomedical Data

Schedule: Thursday, October 22, 2026, 1:00 pm - 2:40 pm

This session presents new theory and methodology for high-dimensional statistical learning in modern biomedical applications, from cross-population genetic prediction to representation learning from electronic health records.

Dr. Ye Tian studies cross-population polygenic risk score (PRS) prediction, which leverages large datasets to improve prediction in smaller populations. Using a high-dimensional sparse linear regression framework with population-specific and genetically correlated shared signals, the work characterizes how optimal estimation depends on genetic correlation, sample sizes, and signal and noise variances, and shows that Bayesian estimators can achieve near-optimal performance where some global-penalization transfer learning methods do not.

Dr. Mengyan Li introduces SCORE, a semi-supervised representation learning framework for clustering and embedding high-dimensional count data such as EHR and RNA sequencing. Built on a Poisson-adapted latent factor mixture model with a hybrid EM and Gaussian variational algorithm, SCORE uses a small labeled subset to refine estimation on large unlabeled data, with convergence guarantees and error rates. It is demonstrated on predicting disability status in multiple sclerosis from EHR data.

Session 3: AI, Machine Learning, and Digital Health in Clinical Trials

Schedule: Thursday, October 22, 2026, 2:50 pm - 4:30 pm

This session examines how advanced technologies are reshaping clinical trials as artificial intelligence (AI), machine learning (ML), and digital health tools become embedded across the drug development life cycle - from trial design and patient selection to endpoint assessment and regulatory decision-making.

The session will share case examples leveraging predictive models, wearable-derived endpoints, and real-world data.

The session will provide a concise roadmap for deploying AI and digital tools responsibly, recognizing common pitfalls, and navigating the evolving regulatory landscape that is redefining the future of quantitative science in drug development.

Session 4: Real-World Impact of Artificial Intelligence in Medicine

Schedule: Thursday, October 22, 2026, 4:40 pm - 5:40 pm

This showcase bridges the gap between academic research and pharmaceutical industry application by bringing together leading experts to highlight the real-world impact of artificial intelligence.

Moving past theoretical hype, the session features a curated selection of high-impact, concrete case studies demonstrating how AI is actively transforming medicine.

Featured presentations will explore verified success stories across accelerated drug discovery, clinical trial optimization, and translational research.

By focusing on evidence-based examples, this event provides a practical blueprint for cross-disciplinary collaboration and offers a clear view of how data-driven innovation is driving the next generation of biomedical solutions.

Session 5: Digital Health Technologies and Wearable Sensors in Clinical Research

Schedule: Friday, October 23, 2026, 9:00 am - 10:30 am

Emerging digital technologies—such as wearable sensors and at-home self-assessments—capture physiological, functional, and behavioral data remotely in patients' everyday environments. By enabling frequent, unbiased measurement across the full functional spectrum, these tools reveal clinical insights that traditional point-in-time visits often miss.

Realizing their potential requires new strategies for study design, quantitative methods, and data and computing infrastructure so that digital biomarkers and outcome measures meet high standards of clinical research and development.

In this session, three digital health industry veterans will share practical guidance for statistically rigorous methods supporting strategy, experimental design, instrument operationalization, and synthesis of results. They will illustrate approaches to monitoring disease progression and to developing and validating measures, with examples from narcolepsy and Parkinson's disease.

Opening remarks will be delivered by Dr. Jaroslaw Harezlak—an academic pioneer in wearable devices and Chair of the Department of Epidemiology and Biostatistics at Indiana University School of Public Health-Bloomington. He will also moderate the panel discussion and Q&A session.

Session 6: Statistical Inference and Machine Learning for High-Dimensional Genomics

Schedule: Friday, October 23, 2026, 10:45 am - 12:15 pm

This session will highlight emerging methods at the intersection of rigorous statistical inference and modern machine learning for analyzing high-dimensional genomic data.

Session 7: Explainable Artificial Intelligence (XAI) for Interpretable Models

Schedule: Friday, October 23, 2026, 1:15 pm - 2:45 pm

Explainable Artificial Intelligence (XAI) refers to a set of methods and techniques in artificial intelligence (AI) that aim to make the decision-making processes of AI systems understandable to human users.

The primary goal of XAI is to provide transparency, accountability, and interpretability in AI models, particularly those that are complex and often considered "black boxes," such as deep learning models.

Session 8: Recent Advancement on Bayesian Methodology in Clinical Trials for Drugs and Biologics

Schedule: Friday, October 23, 2026, 3:00 pm - 4:30 pm

Session Organizers: Dr. Ming-Hui Chen and Dr. Chenguang Wang.

This is going to be a Bayesian session.

This session mainly focuses on recent advancement or reactions in responding to a recent FDA landmark draft guidance, "Use of Bayesian Methodology in Clinical Trials of Drug and Biological Products," jointly released by the Center for Drug Evaluation and Research (CDER) and the Center for Biologics Evaluation and Research (CBER).