2026 Biomedical Informatics Seminars
February 27, 2026
Alex Wellsjo, PhD, Assistant Professor of Economics and Strategy at the Rady School of Management
Abstract: The use of telemedicine has been on the rise. Although telemedicine has increased access to health care, little is known about how the medium changes providers’ medical decision-making. To evaluate how telemedicine differs from in-person care, we compare the quality and cost of in-person clinic visits and virtual visits over the phone for common conditions in Rwanda. To control for patient selection, we conduct an audit study with 2,532 standardized patient visits, where individuals portraying real patients presented standardized cases for malaria and upper respiratory infection (URI). We find that the quality of virtual care is higher than that of in-person care for URI and equally as good for malaria. Telemedicine providers also asked more questions about symptoms and medical history and prescribed more optional medicines for symptom management than in-person providers. We further find that telemedicine is more efficient than in-person care: virtual consultations were faster, had shorter wait times, resulted in fewer unnecessary medications and tests, and cost less for patients. Controlling for a rich set of provider characteristics, we show that provider selection does not appear to drive the results. Instead, better provider-patient communication over the phone emerges as a key mechanism. Providers report that it is easier to treat, get information from, and relate to patients over the phone. Providers also report feeling less social pressure during phone consultations. Consistent with survey evidence, we find that providers prescribe unnecessary antibiotics when asked by patients face-to-face, but not over the phone.
Bio: Alex Wellsjo is an Assistant Professor of Economics and Strategy at the Rady School of Management. Wellsjo's research covers topics in applied microeconomics, with a focus on questions related to health and household finance. As a behavioral economist, she incorporates insights from psychology into economic models to better understand individual behavior. She is especially interested in applications that relate to performance at work, with insights for firms and individuals on how to effectively boost productivity. Wellsjo received her Ph.D. in Economics from UC Berkeley and completed her postdoctoral training at the Haas School of Business.
February 20, 2026
Marc Niethammer, PhD, Professor, Computer Science and Engineering at UC San Diego
Abstract: This talk will cover some recent machine learning approaches for medical image computing with a focus on obtaining practical approaches. One focus will be on Image registration; image registration approaches have been extensively developed over the previous decades with a shift to approaches based on deep learning over the last decade. However, while modern deep learning registration approaches allow for highly accurate and fast registrations many existing approaches are task-specific. Hence, these approaches require extensive retraining or fine tuning for a new registration task. This talk will provide an overview of approaches to obtain deep registration networks that generalize across image types and registration tasks. It will also discuss some recent work on building deep registration models with desirable equivariance properties as a further step towards registration models that work well in practice and that behave more predictably. If time permits this talk will also cover some recent work on multimodal machine learning models motivated by the desire for an early diagnosis of lupus.
Bio: Marc Niethammer is a professor in Computer Science and Engineering with a joint appointment in Neurological Surgery at the University of California, San Diego (UCSD). He holds the Halıcıoğlu Endowed Chair in Health AI and leads the Biomedical Image Analysis Group. Before joining UCSD, he was a Professor of Computer Science at the University of North Carolina at Chapel Hill (UNC) from 2008 to 2024. Dr. Niethammer’s work focuses on methods for statistical shape analysis, image segmentation, image registration, machine learning, and related applications.
February 13, 2026
Abstract: While Artificial Intelligence dominates the healthcare conversation, its practical application often fails to improve patient outcomes. Dr. Michael Tsai argues that this failure stems from a misunderstanding of the technology's role. In this session, Dr. Tsai will share case studies from his work in clinical informatics and digital health to explore the evolution of care delivery. He will analyze the functional difference between "tools" and "agents” and propose that the solution to our healthcare crisis lies not in the technology itself, but in how we leverage it to restore the foundation of healthcare.
Bio: Ming-tse (Michael) Tsai, MD, MHI, FAMIA, ACHIP, is a Taiwanese physician based in San Diego with a multifaceted background in medicine and informatics. Throughout his career in clinical practice and the public sector, he has spearheaded hospital- and city-wide informatics initiatives to advance critical care in both hospital and community settings. Now a co-founder of an international digital health company, he focuses on bridging the gap between patients and clinicians, harnessing AI to empower patients across the U.S., Japan, and Taiwan.
February 6, 2026
Yasir Tarabichi, MD, Chief Health AI Officer and Director of Clinical Research Informatics at MetroHealth, Associate Professor of Medicine at Case Western Reserve University School of Medicine
Abstract: The rapid growth of electronic health record (EHR) data has created unprecedented opportunities for secondary research, yet persistent challenges in interoperability, data quality, and analytic validity limit the reliability and generalizability of findings derived from routine clinical data. This seminar examines vendor-supported platforms for large-scale EHR data aggregation, using national and multinational efforts as case examples to illustrate how pre-existing standards, health information exchange infrastructure, and semi-automated mapping and quality control processes are enabling cross-organizational research at previously unattainable scale. Through applications in chronic disease surveillance, infectious disease monitoring, medication utilization, immunization patterns, and safety signal detection, we evaluate the scientific and public health utility of these systems. We then analyze key limitations, including structured-data dependence, variable data quality and FAIRness, statistical constraints related to privacy protection, limited analytic flexibility, and the difficulty of distinguishing association from causation in platform-mediated observational data. Particular attention is given to emerging large-scale medical foundation models trained on longitudinal EHR event streams, highlighting the distinction between predictive pattern recognition and clinically actionable inference.
Bio: Yasir Tarabichi, MD, MSCR, is a clinical informaticist and physician leader focused on the safe, high-impact use of health data and AI in real-world care. He serves as Chief Health AI Officer and Director of Clinical Research Informatics at MetroHealth and is an Associate Professor of Medicine at Case Western Reserve University School of Medicine.
Dr. Tarabichi’s work centers on interoperability, secondary use of EHR data, and the translation of advanced analytics and AI into clinical and operational practice. He has led and collaborated on multi-institutional studies using large-scale EHR-derived datasets, with a focus on data quality, model validation, and responsible implementation. His perspective blends clinical medicine, research informatics, and health system operations, with a particular interest in how governance, standards, and platform design shape what is scientifically possible - and what is safe - at the point of care.
January 23, 2026
Ji Hoon Kim, MD, MPH, PhD, Associate Professor of Emergency Medicine at Yonsei University College of Medicine; Principal Researcher at the Yonsei Institute for Digital Health
Abstract: Emergency medicine faces time-critical decisions under uncertainty, making it an ideal domain for AI-enabled clinical innovation. In this talk, I will share how our group integrates large-scale health data and multimodal clinical signals (e.g., ECG waveforms, physiologic time-series, medical imaging, and clinical text) to develop and validate AI tools across emergency and pre-hospital care. I will introduce key technical trends driving AI-enabled medicine, including multimodal transformers and foundation models, as well as the growing role of large language models (LLMs) and agentic AI for clinical workflows. Using examples from our national projects in Korea—such as an AI-ambulance program and ED-focused digital solutions—I will discuss a practical framework for digital convergence clinical research: moving from in-silico validation to on-premise integration and prospective evaluation, while addressing model drift, data governance, and real-world impact. The session will emphasize how to generate trustworthy scientific evidence that supports safe adoption in clinical systems and improves both patient outcomes and clinician experience.
Bio: Ji Hoon Kim, MD, MPH, PhD is an Associate Professor of Emergency Medicine at Yonsei University College of Medicine (Seoul, Korea) and a Principal Researcher at the Yonsei Institute for Digital Health. He received an MPH in Epidemiology and a PhD in Preventive Medicine, and his work focuses on AI-enabled emergency and pre-hospital care, multimodal clinical modeling, and real-world validation of digital health solutions—including national-scale projects such as an AI-ambulance program.
January 16, 2026
Andrew Wong, MD MS, Clinical Instructor and Research Fellow, Internal Medicine, University of Michigan Medical School
Abstract: The rapid advancement of clinical AI systems over recent years has raised the challenge of providing adequate governance of these novel technologies. This talk will explore the rapid evolution of clinical AI tools, evaluate their impact on clinical care, and highlight essential considerations for oversight both before and after real-world implementation.
Bio: Dr. Wong is a Research Fellow in the National Clinician Scholars Program and a Clinical Instructor of Internal Medicine at the University of Michigan. His research focuses in applying artificial intelligence and machine learning to solve real-world problems in clinical practice, hospital operations, and medical education. He was a recent recipient of the national Felicia Hill-Briggs GIM Research Award, and his research has been widely cited, including in the White House Blueprint for an AI Bill of Rights. He currently serves on the Michigan Medicine Clinical Intelligence Committee, the University of Michigan Task Force for Generative AI in Research, and the IHPI committee for AI in Health Policy.
January 9, 2026
Daryl Cheng, MBBS MPH, Paediatrician, clinical informatician, board director & health consultant, The Royal Children’s Hospital Melbourne
Abstract: We will explore the digital journey of Australia’s largest paediatric hospital and academic centre, The Royal Children’s Hospital Melbourne. Starting for its EMR implementation, we will discuss the impacts on clinical mortality, as well as impacts on paediatric service provision and responsibility for Victoria, Australia.
Bio: Associate Professor Daryl Cheng is a consultant paediatrician, deputy CMIO and digital health consultant at The Royal Children’s Hospital Melbourne (RCH). He holds clinical research and educational posts at the Murdoch Children’s Research Institute (MCRI) and at the Department of Paediatrics, University of Melbourne. He is engaged in systems thinking, innovation and design work through appointments at the RCH in the Melbourne Children’s Campus Centre for Health Analytics (CHA) and RCH Digital Innovation | EMR Team. A/Prof Cheng is also Chief Medical Officer of Magentus.