Seminars

Abstract: This talk examines maternal and infant health outcomes within the Epic Cosmos dataset, comparing them against established perinatal research to assess validity and clinical utility. Attendees will explore key outcomes such as severe maternal morbidity, postpartum hemorrhage, neonatal length of stay, and hypoxic-ischemic encephalopathy, while gaining insight into the methodology for validating large datasets and their application in real-time surveillance of maternal and neonatal health trends.

Bio: Dr. Kevin Dysart rejoined the Neonatal division at the Children’s Hospital of Philadelphia in April of 2025 after serving at Nemours Children’s Health Wilmington as the Division Chief of Neonatal & Perinatal Medicine from 2021-2025. Previously, he served as the Associate Medical Director of the Newborn/Infant Intensive Care Unit at the Children’s Hospital of Philadelphia. He received his medical degree from Hahnemann University and a Bachelor of Science in Chemistry from Villanova University. Dr. Dysart completed his master’s degree in Biomedical Informatics from the Perelman School of Medicine at the University of Pennsylvania.

He has been actively involved in conducting clinical research, both clinical trials and epidemiology, including pre-translational work. His interests include clinical research focusing on large data sets and the study and applications of machine learning, prediction, and probability.

Abstract: The progression of complex diseases is a dynamic process shaped by the interplay of biology, environment, and lifestyle. Understanding the mechanisms that drive these changes requires the integration of longitudinal, multimodal data. In a preliminary study, our integration of placental multiomics from 321 samples provided insight into the biology of common obstetric syndromes with overlapping clinical features. We ultimately identified a biomarker signature that distinguished fetal growth restriction with gestational hypertension disorder placentas from control and other obstetric syndrome placentas. Building on this, we launched an observational study collecting longitudinal deep phenotyping data from 436 pregnant individuals, from their first prenatal visit to delivery. This comprehensive dataset includes multiomics from blood, urine, placenta, and vaginal and gut microbiome samples; clinical data from electronic health records; and survey, wearable (FitBit), and environmental data (air and water quality). For 133 participants, including 77 with pregnancy complications, we have processed multiomics data across five key timepoints. Our goal is to first define the normal systems-level trajectory of a healthy pregnancy and then identify deviations from this trajectory that are associated with complications. We will then pinpoint the earliest signs of deviation for an individual. Our findings can inform personalized clinical interventions, such as novel biomarker signatures and therapeutic targets. Ultimately, this work is designed to drive the next generation of precision obstetric care.

Bio: Samantha Piekos, Ph.D., is an Assistant Professor of Informatics at the University of Pennsylvania. Her research focuses on applying artificial intelligence and a systems biology approach to analyze longitudinal deep-phenotyping data, including multiomics, wearable, and environmental exposures, to improve maternal-fetal health. Her work aims to define healthy pregnancy trajectories and identify early deviations to inform the development of personalized interventions and precision obstetric care. She completed her doctoral work at Stanford University and postdoctoral work at the Institute for Systems Biology. She has also served as Visiting Research Faculty at Google and is a recipient of the prestigious NIH K99/R00 Pathway to Independence Award.

Abstract: Recent progress in end-to-end learning has led to highly accurate models for automated radiological image interpretation. Yet, most models lack transparency and fail to ground predictions in a clinically meaningful foundation—contrasting sharply with the way radiologists diagnose by referencing shared medical knowledge. In this talk, I will present CheXomni, a foundation model designed to bridge this gap by aligning model predictions with established clinical observations. Trained on over 0.87 million image-report pairs from 227,835 chest X-ray studies, CheXomni uses large language model (LLM) embeddings to project radiological findings into a shared semantic space. This enables interpretable and auditable predictions while maintaining strong diagnostic accuracy. We demonstrate CheXomni’s effectiveness across four large, physician-annotated chest X-ray datasets spanning the U.S., Europe, and Asia. Beyond superior classification performance, CheXomni introduces novel capabilities: zero-shot pathology detection, radiological confounder auditing, and concept bottleneck modeling—all grounded in real clinical data. These advances mark a significant step toward transparent, generalizable, and trustworthy medical AI.

Bio: Dr. Tianyu Han is an Assistant Professor in the Department of Radiology at the University of Pennsylvania and a core faculty member of the AI2D Center. His research focuses on developing interpretable and generalizable AI systems for radiological data understanding. He has led work on multimodal models, medical image synthesis, and adversarial robustness, with applications spanning chest X-rays, breast MRI, and CT imaging. His work is published in Radiology, Nature Communications, JAMA, Science Advances, Cell Reports Medicine, and Nature Machine Intelligence. Dr. Han is also the lead author of MedAlpaca, an open-source framework for medical conversational AI, and is committed to building transparent, clinically useful AI for medical diagnosis.

Abstract: Individuals vary widely in how their brains age and these differences have important clinical consequences. For example, advanced age is the strongest known risk factor for Alzheimer’s disease (AD). This talk focuses on recent findings from our research suggesting that brain aging, as quantified using structural MRI, can modify a person’s risk for AD. Individuals with signs of advanced brain aging are more likely to progress from normal cognition to cognitive impairment (CI), whereas those with slower brain aging appear more resilient. New results highlight that regional brain aging patterns are indicative of CI conversion. We also find that chronic disease risk factors — such as cardiovascular and metabolic conditions, as well as traumatic brain injury — may lead to earlier or more pronounced brain aging. Importantly, we explore how women's health factors, including menopause and reproductive history, influence brain aging trajectories and cognition. These findings offer promising avenues for improving risk stratification and patient-specific disease prevention strategies based on quantification of aging in the brain.

Bio: Andrei Irimia, PhD is an associate professor in the Leonard Davis School of Gerontology at the University of Southern California, with courtesy appointments in biomedical engineering and quantitative biology. His research focuses on brain aging, traumatic brain injury, and Alzheimer’s disease, using advanced neuroimaging and quantitative methods to understand individual variability in aging trajectories and dementia risk. Dr. Irimia leads several NIH-funded studies examining how chronic disease variables and women's health factors influence brain aging and neurodegeneration. His work bridges population neuroscience and clinical neurology, with the goal of improving early detection and stratification of patients at risk for cognitive decline.

Abstract: TBA

Bio: TBA

Abstract: TBA

Bio: TBA

Abstract: TBA

Bio: Yonghyun Nam is a Research Assistant Professor of Informatics at the University of Pennsylvania’s Perelman School of Medicine. His research focuses on integrating multi-omics data—including genomics, proteomics, and electronic health records—to advance risk prediction for complex diseases. He develops machine learning frameworks that combine polygenic risk scores, proteomic biomarkers, and network-based models to identify individuals at elevated risk for a range of complex disorders.