Seminars

Abstract: Periodontitis is a prevalent chronic inflammatory disease and a leading cause of tooth loss worldwide, with substantial implications for systemic health and health care utilization. Despite its burden, early identification of individuals at high risk for disease onset and progression remains challenging in routine clinical practice. This seminar will describe a data-driven framework for developing, testing, and validating prediction models for periodontitis using real-world electronic dental record (EDR) and linked health data. The talk will cover key methodological considerations, including cohort construction from longitudinal clinical data, outcome definition using contemporary periodontal classifications, feature engineering from structured dental and medical records, and model development using machine-learning approaches. Emphasis will be placed on internal and external validation strategies, transportability across heterogeneous populations and care settings, and evaluation of model performance, calibration, and clinical utility. The seminar will also discuss challenges related to data quality, missingness, bias, and fairness, as well as strategies such as federated learning to enable multi-institutional validation without sharing patient-level data. Together, this work highlights how predictive modeling can support earlier risk stratification, inform personalized preventive strategies, and advance the integration of clinical decision support tools into dental practice, while illustrating broader lessons for applying biomedical data science methods to oral health research.

Bio: Jay Patel (BDS, MS, PhD in Informatics) is an Assistant Professor and Director of the Center for Artificial Intelligence (AI), Data Science, and Informatics at Temple University Kornberg School of Dentistry. He is among the few in the US with formal training in both clinical dentistry and a doctorate in informatics, an interdisciplinary background that uniquely positions him to lead transformative AI-driven dental research. Dr. Patel’s work focuses on developing AI models and software applications using large-scale electronic health record data to predict disease onset and progression, with the ultimate goal of improving prevention strategies. His research also emphasizes the integration of oral and systemic health through the linkage of medical and dental records to enable real-time health information exchange and support comprehensive data-driven studies. He has developed over 40 natural language processing pipelines and electronic dental record quality metrics to extract valuable insights from unstructured clinical notes. Dr. Patel has been an invited speaker and panelist at leading institutions, including the National Institute of Dental and Craniofacial Research and the National Institutes of Health, where he has presented his pioneering work in dental AI. He has received several NIH-NIDCR-funded awards as both Principal Investigator and Co-Investigator, along with prestigious foundation awards including the William Buttler Award, Robert Wood Johnson Foundation, New Jersey Health Foundation, and the CareQuest Institute. Dr. Patel has published over 50 peer-reviewed papers and is the recipient of multiple national honors, including a U.S. patent for innovation in dental diagnostics.

Abstract: The past two decades have witnessed remarkable growth in genetic and genomic studies, generating extensive data resources invaluable for genomic research. Various genomic data sources have been generated and utilized for understanding disease mechanisms and genetic risk management. Despite rapid methodological advances and the growing availability of diverse genomic data resources, existing approaches still face important limitations in scalability, transferability, interpretability, and practical deployment, which constrain their use in real-world biomedical applications. I will introduce some recent work on promoting the use of genetic risk models, including the release of cloud computing tools and pre-trained models across thousands of human traits and diseases. I will then introduce work on pathway-level analysis for causal gene prioritization, leveraging gene regulation information to provide more interpretable and mechanistically detailed insights. Together, these efforts aim to bridge statistical genomics, scalable computational infrastructure, and translational biomedical research, seeking to support more actionable genomic discoveries for precision medicine and disease prevention.

Bio: Dr. Jin Jin is an Assistant Professor of Biostatistics at Penn Medicine. Jin’s methodological research spans a number of areas, including statistical genetics/genomics, disease risk prediction, and multi-modal data integration.

Abstract: Clinical images are often acquired for a specific purpose and interpreted within the technical and analytic constraints of their time. However, many existing scans contain additional quantitative and biologic information that was not previously accessible. Now, advances in artificial intelligence, quantitative image analysis, and retrospective computational methods are enabling the extraction of new measurements and signals from previously acquired images, often without the need for new acquisition.

This talk will examine the broader opportunity to repurpose existing imaging data, particularly routine and legacy MRI, to address new scientific and clinical questions. Rather than viewing older datasets primarily through the limitations of their original acquisition or intended use, they can increasingly be approached as sources of underexploited information that can inform independently or can complement ongoing prospective studies. Practical examples including atrophy measurement and cortical lesion detection will be highlighted as examples of how large, pre-existing imaging datasets can support substantially broader analysis than was envisioned when they were first acquired, creating new opportunities for retrospective discovery, biomarker development, and clinical translation

Bio: My research as director of the Neuroinformatics Development Lab at the Buffalo Neuroimaging Analysis Center focuses on developing and applying quantitative image analysis methods to neuroimaging data in order to characterize better the onset, progression, and treatment of neurological diseases. In particular, magnetic resonance imaging (MRI) can provide a vast amount of raw data about a variety of brain and spinal cord tissue characteristics, but extracting meaningful clinical and research metrics from these data is still challenging. Modern computer science techniques, however, can play a transformative role in helping physicians assess data they receive from neuroimaging techniques in order to deliver the best possible care to their patients.
Highlights of my work include translational approaches to measuring brain atrophy in clinical routine MRI, methods for quantification of longitudinal myelin changes in vivo, more precise algorithms for tracking gray and white matter changes over time, and connectomics research elucidating the functional and structural networks involved in cognitive changes in multiple sclerosis. This work has had a substantial impact on our understanding of multiple sclerosis (MS) onset and progression, and these techniques have also been successfully applied in clinical trials to understand better the impact of various therapeutic approaches in MS.
My ongoing research in quantitative image analysis is aimed at increasing our understanding of the data available from state-of-the-art neuroimaging. This increased understanding can be directly translated to clinicians to better inform their patient diagnoses and treatment decisions.

Abstract: TBA

Bio: TBA