The Ohio State University
Abstract: This talk will cover our recent works on subgraph mining and factorization models applied to clinical narrative text, ICU physiologic time series and computational genomics. The common theme of these works aims at building clinical models that improve both prediction accuracy and interpretability, by exploring relational information in each data modality.
The talk will focus on three concrete examples, implicating neurodevelopmentally coregulated exon clusters in phenotypes of Autism Spectrum Disorder (ASD), predicting mortality risk of ICU patients based on their physiologic measurement time series, and identifying subtypes of lymphoma patients based on pathology report text. In each example, I will show how to automatically extract relational information into a graph representation and how to collect important subgraphs that are of interest. Depending on the degree of structure in the data format, heavier machinery of factorization models becomes necessary to reliably group important subgraphs. I will demonstrate that these methods lead to not only improved performance but also better interpretability in each application.
About the Speaker: Yuan Luo is a PhD Candidate at MIT Computer Science and Artificial Intelligence Lab. His research interests include machine learning, natural language processing, time series analysis, computational genomics and big data analytics, with a focus on medical applications. He proposed Subgraph Augmented Non-negative Tensor Factorization (SANTF) for building a clinical model that improves both accuracy and interpretability, by turning narrative text into graph representations and applying tensor factorization to mining graph features. This work was awarded the first prize at NLP Doctoral Consortium in 2013 Annual Symposium of the American Medical Informatics Association. He has extended the subgraph mining and factorization models to time series analysis and computational genomics. He is also a member of the Student Editorial Board for Journal of the American Medical Informatics Association.
When: Monday April 13, 1:00 - 2:00 PM
Where: 245 Lincoln Tower
Lunch will be provided
“I’m grateful for my education and all that OSU has given me,” says Michael J. Patzakis, MD ’63, who chose to say “thank you” to the College of Medicine by establishing the Michael J. Patzakis MD Endowed Lectureship in Orthopaedics at Ohio State, expected to launch in autumn 2015.Dr. Patzakis, says he values the importance of giving back to the “birthplace of his medical career.” He and his wife Susan, a fellow OSU graduate, live in Los Angeles, where he retired after 21 ½ years as Chairman of the Department of Orthopaedic Surgery at Keck School of Medicine, University of Southern California. His time at Ohio State led him to meet his wife, serve as president of the pre-medical honor society, Alpha Epsilon Delta, and launch a lifetime career in medicine. The Ohio native has led the largest orthopaedics department in the country, taught as visiting professor at 60 medical schools and hospitals, issued 65 research studies, authored 170 scientific papers, mentored more than 400 residents, and has received awards and honors from across the country and internationally for his impact on the profession. Because of this success, Ohio State remains close to his heart. “If others have the same feelings, I urge them to find a way to ‘pay it forward’ to support not only the College, but future physicians.”