Advancing Precision Medicine through Machine Learning in Computational Pathology
Saeed Hassanpour
April 5, 2024, Friday, 2:00 PM - 3:00 PM EDT
Abstract
With the recent expansions of whole-slide digital scanning, archiving, and high-throughput tissue banks, the field of digital pathology is primed to benefit dramatically from deep learning technology. This talk will cover several clinical applications of deep learning for characterizing histopathological patterns on high-resolution microscopy images for the classification, prognosis, and treatment of cancerous and precancerous lesions. Furthermore, the current practical challenges of building deep learning models for pathology image analysis will be discussed and new methodological advances to address these bottlenecks will be presented. The internal and external evaluation results show these approaches' high accuracy and generalizability across different cancer and lesion types, data sources, and slide types. These results demonstrate these novel methods could have a significant impact on the current standard of patient care by assisting pathologists in clinical practice, improving access, efficiency, and accuracy of the histopathological interpretation, and providing a framework for integrating histopathological features with other clinical, imaging, and molecular information for the comprehensive modeling of patient outcomes and promoting precision medicine.
Bio
Dr. Saeed Hassanpour is the Founding Director of Dartmouth Center for Precision Health & Artificial Intelligence (CPHAI) and an Associate Professor in the Departments of Biomedical Data Science, Computer Science, and Epidemiology at Dartmouth. His research is focused on building novel machine learning and multimodal data analysis methods to inform precision health, and his lab has been a pioneer in advancing digital pathology through deep learning methodologies. Dr. Hassanpour has led multiple NIH-funded research projects on developing new machine learning models for medical image analysis and clinical text mining to improve diagnosis, prognosis, and personalized therapies. His research has resulted in numerous publications, software, and datasets that are widely recognized and have received multiple awards, including the 2019 Agilent Early Career Professor Award for breakthroughs in digital pathology. Dr. Hassanpour currently serves as a standing member on multiple national and international scientific panels and committees, such as NIH's Clinical Data Management and Analysis (CDMA) Study Section and the JAMIA Open Editorial Board. Before joining Dartmouth, he worked as a Research Engineer at Microsoft. Dr. Hassanpour received his PhD in Electrical Engineering with a minor in Biomedical Informatics from Stanford University and completed his postdoctoral training at the Stanford Center for Artificial Intelligence in Medicine and Imaging.