Center for Biomedical Image Computing and Analytics (CBICA) Seminar
Department of Radiology
University of Pennsylvania
New Frontiers in the Application of Machine Learning to Ultrasound
While there are many applications of traditional machine learning and deep learning to x-rays, CT, and MRI, projects with ultra-sound are comparatively less well explored. This talk will discuss the use of traditional machine learning and deep learning techniques as applied to ultrasound imaging. Applications will include automated hepatic fat quantification, ultrasound elastography, and contrast ultrasound for the early diagnosis of cancer. This talk will also delve into the unique challenges involved in working with ultrasound data for machine learning, as well as the exciting opportunities that this modality offers.
Bio: Dr. Hersh Sagreiya is a newly-appointed Assistant Professor in the Department of Radiology at the University of Pennsylvania. He recently completed the Stanford Cancer Imaging Training Program, a combined clinical and research fellowship, with his research funded by the RSNA, NVIDIA, and the Stanford Society of Physician Scholars. He is collaborating with scientists and engineers on projects that apply machine learning techniques to medical imaging. He is currently working on research projects funded by Yahoo and GE Health, on topics including anomaly detection in medical imaging and automated diagnosis using deep learning and liver ultrasound imaging. Dr. Sagreiya obtained his AB in Biochemical Sciences from Harvard and his MD from Stanford.
* Pizza lunch at 12:45pm – talk at 1pm*