Paul Yushkevich Talk:
Leveraging multi-atlas methodology to build better imaging biomarkers of neurodegenerative disease
Dr. Paul Yushkevich will be giving a presentation this Thursday, June 27th at 3pm in the CIND Conference Room (SFVAMC).
Paul A. Yushkevich, Ph.D.
Associate Professor
Penn Image Computing and Science Laboratory
Department of Radiology
University of Pennsylvania
Title: Leveraging multi-atlas methodology to build better imaging biomarkers of neurodegenerative disease
Abstract:
My talk will summarize our recent work on multi-atlas label fusion, a class of medical image segmentation techniques that has delivered excellent accuracy across a range of problem domains. In particular, I will describe the techniques of joint label fusion and corrective learning, developed by Hongzhi Wang, which achieved the best accuracy in last year's multi-atlas segmentation challenge at MICCAI. I will also touch on more recent work extending label fusion concepts to the problems of groupwise correspondence and tumor segmentation. These methodological developments will be presented in the context of our group's work to develop more detailed quantitative measures of hippocampal and extrahippocampal atrophy in Alzheimer's disease and related disorders.
Please come and enjoy Dr. Yushkevich’s talk!
Leveraging multi-atlas methodology to build better imaging biomarkers of neurodegenerative disease
Dr. Paul Yushkevich will be giving a presentation this Thursday, June 27th at 3pm in the CIND Conference Room (SFVAMC).
Paul A. Yushkevich, Ph.D.
Associate Professor
Penn Image Computing and Science Laboratory
Department of Radiology
University of Pennsylvania
Title: Leveraging multi-atlas methodology to build better imaging biomarkers of neurodegenerative disease
Abstract:
My talk will summarize our recent work on multi-atlas label fusion, a class of medical image segmentation techniques that has delivered excellent accuracy across a range of problem domains. In particular, I will describe the techniques of joint label fusion and corrective learning, developed by Hongzhi Wang, which achieved the best accuracy in last year's multi-atlas segmentation challenge at MICCAI. I will also touch on more recent work extending label fusion concepts to the problems of groupwise correspondence and tumor segmentation. These methodological developments will be presented in the context of our group's work to develop more detailed quantitative measures of hippocampal and extrahippocampal atrophy in Alzheimer's disease and related disorders.
Please come and enjoy Dr. Yushkevich’s talk!