Professor William M. Wells III
Department of Radiology
Harvard Medical School and Brigham and Women's Hospital
- Title of the talk:
Active Mean Fields for Probabilistic Image Segmentation: Connections with Chan-Vese and Rudin-Osher-Fatemi Models
Segmentation is a fundamental task for extracting semantically meaningful regions from an image. The goal of segmentation algorithms is to accurately assign object labels to each image location. However, image-noise, shortcomings of algorithms, and image ambiguities cause uncertainty in label assignment. Estimating the uncertainty in label assignment is important in multiple application domains, such as segmenting tumors from medical images for radiation treatment planning. One way to estimate these uncertainties is through the computation of posteriors of Bayesian models, which is computationally prohibitive for many practical applications. On the other hand, most computationally efficient methods fail to estimate label uncertainty. We therefore propose in this paper the Active Mean Fields (AMF) approach, a technique based on Bayesian modeling that uses a mean-field approximation to efficiently compute a segmentation and its corresponding uncertainty. Based on a variational formulation, the resulting convex model combines any label-likelihood measure with a prior on the length of the segmentation boundary. A specific implementation of that model is the Chan-Vese segmentation model (CV), in which the binary segmentation task is defined by a Gaussian likelihood and a prior regularizing the length of the segmentation boundary. Furthermore, the Euler-Lagrange equations derived from the AMF model are equivalent to those of the popular Rudin-Osher-Fatemi (ROF) model for image denoising. Solutions to the AMF model can thus be implemented by directly utilizing highly-efficient ROF solvers on log-likelihood ratio fields. We qualitatively assess the approach on synthetic data as well as on real natural and medical images. For a quantitative evaluation, we apply our approach to the icgbench dataset.
- Short biography:
William Wells is Professor of Radiology at Harvard Medical School and Brigham and Women's Hospital (BWH), a research scientist at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), and a member of the affiliated faculty of the Harvard-MIT division of Health Sciences and Technology (HST). He received a Ph.D. in computer vision from MIT in 1992 under the supervision of Professor Grimson, and since that time has pursued research in medical image understanding at the BWH Surgical Planning Laboratory, much of it in collaboration with MIT graduate students. Prof. Wells periodically teaches the medical image processing component of HST-582, Biomedical Signal and Image Processing. He is widely known for his ground-breaking work on segmentation of MRI and for his work on multi-modality registration by maximization of Mutual Information, for which he and Paul Viola received the IEEE ICCV Helmholtz "test of time" award.
My research in medical image registration concerns the use of Mutual Information as a criterion for image fusion. This approach has become the de-facto standard for multi-modality problems. Implementations of this method are available in 3D Slicer, our open-source platform for medical image analysis, and in ITK, an NIH sponsored segmentation and registration library.
In addition to morphological analysis, I am also interested in univariate and multivariate analysis of functional MRI.
I recently organized the 2013 meeting of Information Processing In Medical Imaging (IPMI 2013) , held near Monterey California June 29 - July 3.