More about HKUST
Pre- and intra-operative image processing and visualization
Speaker: Prof. William Wells Harvard Medical School, USA Title: Pre- and intra-operative image processing and visualization Date: Tuesday, 17 Feb 2004 Time: 2:00pm - 3:00pm Venure: Room 2404 (Phase I, via lift nos. 17/18) HKUST ABSTRACT: In this talk I will summarize pre-and intra-operative image processing and visualization that are part of a research program in image-guided neurosurgery at Brigham and Women's Hospital. This program is focused on procedures in a .5 Tesla interventional MRI. A central aspect of this work is the "3D Slicer", a freely available open source platform for medical image processing and visualization. After describing these application projects, I will review some recent developments in statistical and information-theoretic image registration. The Mutual Information approach and related model-based methods will be described and contrasted, with the following observation: If you can't move towards the known right thing, then you can at least move away from the most bad thing. In addition, recent experimental results will be described. BIOGRAPHY: Prof. William M. Wells is a well-known researcher in medical image processing. He is a Associate Professor of Radiology at the Harvard Medical School, and Brigham and Women's Hospital. Prof. Wells is also an Affiliated Faculty of the MIT Division of Health Sciences and Technology in Harvard and the Research Scientist of MIT AI Laboratory. Prof. Wells earned his doctoral degree in 1992 during which time he worked with Prof. Eric Grimson on research about Computerized Object Recognition. He joined the surgical planning laboratory at BWH in 1992 as postdoc. Since that time he has worked mostly on MRI segmentation and multi-modality registration, primarily with methods of statistical image processing. He maintains a vigorous collaboration with the MIT AI laboratory faculty and students.