September 09, 2013

Ahmed Saad's PhD Defense

(Seminar: Friday September 6th, 2013 10:00 a.m. TASC1 9408)

Monday September 9th, 2013 10:00 a.m. TASC1 9204 West


We propose novel analysis and visualization techniques for dynamic emission tomography images. First, we propose a semi-supervised, kinetic-modeling-based segmentation technique to identify functional regions of interest. It is an iterative, self-learning algorithm based on uncertainty principles and is designed to be robust to the problems of low signal-to-noise ratio and partial volume effect. Second, we develop an interactive analysis and visualization tool for probabilistic segmentations of medical images. We provide a systematic approach to analyze, interact, and highlight regions of segmentation uncertainty. Finally, we introduce a set of multidimensional transfer function widgets to analyze multivariate probabilistic field data. These widgets furnish the user with contextual information about conformance or deviation from the population statistics. We demonstrate the ability to identify suspicious regions (e.g. tumors) and to correct misclassification results.

Keywords: Uncertainty visualization; medical image analysis; probabilistic segmentation; molecular imaging; PET; SPECT
Ph.D. Examining Committee:
Dr. Ghassan Hamarneh, Senior Supervisor
Dr. Torsten Moller, Supervisor
Dr. Faisal Beg, Internal Examiner
Dr. Christopher Johnson, External Examiner
Dr. Brian Funt, Chair