Diseased Region Detection of Longitudinal Knee Magnetic Resonance Imaging Data

• Activity Number: 655
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Imaging
• Author(s): Chao Huang*+
• Companies: University of North Carolina at Chapel Hill
• Keywords: Diseased regions detection ; Gaussian hidden Markov model ; Longitudinal cartilage thickness ; Pseudo-likelihood method ; EM algorithm

Abstract

Magnetic resonance imaging (MRI) has became an important imaging technique for quantifying the spatial location and magnitude/direction of longitudinal cartilage morphology changes in patients with osteoarthritis (OA). Although several analytical methods, such as subregion-based analysis, have been developed for such quantification, they can be suboptimal due to two major issues including the lack of spatial correspondence across subjects and time and the spatial heterogeneity of cartilage progression across subjects. The aim of this paper is to present an analysis pipeline with two major toolboxs for longitudinal cartilage quantification in OA patients, while addressing these two issues. The first toolbox is to preprocess a 3D knee image data in order to establish spatial correspondence across subjects and/or time. The second toolbox is a Gaussian hidden Markov model (GHMM) for dealing with the spatial heterogeneity of cartilage progression across both time and OA subjects. To estimate unknown parameters in GHMM, we employ a pseudo-likelihood function and optimize it by using an expectation-maximization (EM) algorithm. The proposed model can effectively detect diseased regions in