Spine labeling in axial magnetic resonance imaging via integral kernels. (December 2016)

• Record Type: Journal Article
• Authors: Miles, Brandon; Ben Ayed, Ismail; Hojjat, Seyed-Parsa; Wang, Michael H.; Li, Shuo; Fenster, Aaron; Garvin, Gregory J.
• Is Part Of: Computerized medical imaging and graphics. Volume 54(2016)
• Page Start: 27
• Page End: 34
• Publication Date: 2016-12
• Subjects: Spine labeling -- Magnetic resonance imaging -- Integral kernels -- Geometric priors -- Probability product kernels; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2016.09.004
• Languages: English
• ISSNs: 0895-6111

Abstract

Abstract : Highlights: Novel integral-kernel approach for classifying (labeling) spine structures in axial magnetic resonance images (MRI). The method embeds geometric priors based on anatomical measurements of the spine. Very competitive performance (using a data set of 32 patients) in nearly real-time, along with a graphic processing unit (GPU) implementation. Abstract: This study investigates a fast integral-kernel algorithm for classifying (labeling) the vertebra and disc structures in axial magnetic resonance images (MRI). The method is based on a hierarchy of feature levels, where pixel classifications via non-linear probability product kernels (PPKs) are followed by classifications of 2D slices, individual 3D structures and groups of 3D structures. The algorithm further embeds geometric priors based on anatomical measurements of the spine. Our classifier requires evaluations of computationally expensive integrals at each pixel, and direct evaluations of such integrals would be prohibitively time consuming. We propose an efficient computation of kernel density estimates and PPK evaluations for large images and arbitrary local window sizes via integral kernels . Our method requires a single user click for a whole 3D MRI volume, runs nearly in real-time, and does not require an intensive external training. Comprehensive evaluations over T1-weighted axial lumbar spine data sets from 32 patients demonstrate a competitive structure classification accuracy of 99%, along with a 2D slice classification accuracy of 88%. To the best of our knowledge, such a structure classification accuracy has not been reached by the existing spine labeling algorithms. Furthermore, we believe our work is the first to use integral kernels in the context of medical images.