|
Vol: 55(69) No: 3 / September 2010 A New Cerebral Anatomical-Based Automated Active Segmentation Method Anda Sabau Computer Science Department, “Politehnica” University of Timisoara, Faculty of Automation and Computers, 2 V. Parvan Str. 300223, Timisoara, Romania, e-mail: anda_sab@yahoo.com Roxana Oana Teodorescu Université de Franche-Comté, Besançon, France, e-mail: teodorescu@cs.upt.ro Vladimir Ioan Cretu Computer Science Department, “Politehnica” University of Timisoara, Faculty of Automation and Computers, 2 V. Parvan Str. 300223, Timisoara, Romania, e-mail: vladimir.cretu@cs.upt.ro Keywords: automatic ROI/VOI detection, active contour segmentation, medical image processing Abstract We propose a new method for segmentation based on an active automatic detection approach. We test this method on the putamen anatomical area and we include it as part of an automatic prognosis detection system prototype, PDFibAtl@s. This segmentation method is based on shape constraints that reside at the anatomical structure level. We develop the algorithm on diffusion MRI images, using it further for a deterministic global tractography. Its importance resides at the neural fiber validation and the alignment of the segmented volume augments the accuracy of the tractography. A total of 143 subjects, part of an extended Parkinson\'s Disease (PD) database, with 68 patients clinically diagnosed with PD and 75 control cases, are used for developing this new method. Our method employs a geometrical approach that updates a classic active contour tracking method by eliminating the inter-patient variability. From the technical point of view we are providing an automated tool, with versatility, surmounting not only the demographic variability, but also the anatomical specificity. The medical importance of our approach resides in the possibility to provide a specific anatomical volume for further study and applicability on other cerebral anatomical areas. References R. O. Teodorescu, “Parkinson’s disease prognosis using diffusion tensor imaging features fusion (pronostic de la maladie de Parkinson base sur la fusion des caractristiques d’images par resonance magnetique de diffusion),” Ph.D. dissertation,”Politehnica University of Timisoara, Romania Universite de Franche-Comte, Besancon, France, October 2010. [2] J. M. F. Milan Sonka, Ed., Handbook of Medical Imaging, 3rd ed., ser. Diagnostic Imaging - Handbooks. P.O. box 10 Bellingham, Washington 98227-0010 USA: SPIE Press, 2009, vol. 2 Medical Image Processing and Analysis, ISBN 0-8194-3621-6. [Online]. Available: http://link.aip.org/link/doi/ 10.1117/3.831079 [3] L.-L. Chan, H. Rumperl, and K. Yap, “Case control study of diffusion tensor imaging in Parkinson’s disease,” J. Neurol. Neurosurg. Psychiatry, vol. 78, pp. 1383–1386, 2007. [4] H.-J. Kretschmann and W. Weinrich, Cranial Neuroimaging and Clinical Neuroanatomy - 3rd edition. Thieme Classics, 2003. [5] R. Teodorescu, D. Racoceanu, L. Chan, K. Lovblad, and H. Muller, “Parkinson’s disease detection using 3d brain MRI FA map histograms correlated with tract directions,” RSNA, vol. 8015681, p. 1, 2009, Chicago, IL USA. [6] D. L. Bihan, J.-F. Mangin, C. Pupon, and C. A. Clark, “Diffusion tensor imaging: concepts and application,” Journal of Magnetic Resonance Imaging, vol. 13, pp. 534–546, 2001. [7] Rorden, “DTI analysis,” Workshop Report, October 2008, last accessed on May 2010. [Online]. Available: http: //www.sph.sc.edu/comd/rorden/workshop/fsl/dti/ [8] R. Teodorescu, “H&Y compliant for PD diagnosis and prognosis using EPI and FA images,” Politehnica University of Timisoara, PhD Report No. 2, February 2010. [9] C. Gaser, “Structural brain mapping group,” presentation web-page, December 2008, software presentation and download site. http://dbm.neuro.uni-jena.de/vbm/vbm5-for-spm5/ - last accessed on March 2010. [Online]. Available: http://dbm. neuro.uni-jena.de/vbm/vbm5-for-spm5/ [10] Guillaume, “Spm documentation,” pdf technical report, Trust Center for Neuroimaging, www.fil.ion.ucl.ac.uk, October 2008, trust Center for Neuroimaging. http://www.fil.ion.ucl.ac.uk/spm/doc/ - last accessed on May 2010. [Online]. Available: http://www.fil.ion.ucl.ac.uk/spm/doc/ [11] A. Feldmann, Z. Illes, and et al., “Morphometric changes of gray matter in Parkinson’s disease with depression: A voxel based morphometry study,” Movement Disorders, vol. 23, pp. 42–46, 2008. [12] M. Dijan, “Wfu pickatlas - user manual,” The Functional MRI Laboratory Wake Forest University School of Medicine, Tech. Rep., 2004. [13] I. F. Talos, “Slicer 3 tutorial the spl pnl brain atlas,” Brigham and Women’s Hospital, presentation, March 2003, surgical Planning Laboratory. [Online]. Available: http://www.slicer.org [14] D. T. Michael Kass, Andrew Witkin, “Snakes: Active contour models,” International Journal of the Computer Vision, 1987. [15] R. Teodorescu, D. Racoceanu, N. Smit, V. I. Cretu, E. K. Tan, and L.-L. Chan, “Parkinson’s disease prediction using diffusion based atlas,” 13-18 Feb. 2010, SPIE Medical Imaging, Computer Aided Diagnosis, San Diego, California, USA. [16] A. Sabau, R. O. Teodorescu, and V. I. Cretu, “Automatic putamen detection on DTI images. Application to Parkinson’s disease,” in Proc. ICCC-CONTI 2010, Timisoara, 2010. |