Gautam Prasad1, Anand A Joshi2, Albert Feng3, Arthur W Toga4, Paul M Thompson5, Demetri Terzopoulos6. 1. Imaging Genetics Center & Laboratory of Neuro Imaging, Institute for Neuroimaging and Informatics, Keck School of Medicine of USC, Los Angeles, CA, USA; Department of Psychology, Stanford University, Stanford, CA, USA. Electronic address: gprasad@usc.edu. 2. Signal and Image Processing Institute, USC, Los Angeles, CA, USA. 3. Imaging Genetics Center & Laboratory of Neuro Imaging, Institute for Neuroimaging and Informatics, Keck School of Medicine of USC, Los Angeles, CA, USA. 4. Imaging Genetics Center & Laboratory of Neuro Imaging, Institute for Neuroimaging and Informatics, Keck School of Medicine of USC, Los Angeles, CA, USA; Department of Ophthalmology, Neurology, Psychiatry & Behavioral Sciences, Radiology, and Biomedical Engineering, Keck School of Medicine of USC, Los Angeles, CA, USA. 5. Imaging Genetics Center & Laboratory of Neuro Imaging, Institute for Neuroimaging and Informatics, Keck School of Medicine of USC, Los Angeles, CA, USA; Department of Ophthalmology, Neurology, Psychiatry & Behavioral Sciences, Radiology, and Biomedical Engineering, Keck School of Medicine of USC, Los Angeles, CA, USA; Department of Pediatrics, Keck School of Medicine of USC, Los Angeles, CA, USA. 6. Department of Computer Science, UCLA, Los Angeles, CA, USA.
Abstract
BACKGROUND: Segmentation methods for medical images may not generalize well to new data sets or new tasks, hampering their utility. We attempt to remedy these issues using deformable organisms to create an easily customizable segmentation plan. We validate our framework by creating a plan to locate the brain in 3D magnetic resonance images of the head (skull-stripping). NEW METHOD: Our method borrows ideas from artificial life to govern a set of deformable models. We use control processes such as sensing, proactive planning, reactive behavior, and knowledge representation to segment an image. The image may have landmarks and features specific to that dataset; these may be easily incorporated into the plan. In addition, we use a machine learning method to make our segmentation more accurate. RESULTS: Our method had the least Hausdorff distance error, but included slightly less brain voxels (false negatives). It also had the lowest false positive error and performed on par to skull-stripping specific method on other metrics. COMPARISON WITH EXISTING METHOD(S): We tested our method on 838 T1-weighted images, evaluating results using distance and overlap error metrics based on expert gold standard segmentations. We evaluated the results before and after the learning step to quantify its benefit; we also compare our results to three other widely used methods: BSE, BET, and the Hybrid Watershed algorithm. CONCLUSIONS: Our framework captures diverse categories of information needed for brain segmentation and will provide a foundation for tackling a wealth of segmentation problems.
BACKGROUND: Segmentation methods for medical images may not generalize well to new data sets or new tasks, hampering their utility. We attempt to remedy these issues using deformable organisms to create an easily customizable segmentation plan. We validate our framework by creating a plan to locate the brain in 3D magnetic resonance images of the head (skull-stripping). NEW METHOD: Our method borrows ideas from artificial life to govern a set of deformable models. We use control processes such as sensing, proactive planning, reactive behavior, and knowledge representation to segment an image. The image may have landmarks and features specific to that dataset; these may be easily incorporated into the plan. In addition, we use a machine learning method to make our segmentation more accurate. RESULTS: Our method had the least Hausdorff distance error, but included slightly less brain voxels (false negatives). It also had the lowest false positive error and performed on par to skull-stripping specific method on other metrics. COMPARISON WITH EXISTING METHOD(S): We tested our method on 838 T1-weighted images, evaluating results using distance and overlap error metrics based on expert gold standard segmentations. We evaluated the results before and after the learning step to quantify its benefit; we also compare our results to three other widely used methods: BSE, BET, and the Hybrid Watershed algorithm. CONCLUSIONS: Our framework captures diverse categories of information needed for brain segmentation and will provide a foundation for tackling a wealth of segmentation problems.
Authors: P M Thompson; T D Cannon; K L Narr; T van Erp; V P Poutanen; M Huttunen; J Lönnqvist; C G Standertskjöld-Nordenstam; J Kaprio; M Khaledy; R Dail; C I Zoumalan; A W Toga Journal: Nat Neurosci Date: 2001-12 Impact factor: 24.884
Authors: Susan M Resnick; Dzung L Pham; Michael A Kraut; Alan B Zonderman; Christos Davatzikos Journal: J Neurosci Date: 2003-04-15 Impact factor: 6.167
Authors: Clifford R Jack; Matt A Bernstein; Bret J Borowski; Jeffrey L Gunter; Nick C Fox; Paul M Thompson; Norbert Schuff; Gunnar Krueger; Ronald J Killiany; Charles S Decarli; Anders M Dale; Owen W Carmichael; Duygu Tosun; Michael W Weiner Journal: Alzheimers Dement Date: 2010-05 Impact factor: 21.566
Authors: David W Shattuck; Mubeena Mirza; Vitria Adisetiyo; Cornelius Hojatkashani; Georges Salamon; Katherine L Narr; Russell A Poldrack; Robert M Bilder; Arthur W Toga Journal: Neuroimage Date: 2007-11-26 Impact factor: 6.556
Authors: Hongzhi Wang; Sandhitsu R Das; Jung Wook Suh; Murat Altinay; John Pluta; Caryne Craige; Brian Avants; Paul A Yushkevich Journal: Neuroimage Date: 2011-01-13 Impact factor: 6.556
Authors: Gautam Prasad; Shantanu H Joshi; Neda Jahanshad; Julio Villalon-Reina; Iman Aganj; Christophe Lenglet; Guillermo Sapiro; Katie L McMahon; Greig I de Zubicaray; Nicholas G Martin; Margaret J Wright; Arthur W Toga; Paul M Thompson Journal: Neuroimage Date: 2014-04-18 Impact factor: 6.556