Journal
ACADEMIC RADIOLOGY
Volume 15, Issue 3, Pages 300-313Publisher
ELSEVIER SCIENCE INC
DOI: 10.1016/j.acra.2007.10.012
Keywords
white matter lesion segmentation; support vector machine; machine learning
Funding
- Intramural NIH HHS Funding Source: Medline
- NHLBI NIH HHS [N01HC95178, N01 HC095178, N01-HC-95178] Funding Source: Medline
- NIA NIH HHS [R01 AG014971-05, R01 AG014971, R01-AG-1497] Funding Source: Medline
- NICHD NIH HHS [P01 HD024448-100006] Funding Source: Medline
Ask authors/readers for more resources
Rationale and Objectives. Brain lesions, especially white matter lesions (WMLs), are associated with cardiac and vascular disease, but also with normal aging. Quantitative analysis of WML in large clinical trials is becoming more and more important. Materials and Methods. In this article, we present a computer-assisted WML segmentation method, based on local features extracted from multiparametric magnetic resonance imaging (MRI) sequences (ie, T1-weighted, T2-weighted, proton density-weighted, and fluid attenuation inversion recovery MRI scans). A support vector machine classifier is first trained on expert-defined WMLs, and is then used to classify new scans. Results. Postprocessing analysis further reduces false positives by using anatomic knowledge and measures of distance from the training set. Conclusions. Cross-validation on a population of 35 patients from three different imaging sites with WMLs of varying sizes, shapes, and locations tests the robustness and accuracy of the proposed segmentation method, compared with the manual segmentation results from two experienced neuroradiologists.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available