Measuring Brain Lesion Progression with a Supervised Tissue Classification System

摘要: Brain lesions, especially White Matter Lesions (WMLs), are associated with cardiac and vascular disease, but also normal aging. Quantitative analysis of WML in large clinical trials is becoming more important. In this paper, we present a computer-assisted segmentation method, based on local features extracted from conventional multi-parametric Magnetic Resonance Imaging (MRI) sequences. A framework for preprocessing the temporal data by jointly equalizing histograms reduces spatial variance data, thereby improving longitudinal stability such measurements hence estimate lesion progression. Support Vector Machine (SVM) classifier trained expert-defined WML's applied each scan using AdaBoost algorithm. Validation population 23 patients 3 different imaging sites follow-up studies WMLs varying sizes, shapes locations tests robustness accuracy proposed compared to manual results an experienced neuroradiologist. The show that our CAD-system achieves consistent 4D facilitating disease monitoring.