Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.
Support Vector Machines Classification of Texture Parameters of White Matter Lesions in Systemic Lupus Erythematosus. Possible Mechanism to Distinguish Between Demyelination and Ischemia.
Lapa1, Aline, Bento1, Mariana P., Rittner1, Letcia, Castellano1, Gabriela, Ruocco2, Heloisa, Damasceno3, Benito, Costallat4, Lilian
State University of Campinas, Campinas, Brazil
Medical Faculty of Jundiai, Campinas, Brazil
Medical Faculty of Jundiai, Campinas, United Kingdom
State University of Campinas, Campinas, United Kingdom
State University of Campinas, São Paulo, Brazil
Texture analysis (TA) is a branch of image processing which seeks to reduce image information by extracting texture descriptors from the image. White matter hyperintensities (WMH) are frequently observed in systemic lupus erythematosus (SLE), however the etiology is still unknown. Ischemic and demyelination have been proposed as possible etiologies. Support vector machines (SVM) are a group of supervised learning methods that can be applied to classification or regression. SVM performs classification by constructing a set of hyperplanes in a high dimensional space that optimally separates the data into different categories. A classification task usually involves separating data into training and testing sets. The goal of SVM is to produce a model based on the training data which predicts the target values of the test data given only the test data attributes. Objective: To produce a training model that accurately differentiates WMH of multiple sclerosis (MS) and stroke from normal white matter. To determine attributes that best characterizes WMH in SLE and to analyze clinical and laboratory features that may differentiate SLE patients with demyelination from ischemic WMH.
TA was applied to axial T2-weighted magnetic resonance images (MRI) of 30 SLE, 30 MS, and 10 stroke patients and 30 normal controls, all age and sex-matched. The TA approach used was based on the Gray Level Co-occurrence Matrices (GLCM). The WMH were manually segmented for each subject, classified in periventricular and subcortical WMH and 256 texture parameters were computed for each lesion. A SVM classifier was developed based on texture features of normal white matter and WMH in MS and stroke patients. The classifier was then used to classify WMH in SLE patients. Nature of the classified WMH, demographic, clinical and laboratory features were included in a regression model to determine which variables could help to predict the nature of WMH in clinical practice.
We achieve a accuracy rate of 0.9 on a database of 97 ROIs to training procedure, and 41 ROIs to testing procedure. Of the 24 periventricular WMH, 18 (75%) were classified as ischemic and 6 (25%) as demyelination. Of 44 subcortical lesions, 26 (59%) were classified as ischemic and 18 (41%) as demyelination. Age (odds ratio [OR] 1.7, 95% confidence interval [95% CI] 1.586.72), hypertension (OR=2.6; 95%CI 1.95.3) and positive antiphospholipid antibodies (aPL) (OR=1.9; 95%CI 1.27.3) were variables associated with stroke, whereas shorter disease duration (OR=3.1; 95%CI 2.27.5) and new onset of neurologic symptoms (OR=1.8; 95%CI 1.23.5) were associated with demyelination.
Although 75% of WMH were classified as ischemic in nature, we identified approximately 25% of demyelinating WMH in SLE. SMV of TA is a useful method to help to determine etiology of WMH in SLE. Age, hypertension and aPL were variables associated with ischemic; shorter disease duration and new onset neurologic symptoms were associated with demyelinating lesions in this cohort.
To cite this abstract, please use the following information:
Lapa, Aline, Bento, Mariana P., Rittner, Letcia, Castellano, Gabriela, Ruocco, Heloisa, Damasceno, Benito, et al; Support Vector Machines Classification of Texture Parameters of White Matter Lesions in Systemic Lupus Erythematosus. Possible Mechanism to Distinguish Between Demyelination and Ischemia. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :2257