Multivariate lesion-symptom mapping using support vector regression

• Published in: Human brain mapping 2014-12, Vol.35 (12), p.5861-5876
• Main Authors: Zhang, Yongsheng, Kimberg, Daniel Y., Coslett, H. Branch, Schwartz, Myrna F., Wang, Ze
• Format: Article
• Language: English
• Subjects: Adult; Aged; Algorithms; aphasia; Aphasia - etiology; Aphasia - physiopathology; Biological and medical sciences; Brain - physiopathology; Brain Mapping - methods; Computer Simulation; Feasibility Studies; Female; Humans; Investigative techniques, diagnostic techniques (general aspects); lesion-symptom mapping; Male; Medical sciences; Middle Aged; Models, Neurological; Multivariate Analysis; Nervous system; Neurology; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Regression Analysis; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Stroke - complications; Stroke - physiopathology; Support Vector Machine; support vector regression; total lesion volume control; Vascular diseases and vascular malformations of the nervous system
• ISSN: 1065-9471; 1097-0193
• DOI: 10.1002/hbm.22590

Lesion analysis is a classic approach to study brain functions. Because brain function is a result of coherent activations of a collection of functionally related voxels, lesion‐symptom relations are generally contributed by multiple voxels simultaneously. Although voxel‐based lesion‐symptom mapping (VLSM) has made substantial contributions to the understanding of brain‐behavior relationships, a better understanding of the brain‐behavior relationship contributed by multiple brain regions needs a multivariate lesion‐symptom mapping (MLSM). The purpose of this artilce was to develop an MLSM using a machine learning‐based multivariate regression algorithm: support vector regression (SVR). In the proposed SVR‐LSM, the symptom relation to the entire lesion map as opposed to each isolated voxel is modeled using a nonlinear function, so the intervoxel correlations are intrinsically considered, resulting in a potentially more sensitive way to examine lesion‐symptom relationships. To explore the relative merits of VLSM and SVR‐LSM we used both approaches in the analysis of a synthetic dataset. SVR‐LSM showed much higher sensitivity and specificity for detecting the synthetic lesion‐behavior relations than VLSM. When applied to lesion data and language measures from patients with brain damages, SVR‐LSM reproduced the essential pattern of previous findings identified by VLSM and showed higher sensitivity than VLSM for identifying the lesion‐behavior relations. Our data also showed the possibility of using lesion data to predict continuous behavior scores. Hum Brain Mapp 35:5861–5876, 2014. © 2014 Wiley Periodicals, Inc.