PT  - JOURNAL ARTICLE
AU  - Daniel Wiesen
AU  - Christoph Sperber
AU  - Grigori Yourganov
AU  - Christopher Rorden
AU  - Hans-Otto Karnath
TI  - Using machine learning-based lesion behavior mapping to identify anatomical networks of cognitive dysfunction: spatial neglect and attention
AID  - 10.1101/556753
DP  - 2019 Jan 01
TA  - bioRxiv
PG  - 556753
4099  - http://biorxiv.org/content/early/2019/02/21/556753.short
4100  - http://biorxiv.org/content/early/2019/02/21/556753.full
AB  - Previous lesion behavior studies primarily used univariate lesion behavior mapping techniques to map the anatomical basis of spatial neglect after right brain damage. These studies led to inconsistent results and lively controversies. Given these inconsistencies, the idea of a widespread network that might underlie spatial orientation and neglect has been pushed forward. In such case, univariate lesion behavior mapping methods might have been inherently limited in uncover the presumed network in a single study due to limited statistical power. By using multivariate lesion-mapping based on support vector regression, we aimed to validate the network hypothesis directly in a large sample of 203 newly recruited right brain damaged patients. In a single analysis, this method identified a network of parietal, temporal, frontal, and subcortical regions, which also included white matter tracts connecting these regions. The results were compared to univariate analyses of the same patient sample using different combinations of lesion volume correction and statistical thresholding. The comparison revealed clear benefits of multivariate lesion behavior mapping in identifying brain networks.AbbreviationsVLBM –Voxel-based lesion behavior mapping;MLBM –Multivariate lesion behavior mapping;SVR –Support vector regression;SVR-LSM –Support vector regression based lesion-symptom mapping;FDR –false discovery rate;FWE –Family Wise Error;dTLVC –Direct Total Lesion Volume Control