Automated segmentation of chronic stroke lesions using LINDA: Lesion identification with neighborhood data analysis

Abstract

The gold standard for identifying stroke lesions is manual tracing, a method that is known to be observer dependent and time consuming, thus impractical for big data studies. We propose LINDA (Lesion Identification with Neighborhood Data Analysis), an automated segmentation algorithm capable of learning the relationship between existing manual segmentations and a single T1-weighted MRI. A dataset of 60 left hemispheric chronic stroke patients is used to build the method and test it with k-fold and leave-one-out procedures. With respect to manual tracings, predicted lesion maps showed a mean dice overlap of 0.696±0.16, Hausdorff distance of 17.9±9.8mm, and average displacement of 2.54±1.38mm. The manual and predicted lesion volumes correlated at r=0.961. An additional dataset of 45 patients was utilized to test LINDA with independent data, achieving high accuracy rates and confirming its cross-institutional applicability. To investigate the cost of moving from manual tracings to automated segmentation, we performed comparative lesion-to-symptom mapping (LSM) on five behavioral scores. Predicted and manual lesions produced similar neuro-cognitive maps, albeit with some discussed discrepancies. Of note, region-wise LSM was more robust to the prediction error than voxel-wise LSM. Our results show that, while several limitations exist, our current results compete with or exceed the state-of-the-art, producing consistent predictions, very low failure rates, and transferable knowledge between labs. This work also establishes a new viewpoint on evaluating automated methods not only with segmentation accuracy but also with brain-behavior relationships. LINDA is made available online with trained models from over 100 patients.

Damage to white matter bottlenecks contributes to language impairments after left hemispheric stroke

“…Previous studies using different task paradigms have implicated damage to the left IFG, pSTG/pMTG, and pSMG/AG, in deficits of auditory comprehension and/or discrimination (Bates et al, 2003, Dronkers et al, 2004, Fridriksson et al, 2013, Geva et al, 2012, Pustina et al, 2016, Yourganov et al, 2016), and have implicated anterior temporal lobe lesions in semantic deficits when controlling for deficits in comprehension (Mirman et al, 2015a, Mirman et al, 2015b, Schwartz et al, 2009, Walker et al, 2011). Importantly, all of these regions were identified by our analysis, and are associated with the semantic network recruited by this task (Binder et al, 1997, Binder et al, 2009).…”

Lesion-symptom mapping in the study of spoken language understanding

“…This approach has the advantage of being objective and quick, which is important when studying large groups of patients. The correspondence between automatically and manually delineated lesions is presently only modest (Wilke, de Haan, Juenher, & Karnath, 2011), but continued advances in automated lesion delineation (Griffis, Allendorfer, & Szaflarski, 2016; Pustina et al, 2016) offer the promise of increasingly robust and valid methods that may in time perform as well if not better than manual delineation (Crinion, Holland, Copland, Thompson, & Hillis, 2013). …”

A multivariate lesion symptom mapping toolbox and examination of lesion‐volume biases and correction methods in lesion‐symptom mapping

“…This voxel-based lesion-symptom mapping (VLSM) technique produces a statistical map showing the strength of the relationship between damage at any given voxel and performance on a behavioral measure of interest across a group of individuals with brain lesions (Bates et al, 2003). Analyses are performed on lesion maps delineated by hand (Ashton et al, 2003) or through automated methods (Pustina et al, 2016) on anatomical scans which are then warped to a standard space. Statistical analysis conceptually involves conducting a between-group comparison at each voxel in the sample of spatially normalized lesion maps, with behavioral score binned by lesion status at that voxel.…”