FastSurfer Parcellation Accuracy After Lesion Filling in Moderate-to-Severe Traumatic Brain Injury

EVELYN DEUTSCHER^1,2*Emily L Dennis^3,4Frank Hillary^5,6Elisabeth Wilde^4,7,8Carrie Esopenko⁹Ekaterina Dobryakova^10,11Andrei Irimia^12,13,14Ahmed Radwan¹⁵Phoebe Imms¹²Adam Clemente¹⁶Paul Beech¹⁷Alex Burmester¹⁸Karen Caeyenberghs¹⁸Juan F Domínguez¹⁸

• ¹Deakin University, Geelong, Australia
• ²Deakin University, School of Psychology, Burwood, Australia
• ³TBI and Concussion Center, Department of Neurology, University of Utah, Salt Lake City, United States
• ⁴George E. Wahlen Veterans Affairs Salt Lake City Healthcare System,, Salt Lake City, United States
• ⁵The Pennsylvania State University Department of Psychology, University Park, United States
• ⁶Penn State Health Department of Neurology, Hershey, United States
• ⁷TBI and Concussion Center, Department of Neurology, University of Utah,, Salt Lake City, United States
• ⁸H. Ben Taub Department of Physical Medicine and Rehabilitation, Baylor College of Medicine, Houston, United States
• ⁹Icahn School of Medicine at Mount Sinai, New York, United States
• ¹⁰Center for Traumatic Brain Injury, Kessler Foundation,, East Hanover, United States
• ¹¹Rutgers New Jersey Medical School, Newark, United States
• ¹²University of Southern California Leonard Davis School of Gerontology, Los Angeles, United States
• ¹³Alfred E. Mann Department of Biomedical Engineering, Andrew & Erna Viterbi School of Engineering, University of Southern California,, Los Angeles, United States
• ¹⁴University of Southern California Dana and David Dornsife College of Letters Arts and Sciences, Los Angeles, United States
• ¹⁵Katholieke Universiteit Leuven Departement Beeldvorming & Pathologie, Leuven, Belgium
• ¹⁶School of Behavioural and Health Sciences, Faculty of Health Sciences, Australian Catholic University, Melbourne, Australia
• ¹⁷Alfred Health Radiology Department, Melbourne, Australia
• ¹⁸Cognitive Neuroscience Unit, School of Psychology, Deakin University, Burwood, Australia

Objective: Focal lesions in T1-weighted (T1-w) Magnetic Resonance Images (MRI) of patients with moderate-to-severe Traumatic Brain Injury (ms-TBI) can introduce errors during image processing. We tested whether errors in FastSurfer cortical parcellation could be reduced using lesion filling (Virtual Brain Grafting (VBG)). Methods: T1-w MRIs from 140 healthy controls and 14 ms-TBI patients were shared within ENIGMA's TBI working group. A "ground truth" set of 140 lesion-free images were created by registering 10 HCs onto each of 14 ms-TBI images. Masks indexing focal lesions (small [38mm3] unilateral, to large [164,291mm3] bilateral) were projected onto lesion-free images, creating 140 synthetically lesioned images. Lesioned images underwent VBG filling to replace lesioned regions with simulated healthy brain tissue, creating 140 VBG-filled images. To calculate parcellation accuracy, paired sample t-tests of Mean Dice Similarity Coefficients (DSC) and Percent Volume Differences (PVD) for lesioned, and VBG-filled images, were compared to lesion-free images. Results: Parcellations from lesioned images (DSC M = 0.93, SD = 0.03; PVD M = - 0.40, SD = 1.7) unexpectedly had significantly higher DSCs [t(111) = 19.5, p < 0.001], and lower PVDs [t(111) = 11.3, p < 0.001] than VBG-filled images (DSC M = 0.81, SD = 0.07; PVD M = -9.03, SD = 7.72). Interpretation: Parcellations from lesioned images were more accurate (than VBG-filled images) when compared to lesion-free ground truth images. While likely due to a high frequency of smaller focal lesions in our sample, these results could suggest FastSurfer parcellation may be robust in the presence of such lesions