Answers are grounded in this paper and cite their sources.
We evaluated the performance of the brain segmentation tool SynthSeg on a heterogeneous multi-center cohort of pediatric patients with acute disseminated encephalomyelitis. Brain MRI of the patient cohort was acquired at multiple field strengths, with a diverse range of image spatial resolutions, MR sequences, contrasts, and w/wo administration of gadolinium-based contrast agent (GBCA). SynthSeg performance was generally robust to resolution and contrast differences, especially for larger brain structures. Significant differences in some calculated volumes were detected upon pair-wise comparisons among patients with high- vs. low resolution scans, T1- vs. T2-weighed scans, and w/wo GBCA application, acquired on the same day.
No takes yet. Share an insight, caveat, or question.
Pilgrim et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: