Synthetic Data for Robust Stroke Segmentation

arXiv:2404.01946v1 Announce Type: cross
Abstract: Deep learning-based semantic segmentation in neuroimaging currently requires high-resolution scans and extensive annotated datasets, posing significant barriers to clinical applicability. We present a novel synthetic framework for the task of lesion segmentation, extending the capabilities of the established SynthSeg approach to accommodate large heterogeneous pathologies with lesion-specific augmentation strategies. Our method trains deep learning models, demonstrated here with the UNet architecture, using label maps derived from healthy and stroke datasets, facilitating the segmentation of both …

arxiv cs.cv cs.lg data eess.iv robust segmentation stroke synthetic synthetic data type