Clinical performance review for 3-D Deep Learning segmentation of stroke infarct from diffusion-weighted images

•Deep Learning (DL) segmented stroke infarct from MR Diffusion-weighted images with high accuracy.•Using two different DL models, we conducted a clinical review to derive the clinical significance of a change in performance.•When the Dice Similarity Coefficient (DSC) increased, clinician agreement with the model segmentations increased.•A DSC increase also increased the inter-rater agreement between clinicians around model performance.