Skip to main content
Fig. 3 | Journal of Big Data

Fig. 3

From: DFA-Net: Dual multi-scale feature aggregation network for vessel segmentation in X-ray digital subtraction angiography

Fig. 3

For a representative DSA image (a), segmented results obtained from U-net (b), Deeplabv3+ (c), U-net++ (d), SA-Unet (e), U-net 3+ (f), CMU-Net (g), CA-Net (h), and DFA-Net (i), respectively, and ground truth (j), in which white rectangles denote ROIs (e.g., tiny sub-branches), yellow ellipses indicate false detections, and yellow rectangles indicate capillary vessels that are not manually labelled in ground truth

Back to article page