UI-VISA: U-Net Initialized Vascular Image Segmentation Architecture
2026-09-01 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the challenge of accurately outlining blood vessels in special X-ray images called digital subtraction angiography (DSA), where vessels are thin and branched. They created a new method called UI-VISA that first uses a type of AI (U-Net) to roughly find vessels, then improves these guesses by growing connected regions step-by-step to make the outlines more continuous and complete. When they tested UI-VISA against just using U-Net or older methods, UI-VISA performed better in keeping the vessels connected, which is important for understanding their shape. Their tests showed that this improvement in connectivity was statistically significant.
Digital Subtraction Angiography (DSA)Vascular SegmentationU-NetRegion Growing AlgorithmCNN (Convolutional Neural Network)Dice ScoreclDice ScoreWilcoxon Signed-Rank TestVascular ConnectivityImage Segmentation
Authors
Asees Kaur, Suzanne S. Sindi, Erica M. Rutter
Abstract
Accurate segmentation of vascular structures in digital subtraction angiography (DSA) images remains challenging due to the thin, elongated, and branching nature of blood vessels. Pixel-wise deep learning approaches such as U-Net achieve strong general-purpose segmentation performance but often produce fragmented or discontinuous predictions in fine vascular regions, since they do not explicitly enforce structural connectivity. Region growing algorithms preserve spatial context and topological continuity, but are highly sensitive to seed point initialization and can be computationally expensive. We propose UI-VISA (U-Net Initialized Vascular Image Segmentation Architecture), a hybrid pipeline that combines the complementary strengths of both approaches. UI-VISA uses U-Net's foreground predictions as informed seed points for a CNN-guided region growing algorithm, which then iteratively refines the segmentation by enforcing local connectivity and recovering fine vessel details that U-Net alone tends to miss or over-predict. We evaluate UI-VISA against standalone U-Net and a prior region-growing-based method (VISA) using 5-fold cross-validation on 26 DSA images. UI-VISA achieves the highest mean Dice and clDice scores across folds, and a paired Wilcoxon signed-rank test shows the improvement in clDice is statistically significant ($p=0.023$), consistent with the method's design goal of preserving vascular connectivity, while the improvement in Dice does not reach significance ($p=0.104$).