DIAS: A Dataset and Benchmark for Intracranial Artery Segmentation in DSA sequences

TL;DR

This work introduces DIAS, a public dataset for intracranial artery segmentation in DSA sequences, and establishes a benchmark for fully-, weakly-, and semi-supervised segmentation on $2D+Time$ data. It proposes three core contributions: VSS-Net for efficient spatiotemporal segmentation of DSA sequences, SSCR for scribble-based weak supervision with cross pseudo supervision and consistency regularization, and RPST for random patch-based self-training to leverage unlabeled data. Extensive experiments on DIAS demonstrate that VSS-Net achieves leading performance among a wide range of baselines, while SSCR and RPST provide robust, annotation-efficient alternatives with competitive accuracy. The dataset and code are publicly available, enabling researchers and clinicians to explore advanced algorithms for IA segmentation and broader vascular analysis in DSA sequences.

Abstract

The automated segmentation of Intracranial Arteries (IA) in Digital Subtraction Angiography (DSA) plays a crucial role in the quantification of vascular morphology, significantly contributing to computer-assisted stroke research and clinical practice. Current research primarily focuses on the segmentation of single-frame DSA using proprietary datasets. However, these methods face challenges due to the inherent limitation of single-frame DSA, which only partially displays vascular contrast, thereby hindering accurate vascular structure representation. In this work, we introduce DIAS, a dataset specifically developed for IA segmentation in DSA sequences. We establish a comprehensive benchmark for evaluating DIAS, covering full, weak, and semi-supervised segmentation methods. Specifically, we propose the vessel sequence segmentation network, in which the sequence feature extraction module effectively captures spatiotemporal representations of intravascular contrast, achieving intracranial artery segmentation in 2D+Time DSA sequences. For weakly-supervised IA segmentation, we propose a novel scribble learning-based image segmentation framework, which, under the guidance of scribble labels, employs cross pseudo-supervision and consistency regularization to improve the performance of the segmentation network. Furthermore, we introduce the random patch-based self-training framework, aimed at alleviating the performance constraints encountered in IA segmentation due to the limited availability of annotated DSA data. Our extensive experiments on the DIAS dataset demonstrate the effectiveness of these methods as potential baselines for future research and clinical applications. The dataset and code are publicly available at https://doi.org/10.5281/zenodo.11396520 and https://github.com/lseventeen/DIAS.

Figures (10)

• Figure 1: A representative sample of intracranial vessels in lateral view from DSA, where t to t+5 are consecutive frames of the arterial phase. The red arrow in the upper left image indicates segments C1-C3 of the intracranial internal carotid artery (ICA), while the lower left image represents the anterior cerebral artery (ACA) and the middle cerebral artery (MCA).
• Figure 2: Full annotated process of DIAS.
• Figure 3: Full annotated representative samples of DIAS at anteroposterior view and lateral view.
• Figure 4: Examples of two forms scribble annotations: the Scribbles Annotation with Little clinical Experience (SALE) and Random Drawing based the Full Annotation (RDFA). Red represents annotated vessel pixels, blue represents annotated background pixels.
• Figure 5: Illustration of the proposed benchmark for DSA-sequence intracranial artery segmentation. (a) Vessel sequence segmentation network with (b) sequence feature extraction module; (c) WSS: scribble learning-based segmentation composed of scribble supervision and consistency regularization; (d) SSS: random patch-based self-training framework. (WSC: Weight Sharing Convolution, SF: Sequence Fusion, GMP: Global Max Pooling, WSD: Weight Sharing Downsampling, DA: Data augmentation, WA: Weak Augmentation, SA: Strong Augmentation)