OccluNet: Spatio-Temporal Deep Learning for Occlusion Detection on DSA

2025-08-19

AI Summary

To address the challenge of detecting vessel occlusion in digital subtraction angiography (DSA) sequences during endovascular thrombectomy (EVT) for acute ischemic stroke (AIS), this paper proposes a spatiotemporal deep learning model integrating YOLOX-based spatial detection with Transformer-based temporal attention. It is the first work to combine a single-stage detector with either pure temporal or stepwise spatiotemporal attention mechanisms, enabling effective modeling of dynamic cerebral perfusion patterns and improving cross-frame detection consistency. The model employs minimum-density projection preprocessing and single-frame sequence contrastive training for end-to-end learning on the MR CLEAN Registry dataset. It achieves 89.02% precision and 74.87% recall—significantly outperforming the YOLOv11 baseline—while the two attention variants exhibit comparable performance. The source code is publicly available.

Technology Category

• Computer Vision: Segmentation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Security and Privacy: Data transparency and provenance
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate detection of vascular occlusions during endovascular thrombectomy (EVT) is critical in acute ischemic stroke (AIS). Interpretation of digital subtraction angiography (DSA) sequences poses challenges due to anatomical complexity and time constraints. This work proposes OccluNet, a spatio-temporal deep learning model that integrates YOLOX, a single-stage object detector, with transformer-based temporal attention mechanisms to automate occlusion detection in DSA sequences. We compared OccluNet with a YOLOv11 baseline trained on either individual DSA frames or minimum intensity projections. Two spatio-temporal variants were explored for OccluNet: pure temporal attention and divided space-time attention. Evaluation on DSA images from the MR CLEAN Registry revealed the model's capability to capture temporally consistent features, achieving precision and recall of 89.02% and 74.87%, respectively. OccluNet significantly outperformed the baseline models, and both attention variants attained similar performance. Source code is available at https://github.com/anushka-kore/OccluNet.git

Problem

• Automating vascular occlusion detection in DSA sequences
• Addressing anatomical complexity and time constraints in DSA interpretation
• Integrating spatio-temporal deep learning for acute ischemic stroke

Innovation

• Integrates YOLOX with transformer temporal attention
• Uses spatio-temporal deep learning for occlusion detection
• Compares pure temporal and divided space-time attention variants