Automatic ultrasound vessel segmentation with deep spatiotemporal context learning

TL;DR

This paper introduces a deep learning approach that leverages spatiotemporal context in ultrasound sequences to improve real-time segmentation of small vessel structures, aiding vascular disease assessment.

Contribution

It presents a novel deep learning method that integrates temporal, spatial, and feature-aware contextual embeddings across multiple resolutions for ultrasound vessel segmentation.

Findings

Real-time segmentation achieved with the proposed models.

Significant improvement over baseline approaches.

Effective segmentation of small vessels in ultrasound sequences.

Abstract

Accurate, real-time segmentation of vessel structures in ultrasound image sequences can aid in the measurement of lumen diameters and assessment of vascular diseases. This, however, remains a challenging task, particularly for extremely small vessels that are difficult to visualize. We propose to leverage the rich spatiotemporal context available in ultrasound to improve segmentation of small-scale lower-extremity arterial vasculature. We describe efficient deep learning methods that incorporate temporal, spatial, and feature-aware contextual embeddings at multiple resolution scales while jointly utilizing information from B-mode and Color Doppler signals. Evaluating on femoral and tibial artery scans performed on healthy subjects by an expert ultrasonographer, and comparing to consensus expert ground-truth annotations of inner lumen boundaries, we demonstrate real-time segmentation using the context-aware models and show that they significantly outperform comparable baseline approaches.