GDKVM: Echocardiography Video Segmentation via Spatiotemporal Key-Value Memory with Gated Delta Rule

TL;DR

GDKVM is a novel neural network architecture that improves echocardiography video segmentation by modeling inter-frame correlations, efficiently storing memory, and fusing multi-scale features, achieving higher accuracy and robustness in real-time.

Contribution

The paper introduces GDKVM, combining LKVA, GDR, and KPFF modules for enhanced, efficient, and robust cardiac video segmentation, surpassing existing methods.

Findings

Outperforms state-of-the-art methods in accuracy and robustness

Achieves real-time segmentation performance

Validated on CAMUS and EchoNet-Dynamic datasets

Abstract

Accurate segmentation of cardiac chambers in echocardiography sequences is crucial for the quantitative analysis of cardiac function, aiding in clinical diagnosis and treatment. The imaging noise, artifacts, and the deformation and motion of the heart pose challenges to segmentation algorithms. While existing methods based on convolutional neural networks, Transformers, and space-time memory networks have improved segmentation accuracy, they often struggle with the trade-off between capturing long-range spatiotemporal dependencies and maintaining computational efficiency with fine-grained feature representation. In this paper, we introduce GDKVM, a novel architecture for echocardiography video segmentation. The model employs Linear Key-Value Association (LKVA) to effectively model inter-frame correlations, and introduces Gated Delta Rule (GDR) to efficiently store intermediate memory states. Key-Pixel Feature Fusion (KPFF) module is designed to integrate local and global features at multiple scales, enhancing robustness against boundary blurring and noise interference. We validated GDKVM on two mainstream echocardiography video datasets (CAMUS and EchoNet-Dynamic) and compared it with various state-of-the-art methods. Experimental results show that GDKVM outperforms existing approaches in terms of segmentation accuracy and robustness, while ensuring real-time performance. Code is available at https://github.com/wangrui2025/GDKVM.