OSA: Echocardiography Video Segmentation via Orthogonalized State Update and Anatomical Prior-aware Feature Enhancement

2026-03-27

AI Summary

This work addresses the challenges of spatiotemporal modeling in left ventricular segmentation from echocardiography, which are exacerbated by speckle noise, non-rigid deformations, and rank collapse in the state transition matrix. To mitigate these issues, the authors propose a Stiefel manifold-constrained state evolution framework that incorporates an Orthogonalized State Update (OSU) mechanism to prevent rank collapse. Additionally, an anatomy-aware feature enhancement module is introduced to effectively disentangle structural information from noise, thereby improving temporal consistency and preserving boundary details. Evaluated on the CAMUS and EchoNet-Dynamic datasets, the method achieves state-of-the-art performance in both segmentation accuracy and temporal stability, while maintaining computational efficiency suitable for real-time clinical inference.

Abstract

Accurate and temporally consistent segmentation of the left ventricle from echocardiography videos is essential for estimating the ejection fraction and assessing cardiac function. However, modeling spatiotemporal dynamics remains difficult due to severe speckle noise and rapid non-rigid deformations. Existing linear recurrent models offer efficient in-context associative recall for temporal tracking, but rely on unconstrained state updates, which cause progressive singular value decay in the state matrix, a phenomenon known as rank collapse, resulting in anatomical details being overwhelmed by noise. To address this, we propose OSA, a framework that constrains the state evolution on the Stiefel manifold. We introduce the Orthogonalized State Update (OSU) mechanism, which formulates the memory evolution as Euclidean projected gradient descent on the Stiefel manifold to prevent rank collapse and maintain stable temporal transitions. Furthermore, an Anatomical Prior-aware Feature Enhancement module explicitly separates anatomical structures from speckle noise through a physics-driven process, providing the temporal tracker with noise-resilient structural cues. Comprehensive experiments on the CAMUS and EchoNet-Dynamic datasets show that OSA achieves state-of-the-art segmentation accuracy and temporal stability, while maintaining real-time inference efficiency for clinical deployment. Codes are available at https://github.com/wangrui2025/OSA.

Problem

• echocardiography video segmentation
• speckle noise
• non-rigid deformation
• temporal consistency
• left ventricle segmentation

Innovation

• Orthogonalized State Update
• Stiefel manifold
• Anatomical Prior-aware Feature Enhancement
• Rank collapse prevention
• Echocardiography video segmentation