UltraStar: Semantic-Aware Star Graph Modeling for Echocardiography Navigation

TL;DR

UltraStar introduces a semantic-aware star graph model for echocardiography probe navigation, improving robustness and accuracy by leveraging spatial anchors and reducing noise impact in historical data.

Contribution

It reformulates probe navigation as anchor-based localization using a star graph, incorporating semantic-aware sampling to handle noisy historical data effectively.

Findings

Outperforms baseline methods on a large dataset

Scales better with longer input sequences

Provides more accurate and robust navigation results

Abstract

Echocardiography is critical for diagnosing cardiovascular diseases, yet the shortage of skilled sonographers hinders timely patient care, due to high operational difficulties. Consequently, research on automated probe navigation has significant clinical potential. To achieve robust navigation, it is essential to leverage historical scanning information, mimicking how experts rely on past feedback to adjust subsequent maneuvers. Practical scanning data collected from sonographers typically consists of noisy trajectories inherently generated through trial-and-error exploration. However, existing methods typically model this history as a sequential chain, forcing models to overfit these noisy paths, leading to performance degradation on long sequences. In this paper, we propose UltraStar, which reformulates probe navigation from path regression to anchor-based global localization. By establishing a Star Graph, UltraStar treats historical keyframes as spatial anchors connected directly to the current view, explicitly modeling geometric constraints for precise positioning. We further enhance the Star Graph with a semantic-aware sampling strategy that actively selects the representative landmarks from massive history logs, reducing redundancy for accurate anchoring. Extensive experiments on a dataset with over 1.31 million samples demonstrate that UltraStar outperforms baselines and scales better with longer input lengths, revealing a more effective topology for history modeling under noisy exploration.