LSDM: Long-Short Diffeomorphic Motion for Weakly-Supervised Ultrasound Landmark Tracking

TL;DR

This paper introduces LSDM, a weakly-supervised, multi-task diffeomorphic motion network for accurate and robust ultrasound landmark tracking, effectively handling deformation, ambiguity, and partial observations.

Contribution

The paper presents a novel long-short diffeomorphic motion framework with a learnable deformation prior and an EM-based alignment module, enabling effective weakly-supervised landmark tracking.

Findings

Achieves superior or competitive tracking accuracy compared to state-of-the-art methods.

Demonstrates strong generalization across different ultrasound scanners and modalities.

Operates effectively with minimal landmark annotations.

Abstract

Accurate tracking of an anatomical landmark over time has been of high interests for disease assessment such as minimally invasive surgery and tumor radiation therapy. Ultrasound imaging is a promising modality benefiting from low-cost and real-time acquisition. However, generating a precise landmark tracklet is very challenging, as attempts can be easily distorted by different interference such as landmark deformation, visual ambiguity and partial observation. In this paper, we propose a long-short diffeomorphic motion network, which is a multi-task framework with a learnable deformation prior to search for the plausible deformation of landmark. Specifically, we design a novel diffeomorphism representation in both long and short temporal domains for delineating motion margins and reducing long-term cumulative tracking errors. To further mitigate local anatomical ambiguity, we propose an expectation maximisation motion alignment module to iteratively optimize both long and short deformation, aligning to the same directional and spatial representation. The proposed multi-task system can be trained in a weakly-supervised manner, which only requires few landmark annotations for tracking and zero annotation for long-short deformation learning. We conduct extensive experiments on two ultrasound landmark tracking datasets. Experimental results show that our proposed method can achieve better or competitive landmark tracking performance compared with other state-of-the-art tracking methods, with a strong generalization capability across different scanner types and different ultrasound modalities.