Relative distance matters for one-shot landmark detection

TL;DR

This paper improves one-shot landmark detection by incorporating relative distance bias into contrastive learning, enhancing accuracy and robustness, and introduces a new dataset for biomechanical measurement.

Contribution

The paper proposes CC2Dv2, an enhanced contrastive learning method that uses relative distance bias, and provides a new dataset for biomechanical landmark measurement.

Findings

CC2Dv2 outperforms state-of-the-art methods on public and new datasets.

Incorporating relative distance bias reduces false detections.

The new dataset aids orthopedic research and landmark measurement.

Abstract

Contrastive learning based methods such as cascade comparing to detect (CC2D) have shown great potential for one-shot medical landmark detection. However, the important cue of relative distance between landmarks is ignored in CC2D. In this paper, we upgrade CC2D to version II by incorporating a simple-yet-effective relative distance bias in the training stage, which is theoretically proved to encourage the encoder to project the relatively distant landmarks to the embeddings with low similarities. As consequence, CC2Dv2 is less possible to detect a wrong point far from the correct landmark. Furthermore, we present an open-source, landmark-labeled dataset for the measurement of biomechanical parameters of the lower extremity to alleviate the burden of orthopedic surgeons. The effectiveness of CC2Dv2 is evaluated on the public dataset from the ISBI 2015 Grand-Challenge of cephalometric radiographs and our new dataset, which greatly outperforms the state-of-the-art one-shot landmark detection approaches.