One-Shot Medical Landmark Localization by Edge-Guided Transform and Noisy Landmark Refinement

TL;DR

This paper introduces a two-stage one-shot medical landmark localization framework that leverages unsupervised registration, edge-guided transformations, and noisy pseudo labels to achieve state-of-the-art results with minimal annotation effort.

Contribution

It proposes a novel two-stage approach combining unsupervised registration and robust detector training with edge-guided transformations and self/ cross-consistency for semi-supervised learning.

Findings

Achieves state-of-the-art performance on multiple datasets.

Effectively handles structure variations with edge-guided transformations.

Reduces annotation costs for medical landmark localization.

Abstract

As an important upstream task for many medical applications, supervised landmark localization still requires non-negligible annotation costs to achieve desirable performance. Besides, due to cumbersome collection procedures, the limited size of medical landmark datasets impacts the effectiveness of large-scale self-supervised pre-training methods. To address these challenges, we propose a two-stage framework for one-shot medical landmark localization, which first infers landmarks by unsupervised registration from the labeled exemplar to unlabeled targets, and then utilizes these noisy pseudo labels to train robust detectors. To handle the significant structure variations, we learn an end-to-end cascade of global alignment and local deformations, under the guidance of novel loss functions which incorporate edge information. In stage II, we explore self-consistency for selecting reliable pseudo labels and cross-consistency for semi-supervised learning. Our method achieves state-of-the-art performances on public datasets of different body parts, which demonstrates its general applicability.