Which images to label for few-shot medical landmark detection?

TL;DR

This paper introduces a novel Sample Choosing Policy (SCP) for selecting the most informative images for annotation in few-shot medical landmark detection, significantly improving accuracy by reducing mean radial errors.

Contribution

It proposes a new SCP method that combines self-supervised training, key point proposal, and representative scoring to select optimal images for annotation in few-shot learning.

Findings

Reduces mean radial errors by 14.2% on Cephalometric dataset.

Reduces mean radial errors by 35.5% on HandXray dataset.

Demonstrates effectiveness across three public datasets.

Abstract

The success of deep learning methods relies on the availability of well-labeled large-scale datasets. However, for medical images, annotating such abundant training data often requires experienced radiologists and consumes their limited time. Few-shot learning is developed to alleviate this burden, which achieves competitive performances with only several labeled data. However, a crucial yet previously overlooked problem in few-shot learning is about the selection of template images for annotation before learning, which affects the final performance. We herein propose a novel Sample Choosing Policy (SCP) to select "the most worthy" images for annotation, in the context of few-shot medical landmark detection. SCP consists of three parts: 1) Self-supervised training for building a pre-trained deep model to extract features from radiological images, 2) Key Point Proposal for localizing informative patches, and 3) Representative Score Estimation for searching the most representative samples or templates. The advantage of SCP is demonstrated by various experiments on three widely-used public datasets. For one-shot medical landmark detection, its use reduces the mean radial errors on Cephalometric and HandXray datasets by 14.2% (from 3.595mm to 3.083mm) and 35.5% (4.114mm to 2.653mm), respectively.