Warm Start Active Learning with Proxy Labels \& Selection via Semi-Supervised Fine-Tuning

TL;DR

This paper introduces novel active learning strategies for 3D medical image segmentation, addressing cold start issues with proxy tasks and enhancing performance through semi-supervised fine-tuning, demonstrated on public datasets.

Contribution

Proposes a proxy task-based cold start solution and a two-stage semi-supervised framework for active learning in 3D medical imaging segmentation.

Findings

Significant improvement in AL performance with initial data selection.

Effective semi-supervised fine-tuning enhances segmentation accuracy.

Approach validated on large public medical datasets.

Abstract

Which volume to annotate next is a challenging problem in building medical imaging datasets for deep learning. One of the promising methods to approach this question is active learning (AL). However, AL has been a hard nut to crack in terms of which AL algorithm and acquisition functions are most useful for which datasets. Also, the problem is exacerbated with which volumes to label first when there is zero labeled data to start with. This is known as the cold start problem in AL. We propose two novel strategies for AL specifically for 3D image segmentation. First, we tackle the cold start problem by proposing a proxy task and then utilizing uncertainty generated from the proxy task to rank the unlabeled data to be annotated. Second, we craft a two-stage learning framework for each active iteration where the unlabeled data is also used in the second stage as a semi-supervised fine-tuning strategy. We show the promise of our approach on two well-known large public datasets from medical segmentation decathlon. The results indicate that the initial selection of data and semi-supervised framework both showed significant improvement for several AL strategies.