Diminishing Uncertainty within the Training Pool: Active Learning for Medical Image Segmentation

TL;DR

This paper introduces a novel active learning framework for medical image segmentation, reducing data requirements by selectively annotating the most informative images, and demonstrates significant data efficiency improvements on MRI and CT datasets.

Contribution

It proposes three new active learning strategies and a query-by-committee approach tailored for medical image segmentation, advancing beyond existing classification-focused methods.

Findings

Achieved full accuracy using only 22.69% of MRI data.

Achieved full accuracy using only 48.85% of CT data.

Improved data efficiency in medical image segmentation tasks.

Abstract

Active learning is a unique abstraction of machine learning techniques where the model/algorithm could guide users for annotation of a set of data points that would be beneficial to the model, unlike passive machine learning. The primary advantage being that active learning frameworks select data points that can accelerate the learning process of a model and can reduce the amount of data needed to achieve full accuracy as compared to a model trained on a randomly acquired data set. Multiple frameworks for active learning combined with deep learning have been proposed, and the majority of them are dedicated to classification tasks. Herein, we explore active learning for the task of segmentation of medical imaging data sets. We investigate our proposed framework using two datasets: 1.) MRI scans of the hippocampus, 2.) CT scans of pancreas and tumors. This work presents a query-by-committee approach for active learning where a joint optimizer is used for the committee. At the same time, we propose three new strategies for active learning: 1.) increasing frequency of uncertain data to bias the training data set; 2.) Using mutual information among the input images as a regularizer for acquisition to ensure diversity in the training dataset; 3.) adaptation of Dice log-likelihood for Stein variational gradient descent (SVGD). The results indicate an improvement in terms of data reduction by achieving full accuracy while only using 22.69 % and 48.85 % of the available data for each dataset, respectively.