O-MedAL: Online Active Deep Learning for Medical Image Analysis

TL;DR

This paper introduces O-MedAL, an online active deep learning framework for medical image analysis that improves model accuracy and efficiency by smart sampling and online training, especially in imbalanced datasets.

Contribution

It presents a novel online active learning method with a new sampling strategy that enhances deep network performance and reduces labeling effort in medical imaging.

Findings

Model accuracy improved by 6.30%

Achieved baseline accuracy with only 25% labeled data

Reduced training images by up to 67%

Abstract

Active Learning methods create an optimized labeled training set from unlabeled data. We introduce a novel Online Active Deep Learning method for Medical Image Analysis. We extend our MedAL active learning framework to present new results in this paper. Our novel sampling method queries the unlabeled examples that maximize the average distance to all training set examples. Our online method enhances performance of its underlying baseline deep network. These novelties contribute significant performance improvements, including improving the model's underlying deep network accuracy by 6.30%, using only 25% of the labeled dataset to achieve baseline accuracy, reducing backpropagated images during training by as much as 67%, and demonstrating robustness to class imbalance in binary and multi-class tasks.