Annotation-efficient deep learning for automatic medical image segmentation

TL;DR

This paper introduces AIDE, a framework for medical image segmentation that performs well with limited or noisy annotations, reducing the need for extensive manual labeling and enabling efficient biomedical applications.

Contribution

AIDE is a novel open-source framework that improves medical image segmentation accuracy using scarce or imperfect annotations, surpassing traditional fully-supervised models.

Findings

AIDE outperforms conventional models on datasets with noisy or limited annotations.

Using only 10% of annotations, AIDE achieves segmentation quality comparable to fully-supervised methods.

AIDE enhances annotation efficiency by tenfold, facilitating broader biomedical applications.

Abstract

Automatic medical image segmentation plays a critical role in scientific research and medical care. Existing high-performance deep learning methods typically rely on large training datasets with high-quality manual annotations, which are difficult to obtain in many clinical applications. Here, we introduce Annotation-effIcient Deep lEarning (AIDE), an open-source framework to handle imperfect training datasets. Methodological analyses and empirical evaluations are conducted, and we demonstrate that AIDE surpasses conventional fully-supervised models by presenting better performance on open datasets possessing scarce or noisy annotations. We further test AIDE in a real-life case study for breast tumor segmentation. Three datasets containing 11,852 breast images from three medical centers are employed, and AIDE, utilizing 10% training annotations, consistently produces segmentation maps comparable to those generated by fully-supervised counterparts or provided by independent radiologists. The 10-fold enhanced efficiency in utilizing expert labels has the potential to promote a wide range of biomedical applications.