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.
@article{arxiv.2012.04885,
title = {Annotation-efficient deep learning for automatic medical image segmentation},
author = {Shanshan Wang and Cheng Li and Rongpin Wang and Zaiyi Liu and Meiyun Wang and Hongna Tan and Yaping Wu and Xinfeng Liu and Hui Sun and Rui Yang and Xin Liu and Jie Chen and Huihui Zhou and Ismail Ben Ayed and Hairong Zheng},
journal= {arXiv preprint arXiv:2012.04885},
year = {2021}
}