Segmentation is a prerequisite yet challenging task for medical image analysis. In this paper, we introduce a novel deeply supervised active learning approach for finger bones segmentation. The proposed architecture is fine-tuned in an iterative and incremental learning manner. In each step, the deep supervision mechanism guides the learning process of hidden layers and selects samples to be labeled. Extensive experiments demonstrated that our method achieves competitive segmentation results using less labeled samples as compared with full annotation.
@article{arxiv.2005.03225,
title = {Deeply Supervised Active Learning for Finger Bones Segmentation},
author = {Ziyuan Zhao and Xiaoyan Yang and Bharadwaj Veeravalli and Zeng Zeng},
journal= {arXiv preprint arXiv:2005.03225},
year = {2022}
}
Comments
Accepted version to be published in the 42nd IEEE Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2020, Montreal, Canada