Multi-Task Self-Supervised Learning for Medical Image Segmentation

Bo Wang; Hang Zhao; Xiongfei Li; Mingjie Tian; Bo Huang; Feiyang Yang

ICASSP 2024accepted0 citations

Multi-Task Self-Supervised Learning for Medical Image Segmentation

Bo Wang, Hang Zhao, Xiongfei Li, Mingjie Tian, Bo Huang, Feiyang Yang

Abstract

Although medical image segmentation has achieved remarkable results with supervised learning, obtaining labeled data remains challenging and costly. To counteract this, we present the MTSPSeg, a multi-task self-supervised learning framework. We establish the dynamic gradient learning rate (DGLR) strategy and the dynamic multi-task loss weight adjustment method (DWL) to relieve potential early convergence and diverse learning difficulties of self-supervised learning. Our method deeply integrates the prior knowledge inherent in medical images, facilitates collaborative learning among tasks, and delves further into the latent information within the images. Experimental results demonstrate that regardless of the proportion of labeled data, our method outperforms baseline models on the KiTS19 and LiTS datasets, achieving state-of-the-art performance on the KiTS19 dataset.

BibTeX

@inproceedings{icassp2024_multitaskselfsup,
  title = {Multi-Task Self-Supervised Learning for Medical Image Segmentation},
  author = {Bo Wang and Hang Zhao and Xiongfei Li and Mingjie Tian and Bo Huang and Feiyang Yang},
  booktitle = {ICASSP 2024},
  year = {2024}
}