EVIL: Evidential Inference Learning for Trustworthy Semi-Supervised Medical Image Segmentation

• Published in: 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI) pp. 1 - 5
• Main Authors: Chen, Yingyu, Yang, Ziyuan, Shen, Chenyu, Wang, Zhiwen, Qin, Yang, Zhang, Yi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 18.04.2023
• Subjects: Biomedical imaging; Computational efficiency; Estimation; Evidence theory; Evidential Learning; Image segmentation; Medical Image Segmentation; Semi-Supervised Learning; Solids; Training
• ISSN: 1945-8452
• DOI: 10.1109/ISBI53787.2023.10230589

Recently, uncertainty-aware methods have attracted increasing attention in semi-supervised medical image segmentation. However, current methods usually suffer from the drawback that it is difficult to balance the computational cost, estimation accuracy, and theoretical support in a unified framework. To alleviate this problem, we introduce the Dempster-Shafer Theory of Evidence (DST) into semi-supervised medical image segmentation, dubbed EVidential Inference Learning (EVIL). EVIL provides a theoretically guaranteed solution to infer accurate uncertainty quantification in a single forward pass. Trustworthy pseudo labels on unlabeled data are generated after uncertainty estimation. The recently proposed consistency regularization-based training paradigm is adopted in our framework, which enforces the consistency on the perturbed predictions to enhance the generalization with few labeled data. Experimental results show that EVIL achieves competitive performance in comparison with several state-of-the-art methods on the public dataset.