Enhancing Weakly Supervised 3D Medical Image Segmentation through Probabilistic-aware Learning
Keywords: Weakly Supervised Learning; Medical Image Segmentation; Probabilistic Methods
Abstract: 3D medical image segmentation is a challenging task with crucial implications for disease diagnosis and treatment planning. Recent advances in deep learning have significantly enhanced fully supervised medical image segmentation. However, this approach heavily relies on labor-intensive and time-consuming fully annotated ground-truth labels, particularly for 3D volumes. To overcome this limitation, we propose an innovative probabilistic-aware weakly supervised learning pipeline tailored for 3D medical image segmentation. Our pipeline consists of three key components. Firstly, we introduce a Probability-based Pseudo Label Generation scheme that synthesizes dense 3D segmentation masks from sparsely annotated point annotations. Secondly, we develop a Probabilistic Multi-head Self-Attention network to extract robust probability-driven features, forming the foundation of our Probabilistic Transformer Network. Lastly, we incorporate a Probability-informed Segmentation Loss Function that effectively guides the training process by incorporating annotation confidence. Experimental results demonstrate significant improvements in weakly supervised segmentation, surpassing state-of-the-art methods.
Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 4922
Loading