Weakly Supervised Medical Image Segmentation With Soft Labels and Noise Robust Loss

作     者：Felfeliyan, Banafshe Hareendranathan, Abhilash Kuntze, Gregor Wichuk, Stephanie Forkert, Nils D. Jaremko, Jacob L. Ronsky, Janet L. 

作者机构：Department of Biomedical Engineering University of Calgary CalgaryAB Canada McCaig Institute for Bone Joint Health University of Calgary CalgaryAB Canada Department of Radiology & Diagnostic Imaging University of Alberta EdmontonAB Canada  University of Alberta EdmontonAB Canada Mechanical and Manufacturing Engineering University of Calgary Calgary Canada 

出 版 物：《arXiv》 (arXiv)

年 卷 期：2022年

核心收录：

主　　题：Diagnosis 

摘      要：Recent advances in deep learning algorithms have led to significant benefits for solving many medical image analysis problems. Training deep learning models commonly requires large datasets with expert-labeled annotations. However, acquiring expert-labeled annotation is not only expensive but also is subjective, error-prone, and inter-/intra- observer variability introduces noise to labels. This is particularly a problem when using deep learning models for segmenting medical images due to the ambiguous anatomical boundaries. Image-based medical diagnosis tools using deep learning models trained with incorrect segmentation labels can lead to false diagnoses and treatment suggestions. Multi-rater annotations might be better suited to train deep learning models with small training sets compared to single-rater annotations. The aim of this paper was to develop and evaluate a method to generate probabilistic labels based on multi-rater annotations and anatomical knowledge of the lesion features in MRI and a method to train segmentation models using probabilistic labels using normalized active-passive loss as anoise-tolerant loss function. The model was evaluated by comparing it to binary ground truth for 17 knees MRI scans for clinical segmentation and detection of bone marrow lesions (BML). The proposed method successfully improved precision 14, recall 22, and Dice score 8 percent compared to a binary cross-entropy loss function. Overall, the results of this work suggest that the proposed normalized active-passive loss using soft labels successfully mitigated the effects of noisy labels. © 2022, CC BY-NC-ND.