1
|
Bai L, Wang D, Wang H, Barnett M, Cabezas M, Cai W, Calamante F, Kyle K, Liu D, Ly L, Nguyen A, Shieh CC, Sullivan R, Zhan G, Ouyang W, Wang C. Improving multiple sclerosis lesion segmentation across clinical sites: A federated learning approach with noise-resilient training. Artif Intell Med 2024; 152:102872. [PMID: 38701636 DOI: 10.1016/j.artmed.2024.102872] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/26/2023] [Revised: 03/28/2024] [Accepted: 04/15/2024] [Indexed: 05/05/2024]
Abstract
Accurately measuring the evolution of Multiple Sclerosis (MS) with magnetic resonance imaging (MRI) critically informs understanding of disease progression and helps to direct therapeutic strategy. Deep learning models have shown promise for automatically segmenting MS lesions, but the scarcity of accurately annotated data hinders progress in this area. Obtaining sufficient data from a single clinical site is challenging and does not address the heterogeneous need for model robustness. Conversely, the collection of data from multiple sites introduces data privacy concerns and potential label noise due to varying annotation standards. To address this dilemma, we explore the use of the federated learning framework while considering label noise. Our approach enables collaboration among multiple clinical sites without compromising data privacy under a federated learning paradigm that incorporates a noise-robust training strategy based on label correction. Specifically, we introduce a Decoupled Hard Label Correction (DHLC) strategy that considers the imbalanced distribution and fuzzy boundaries of MS lesions, enabling the correction of false annotations based on prediction confidence. We also introduce a Centrally Enhanced Label Correction (CELC) strategy, which leverages the aggregated central model as a correction teacher for all sites, enhancing the reliability of the correction process. Extensive experiments conducted on two multi-site datasets demonstrate the effectiveness and robustness of our proposed methods, indicating their potential for clinical applications in multi-site collaborations to train better deep learning models with lower cost in data collection and annotation.
Collapse
Affiliation(s)
- Lei Bai
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; School of Electrical and Information Engineering, The University of Sydney, NSW 2006, Australia
| | - Dongang Wang
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia.
| | - Hengrui Wang
- Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Michael Barnett
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia; Royal Prince Alfred Hospital, NSW, 2050, Australia
| | - Mariano Cabezas
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia
| | - Weidong Cai
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; School of Computer Science, The University of Sydney, NSW 2006, Australia
| | - Fernando Calamante
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; School of Biomedical Engineering, The University of Sydney, NSW 2006, Australia; Sydney Imaging, The University of Sydney, NSW 2006, Australia
| | - Kain Kyle
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Dongnan Liu
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; School of Computer Science, The University of Sydney, NSW 2006, Australia
| | - Linda Ly
- Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Aria Nguyen
- Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Chun-Chien Shieh
- Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Ryan Sullivan
- School of Biomedical Engineering, The University of Sydney, NSW 2006, Australia; Australian Imaging Service, NSW 2006, Australia
| | - Geng Zhan
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia
| | - Wanli Ouyang
- School of Electrical and Information Engineering, The University of Sydney, NSW 2006, Australia
| | - Chenyu Wang
- Brain and Mind Centre, The University of Sydney, NSW 2050, Australia; Sydney Neuroimaging Analysis Centre, 94 Mallett Street, NSW 2050, Australia.
| |
Collapse
|
2
|
Deng L, Yang B, Kang Z, Xiang Y. Invariant feature based label correction for DNN when Learning with Noisy Labels. Neural Netw 2024; 172:106137. [PMID: 38309136 DOI: 10.1016/j.neunet.2024.106137] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/06/2023] [Revised: 11/14/2023] [Accepted: 01/17/2024] [Indexed: 02/05/2024]
Abstract
Learning with Noisy Labels (LNL) methods have been widely studied in recent years, which aims to improve the performance of Deep Neural Networks (DNNs) when the training dataset contains incorrectly annotated labels. Popular existing LNL methods rely on semantic features extracted by the DNN to detect and mitigate label noise. However, these extracted features are often spurious and contain unstable correlations with the label across different environments (domains), which can occasionally lead to incorrect prediction and compromise the efficacy of LNL methods. To mitigate this insufficiency, we propose Invariant Feature based Label Correction (IFLC), which reduces spurious features and accurately utilizes the learned invariant features that contain stable correlation to correct label noise. To the best of our knowledge, this is the first attempt to mitigate the issue of spurious features for LNL methods. IFLC consists of two critical processes: The Label Disturbing (LD) process and the Representation Decorrelation (RD) process. The LD process aims to encourage DNN to attain stable performance across different environments, thus reducing the captured spurious features. The RD process strengthens independence between each dimension of the representation vector, thus enabling accurate utilization of the learned invariant features for label correction. We then utilize robust linear regression for the feature representation to conduct label correction. We evaluated the effectiveness of our proposed method and compared it with state-of-the-art (sota) LNL methods on four benchmark datasets, CIFAR-10, CIFAR-100, Animal-10N, and Clothing1M. The experimental results show that our proposed method achieved comparable or even better performance than the existing sota methods. The source codes are available at https://github.com/yangbo1973/IFLC.
Collapse
Affiliation(s)
- Lihui Deng
- School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China.
| | - Bo Yang
- School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China.
| | - Zhongfeng Kang
- School of Information Science and Engineering, Lanzhou University, Lanzhou, Gansu 730000, China.
| | - Yanping Xiang
- School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan 611731, China.
| |
Collapse
|
3
|
Liu L, Zhang Z, Li S, Ma K, Zheng Y. S-CUDA: Self-cleansing unsupervised domain adaptation for medical image segmentation. Med Image Anal 2021; 74:102214. [PMID: 34464837 DOI: 10.1016/j.media.2021.102214] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/15/2020] [Revised: 08/09/2021] [Accepted: 08/10/2021] [Indexed: 01/08/2023]
Abstract
Medical image segmentation tasks hitherto have achieved excellent progresses with large-scale datasets, which empowers us to train potent deep convolutional neural networks (DCNNs). However, labeling such large-scale datasets is laborious and error-prone, which leads the noisy (or incorrect) labels to be an ubiquitous problem in the real-world scenarios. In addition, data collected from different sites usually exhibit significant data distribution shift (or domain shift). As a result, noisy label and domain shift become two common problems in medical imaging application scenarios, especially in medical image segmentation, which degrade the performance of deep learning models significantly. In this paper, we identify a novel problem hidden in medical image segmentation, which is unsupervised domain adaptation on noisy labeled data, and propose a novel algorithm named "Self-Cleansing Unsupervised Domain Adaptation" (S-CDUA) to address such issue. S-CUDA sets up a realistic scenario to solve the above problems simultaneously where training data (i.e., source domain) not only shows domain shift w.r.t. unsupervised test data (i.e., target domain) but also contains noisy labels. The key idea of S-CUDA is to learn noise-excluding and domain invariant knowledge from noisy supervised data, which will be applied on the highly corrupted data for label cleansing and further data-recycling, as well as on the test data with domain shift for supervised propagation. To this end, we propose a novel framework leveraging noisy-label learning and domain adaptation techniques to cleanse the noisy labels and learn from trustable clean samples, thus enabling robust adaptation and prediction on the target domain. Specifically, we train two peer adversarial networks to identify high-confidence clean data and exchange them in companions to eliminate the error accumulation problem and narrow the domain gap simultaneously. In the meantime, the high-confidence noisy data are detected and cleansed in order to reuse the contaminated training data. Therefore, our proposed method can not only cleanse the noisy labels in the training set but also take full advantage of the existing noisy data to update the parameters of the network. For evaluation, we conduct experiments on two popular datasets (REFUGE and Drishti-GS) for optic disc (OD) and optic cup (OC) segmentation, and on another public multi-vendor dataset for spinal cord gray matter (SCGM) segmentation. Experimental results show that our proposed method can cleanse noisy labels efficiently and obtain a model with better generalization performance at the same time, which outperforms previous state-of-the-art methods by large margin. Our code can be found at https://github.com/zzdxjtu/S-cuda.
Collapse
Affiliation(s)
- Luyan Liu
- Tencent Jarvis Lab, Shenzhen 518040, China; Tencent Healthcare (Shenzhen) Co., LTD, China.
| | - Zhengdong Zhang
- State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing 100191, China
| | - Shuai Li
- State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing 100191, China
| | - Kai Ma
- Tencent Jarvis Lab, Shenzhen 518040, China; Tencent Healthcare (Shenzhen) Co., LTD, China
| | - Yefeng Zheng
- Tencent Jarvis Lab, Shenzhen 518040, China; Tencent Healthcare (Shenzhen) Co., LTD, China
| |
Collapse
|