1
|
Liu X, Qu L, Xie Z, Zhao J, Shi Y, Song Z. Towards more precise automatic analysis: a systematic review of deep learning-based multi-organ segmentation. Biomed Eng Online 2024; 23:52. [PMID: 38851691 PMCID: PMC11162022 DOI: 10.1186/s12938-024-01238-8] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/08/2023] [Accepted: 04/11/2024] [Indexed: 06/10/2024] Open
Abstract
Accurate segmentation of multiple organs in the head, neck, chest, and abdomen from medical images is an essential step in computer-aided diagnosis, surgical navigation, and radiation therapy. In the past few years, with a data-driven feature extraction approach and end-to-end training, automatic deep learning-based multi-organ segmentation methods have far outperformed traditional methods and become a new research topic. This review systematically summarizes the latest research in this field. We searched Google Scholar for papers published from January 1, 2016 to December 31, 2023, using keywords "multi-organ segmentation" and "deep learning", resulting in 327 papers. We followed the PRISMA guidelines for paper selection, and 195 studies were deemed to be within the scope of this review. We summarized the two main aspects involved in multi-organ segmentation: datasets and methods. Regarding datasets, we provided an overview of existing public datasets and conducted an in-depth analysis. Concerning methods, we categorized existing approaches into three major classes: fully supervised, weakly supervised and semi-supervised, based on whether they require complete label information. We summarized the achievements of these methods in terms of segmentation accuracy. In the discussion and conclusion section, we outlined and summarized the current trends in multi-organ segmentation.
Collapse
Affiliation(s)
- Xiaoyu Liu
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China
| | - Linhao Qu
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China
| | - Ziyue Xie
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China
| | - Jiayue Zhao
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China
| | - Yonghong Shi
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China.
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China.
| | - Zhijian Song
- Digital Medical Research Center, School of Basic Medical Sciences, Fudan University, 138 Yixueyuan Road, Shanghai, 200032, People's Republic of China.
- Shanghai Key Laboratory of Medical Image Computing and Computer Assisted Intervention, Shanghai, 200032, China.
| |
Collapse
|
2
|
Thwal CM, Nguyen MNH, Tun YL, Kim ST, Thai MT, Hong CS. OnDev-LCT: On-Device Lightweight Convolutional Transformers towards federated learning. Neural Netw 2024; 170:635-649. [PMID: 38100846 DOI: 10.1016/j.neunet.2023.11.044] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/21/2023] [Revised: 10/26/2023] [Accepted: 11/20/2023] [Indexed: 12/17/2023]
Abstract
Federated learning (FL) has emerged as a promising approach to collaboratively train machine learning models across multiple edge devices while preserving privacy. The success of FL hinges on the efficiency of participating models and their ability to handle the unique challenges of distributed learning. While several variants of Vision Transformer (ViT) have shown great potential as alternatives to modern convolutional neural networks (CNNs) for centralized training, the unprecedented size and higher computational demands hinder their deployment on resource-constrained edge devices, challenging their widespread application in FL. Since client devices in FL typically have limited computing resources and communication bandwidth, models intended for such devices must strike a balance between model size, computational efficiency, and the ability to adapt to the diverse and non-IID data distributions encountered in FL. To address these challenges, we propose OnDev-LCT: Lightweight Convolutional Transformers for On-Device vision tasks with limited training data and resources. Our models incorporate image-specific inductive biases through the LCT tokenizer by leveraging efficient depthwise separable convolutions in residual linear bottleneck blocks to extract local features, while the multi-head self-attention (MHSA) mechanism in the LCT encoder implicitly facilitates capturing global representations of images. Extensive experiments on benchmark image datasets indicate that our models outperform existing lightweight vision models while having fewer parameters and lower computational demands, making them suitable for FL scenarios with data heterogeneity and communication bottlenecks.
Collapse
Affiliation(s)
- Chu Myaet Thwal
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - Minh N H Nguyen
- Vietnam - Korea University of Information and Communication Technology, Danang, Viet Nam.
| | - Ye Lin Tun
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - Seong Tae Kim
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - My T Thai
- Department of Computer and Information Science and Engineering, University of Florida, Gainesville, Florida 32611, USA.
| | - Choong Seon Hong
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| |
Collapse
|
3
|
Chen X, Niu Y, Zhao Y, Qin X. An Efficient Group Federated Learning Framework for Large-Scale EEG-Based Driver Drowsiness Detection. Int J Neural Syst 2024; 34:2450003. [PMID: 37964570 DOI: 10.1142/s0129065724500035] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2023]
Abstract
To avoid traffic accidents, monitoring the driver's electroencephalogram (EEG) signals to assess drowsiness is an effective solution. However, aggregating the personal data of these drivers may lead to insufficient data usage and pose a risk of privacy breaches. To address these issues, a framework called Group Federated Learning (Group-FL) for large-scale driver drowsiness detection is proposed, which can efficiently utilize diverse client data while protecting privacy. First, by arranging the clients into different levels of groups and gradually aggregating their model parameters from low-level groups to high-level groups, communication and time costs are reduced. In addition, to solve the problem of notable variations in EEG signals among different clients, a global-personalized deep neural network is designed. The global model extracts shared features from various clients, while the personalized model extracts fine-grained features from each client and outputs classification results. Finally, to address special issues such as scale/category imbalance and data pollution, three checking modules are designed for adjusting grouping, evaluating client data, and effectively applying personalized models. Through extensive experimentation, the effectiveness of each component within the framework was validated, and a mean accuracy, F1-score, and Area Under Curve (AUC) of 81.0%, 82.0%, and 87.9% was achieved, respectively, on a publicly available dataset comprising 11 subjects.
Collapse
Affiliation(s)
- Xinyuan Chen
- School of Information Science and Engineering, Shandong Normal University, Jinan 250014, P. R. China
| | - Yi Niu
- School of Information Science and Engineering, Shandong Normal University, Jinan 250014, P. R. China
| | - Yanna Zhao
- School of Information Science and Engineering, Shandong Normal University, Jinan 250014, P. R. China
| | - Xue Qin
- School of Information Science and Engineering, Shandong Normal University, Jinan 250014, P. R. China
| |
Collapse
|
4
|
Wang J, Jin Y, Stoyanov D, Wang L. FedDP: Dual Personalization in Federated Medical Image Segmentation. IEEE TRANSACTIONS ON MEDICAL IMAGING 2024; 43:297-308. [PMID: 37494156 DOI: 10.1109/tmi.2023.3299206] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 07/28/2023]
Abstract
Personalized federated learning (PFL) addresses the data heterogeneity challenge faced by general federated learning (GFL). Rather than learning a single global model, with PFL a collection of models are adapted to the unique feature distribution of each site. However, current PFL methods rarely consider self-attention networks which can handle data heterogeneity by long-range dependency modeling and they do not utilize prediction inconsistencies in local models as an indicator of site uniqueness. In this paper, we propose FedDP, a novel fed erated learning scheme with d ual p ersonalization, which improves model personalization from both feature and prediction aspects to boost image segmentation results. We leverage long-range dependencies by designing a local query (LQ) that decouples the query embedding layer out of each local model, whose parameters are trained privately to better adapt to the respective feature distribution of the site. We then propose inconsistency-guided calibration (IGC), which exploits the inter-site prediction inconsistencies to accommodate the model learning concentration. By encouraging a model to penalize pixels with larger inconsistencies, we better tailor prediction-level patterns to each local site. Experimentally, we compare FedDP with the state-of-the-art PFL methods on two popular medical image segmentation tasks with different modalities, where our results consistently outperform others on both tasks. Our code and models are available at https://github.com/jcwang123/PFL-Seg-Trans.
Collapse
|
5
|
Hsu HY, Keoy KH, Chen JR, Chao HC, Lai CF. Personalized Federated Learning Algorithm with Adaptive Clustering for Non-IID IoT Data Incorporating Multi-Task Learning and Neural Network Model Characteristics. SENSORS (BASEL, SWITZERLAND) 2023; 23:9016. [PMID: 38005404 PMCID: PMC10675707 DOI: 10.3390/s23229016] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 09/27/2023] [Revised: 10/30/2023] [Accepted: 11/03/2023] [Indexed: 11/26/2023]
Abstract
The proliferation of IoT devices has led to an unprecedented integration of machine learning techniques, raising concerns about data privacy. To address these concerns, federated learning has been introduced. However, practical implementations face challenges, including communication costs, data and device heterogeneity, and privacy security. This paper proposes an innovative approach within the context of federated learning, introducing a personalized joint learning algorithm for Non-IID IoT data. This algorithm incorporates multi-task learning principles and leverages neural network model characteristics. To overcome data heterogeneity, we present a novel clustering algorithm designed specifically for federated learning. Unlike conventional methods that require a predetermined number of clusters, our approach utilizes automatic clustering, eliminating the need for fixed cluster specifications. Extensive experimentation demonstrates the exceptional performance of the proposed algorithm, particularly in scenarios with specific client distributions. By significantly improving the accuracy of trained models, our approach not only addresses data heterogeneity but also strengthens privacy preservation in federated learning. In conclusion, we offer a robust solution to the practical challenges of federated learning in IoT environments. By combining personalized joint learning, automatic clustering, and neural network model characteristics, we facilitate more effective and privacy-conscious machine learning in Non-IID IoT data settings.
Collapse
Affiliation(s)
- Hua-Yang Hsu
- Shenzhen Graduate School, Peking University, Beijing 100191, China;
| | - Kay Hooi Keoy
- UCSI Graduate Business School, UCSI University, Kuala Lumpur 56000, Malaysia;
| | - Jun-Ru Chen
- Department of Engineering Science, National Cheng Kung University, Tainan 70101, Taiwan;
| | - Han-Chieh Chao
- Department of Electrical Engineering, National Dong Hwa University, Hualien 974301, Taiwan
| | - Chin-Feng Lai
- Department of Engineering Science, National Cheng Kung University, Tainan 70101, Taiwan;
| |
Collapse
|
6
|
Asad M, Shaukat S, Hu D, Wang Z, Javanmardi E, Nakazato J, Tsukada M. Limitations and Future Aspects of Communication Costs in Federated Learning: A Survey. SENSORS (BASEL, SWITZERLAND) 2023; 23:7358. [PMID: 37687814 PMCID: PMC10490700 DOI: 10.3390/s23177358] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/31/2023] [Revised: 08/19/2023] [Accepted: 08/20/2023] [Indexed: 09/10/2023]
Abstract
This paper explores the potential for communication-efficient federated learning (FL) in modern distributed systems. FL is an emerging distributed machine learning technique that allows for the distributed training of a single machine learning model across multiple geographically distributed clients. This paper surveys the various approaches to communication-efficient FL, including model updates, compression techniques, resource management for the edge and cloud, and client selection. We also review the various optimization techniques associated with communication-efficient FL, such as compression schemes and structured updates. Finally, we highlight the current research challenges and discuss the potential future directions for communication-efficient FL.
Collapse
Affiliation(s)
- Muhammad Asad
- Graduate School of Information Science and Technology, Department of Creative Informatics, The University of Tokyo, Tokyo 113-8654, Japan; (S.S.); (D.H.); (Z.W.); (E.J.); (J.N.); (M.T.)
| | | | | | | | | | | | | |
Collapse
|
7
|
Tun YL, Nguyen MNH, Thwal CM, Choi J, Hong CS. Contrastive encoder pre-training-based clustered federated learning for heterogeneous data. Neural Netw 2023; 165:689-704. [PMID: 37385023 DOI: 10.1016/j.neunet.2023.06.010] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/23/2022] [Revised: 05/26/2023] [Accepted: 06/05/2023] [Indexed: 07/01/2023]
Abstract
Federated learning (FL) is a promising approach that enables distributed clients to collaboratively train a global model while preserving their data privacy. However, FL often suffers from data heterogeneity problems, which can significantly affect its performance. To address this, clustered federated learning (CFL) has been proposed to construct personalized models for different client clusters. One effective client clustering strategy is to allow clients to choose their own local models from a model pool based on their performance. However, without pre-trained model parameters, such a strategy is prone to clustering failure, in which all clients choose the same model. Unfortunately, collecting a large amount of labeled data for pre-training can be costly and impractical in distributed environments. To overcome this challenge, we leverage self-supervised contrastive learning to exploit unlabeled data for the pre-training of FL systems. Together, self-supervised pre-training and client clustering can be crucial components for tackling the data heterogeneity issues of FL. Leveraging these two crucial strategies, we propose contrastive pre-training-based clustered federated learning (CP-CFL) to improve the model convergence and overall performance of FL systems. In this work, we demonstrate the effectiveness of CP-CFL through extensive experiments in heterogeneous FL settings, and present various interesting observations.
Collapse
Affiliation(s)
- Ye Lin Tun
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - Minh N H Nguyen
- Vietnam - Korea University of Information and Communication Technology, Danang, Viet Nam.
| | - Chu Myaet Thwal
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - Jinwoo Choi
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| | - Choong Seon Hong
- Department of Computer Science and Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 17104, South Korea.
| |
Collapse
|
8
|
Yan R, Qu L, Wei Q, Huang SC, Shen L, Rubin DL, Xing L, Zhou Y. Label-Efficient Self-Supervised Federated Learning for Tackling Data Heterogeneity in Medical Imaging. IEEE TRANSACTIONS ON MEDICAL IMAGING 2023; 42:1932-1943. [PMID: 37018314 PMCID: PMC10880587 DOI: 10.1109/tmi.2022.3233574] [Citation(s) in RCA: 5] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
The collection and curation of large-scale medical datasets from multiple institutions is essential for training accurate deep learning models, but privacy concerns often hinder data sharing. Federated learning (FL) is a promising solution that enables privacy-preserving collaborative learning among different institutions, but it generally suffers from performance deterioration due to heterogeneous data distributions and a lack of quality labeled data. In this paper, we present a robust and label-efficient self-supervised FL framework for medical image analysis. Our method introduces a novel Transformer-based self-supervised pre-training paradigm that pre-trains models directly on decentralized target task datasets using masked image modeling, to facilitate more robust representation learning on heterogeneous data and effective knowledge transfer to downstream models. Extensive empirical results on simulated and real-world medical imaging non-IID federated datasets show that masked image modeling with Transformers significantly improves the robustness of models against various degrees of data heterogeneity. Notably, under severe data heterogeneity, our method, without relying on any additional pre-training data, achieves an improvement of 5.06%, 1.53% and 4.58% in test accuracy on retinal, dermatology and chest X-ray classification compared to the supervised baseline with ImageNet pre-training. In addition, we show that our federated self-supervised pre-training methods yield models that generalize better to out-of-distribution data and perform more effectively when fine-tuning with limited labeled data, compared to existing FL algorithms. The code is available at https://github.com/rui-yan/SSL-FL.
Collapse
|
9
|
Shiri I, Vafaei Sadr A, Akhavan A, Salimi Y, Sanaat A, Amini M, Razeghi B, Saberi A, Arabi H, Ferdowsi S, Voloshynovskiy S, Gündüz D, Rahmim A, Zaidi H. Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning. Eur J Nucl Med Mol Imaging 2023; 50:1034-1050. [PMID: 36508026 PMCID: PMC9742659 DOI: 10.1007/s00259-022-06053-8] [Citation(s) in RCA: 12] [Impact Index Per Article: 12.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/13/2022] [Accepted: 11/18/2022] [Indexed: 12/15/2022]
Abstract
PURPOSE Attenuation correction and scatter compensation (AC/SC) are two main steps toward quantitative PET imaging, which remain challenging in PET-only and PET/MRI systems. These can be effectively tackled via deep learning (DL) methods. However, trustworthy, and generalizable DL models commonly require well-curated, heterogeneous, and large datasets from multiple clinical centers. At the same time, owing to legal/ethical issues and privacy concerns, forming a large collective, centralized dataset poses significant challenges. In this work, we aimed to develop a DL-based model in a multicenter setting without direct sharing of data using federated learning (FL) for AC/SC of PET images. METHODS Non-attenuation/scatter corrected and CT-based attenuation/scatter corrected (CT-ASC) 18F-FDG PET images of 300 patients were enrolled in this study. The dataset consisted of 6 different centers, each with 50 patients, with scanner, image acquisition, and reconstruction protocols varying across the centers. CT-based ASC PET images served as the standard reference. All images were reviewed to include high-quality and artifact-free PET images. Both corrected and uncorrected PET images were converted to standardized uptake values (SUVs). We used a modified nested U-Net utilizing residual U-block in a U-shape architecture. We evaluated two FL models, namely sequential (FL-SQ) and parallel (FL-PL) and compared their performance with the baseline centralized (CZ) learning model wherein the data were pooled to one server, as well as center-based (CB) models where for each center the model was built and evaluated separately. Data from each center were divided to contribute to training (30 patients), validation (10 patients), and test sets (10 patients). Final evaluations and reports were performed on 60 patients (10 patients from each center). RESULTS In terms of percent SUV absolute relative error (ARE%), both FL-SQ (CI:12.21-14.81%) and FL-PL (CI:11.82-13.84%) models demonstrated excellent agreement with the centralized framework (CI:10.32-12.00%), while FL-based algorithms improved model performance by over 11% compared to CB training strategy (CI: 22.34-26.10%). Furthermore, the Mann-Whitney test between different strategies revealed no significant differences between CZ and FL-based algorithms (p-value > 0.05) in center-categorized mode. At the same time, a significant difference was observed between the different training approaches on the overall dataset (p-value < 0.05). In addition, voxel-wise comparison, with respect to reference CT-ASC, exhibited similar performance for images predicted by CZ (R2 = 0.94), FL-SQ (R2 = 0.93), and FL-PL (R2 = 0.92), while CB model achieved a far lower coefficient of determination (R2 = 0.74). Despite the strong correlations between CZ and FL-based methods compared to reference CT-ASC, a slight underestimation of predicted voxel values was observed. CONCLUSION Deep learning-based models provide promising results toward quantitative PET image reconstruction. Specifically, we developed two FL models and compared their performance with center-based and centralized models. The proposed FL-based models achieved higher performance compared to center-based models, comparable with centralized models. Our work provided strong empirical evidence that the FL framework can fully benefit from the generalizability and robustness of DL models used for AC/SC in PET, while obviating the need for the direct sharing of datasets between clinical imaging centers.
Collapse
Affiliation(s)
- Isaac Shiri
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Alireza Vafaei Sadr
- Department of Theoretical Physics and Center for Astroparticle Physics, University of Geneva, Geneva, Switzerland.,Institute of Pathology, RWTH Aachen University Hospital, Aachen, Germany
| | - Azadeh Akhavan
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Yazdan Salimi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Amirhossein Sanaat
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Mehdi Amini
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Behrooz Razeghi
- Department of Computer Science, University of Geneva, Geneva, Switzerland
| | - Abdollah Saberi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | - Hossein Arabi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland
| | | | | | - Deniz Gündüz
- Department of Electrical and Electronic Engineering, Imperial College London, London, UK
| | - Arman Rahmim
- Departments of Radiology and Physics, University of British Columbia, Vancouver, Canada.,Department of Integrative Oncology, BC Cancer Research Institute, Vancouver, BC, Canada
| | - Habib Zaidi
- Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland. .,Geneva University Neurocenter, Geneva University, Geneva, Switzerland. .,Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, Groningen, Netherlands. .,Department of Nuclear Medicine, University of Southern Denmark, Odense, Denmark.
| |
Collapse
|