pith. sign in

arxiv: 2411.05824 · v3 · submitted 2024-11-05 · 📡 eess.IV · cs.CV· cs.LG

Navigating Distribution Shifts in Medical Image Analysis: A Survey

Zixian Su , Jingwei Guo , Xi Yang , Qiufeng Wang , Frans Coenen , Amir Hussain , Kaizhu Huang This is my paper

Pith reviewed 2026-05-23 17:59 UTC · model grok-4.3

classification 📡 eess.IV cs.CVcs.LG
keywords distribution shiftsmedical image analysisdomain generalizationfederated learningfine-tuningjoint traininguncertainty modelingdeployability
0
0 comments X

The pith

As domain information becomes less accessible in medical image analysis, performance gains are constrained and methods shift from explicit alignment to uncertainty-aware modeling.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper reviews strategies for deep learning models in medical image analysis facing distribution shifts due to varying hospitals and populations. It connects real-world constraints such as limited data access and privacy to four technical paradigms: Joint Training, Federated Learning, Fine-tuning, and Domain Generalization. The review shows that performance improvements become more limited as domain information access decreases across these paradigms. It also notes a methodological evolution toward uncertainty-aware modeling. This framework helps understand how to make models more adaptable in practical healthcare settings with strict protocols.

Core claim

The central claim is that categorizing approaches by alignment with clinical scenarios reveals a trend where reduced domain accessibility constrains performance and shifts focus from distribution alignment to uncertainty modeling, indicating a need for deployability-aware designs in real-world medical image analysis.

What carries the argument

The four-paradigm taxonomy (Joint Training, Federated Learning, Fine-tuning, Domain Generalization) that links operational constraints to methodological choices for handling distribution shifts.

If this is right

  • Performance improvements are increasingly constrained as one progresses from paradigms with more domain access to those with less.
  • Methodological focus gradually shifts from explicit distribution alignment to uncertainty-aware modeling.
  • Real-world MedIA requires more emphasis on deployability-aware design.
  • Each paradigm corresponds to specific healthcare scenarios involving data sharing and collaboration.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Future surveys could test if the same trend appears in non-imaging medical AI tasks.
  • Deployments might benefit from combining uncertainty modeling with elements of earlier paradigms where possible.
  • Researchers could quantify the performance drop in specific clinical settings to validate the constraint claim.

Load-bearing premise

The selected literature provides an unbiased and representative sample for the categorization into Joint Training, Federated Learning, Fine-tuning, and Domain Generalization.

What would settle it

A different selection or categorization of the literature that shows no increasing constraint on performance or no shift to uncertainty modeling would falsify the empirical analysis.

Figures

Figures reproduced from arXiv: 2411.05824 by Amir Hussain, Frans Coenen, Jingwei Guo, Kaizhu Huang, Qiufeng Wang, Xi Yang, Zixian Su.

Figure 1
Figure 1. Figure 1: The diagram categorizes existing deep learning techniques into four main approaches, each addressing [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of medical imaging distribution shifts, showcasing from Imaging Modalities (Cardiac [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Joint Training in MedIA: It enables data sharing among healthcare institutions without privacy [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Federated Learning in MedIA. It facilitates collaborative model training across healthcare institutions [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Fine-tuning in MedIA: It adapts pre-trained models to specialized datasets, particularly when privacy [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Domain Generalization in MedIA: It prepares models for unseen data by generalizing from source [PITH_FULL_IMAGE:figures/full_fig_p020_6.png] view at source ↗
read the original abstract

Medical Image Analysis (MedIA) has become indispensable in modern healthcare, enhancing clinical diagnostics and personalized treatment. Despite the remarkable advancements supported by deep learning (DL) technologies, their practical deployment faces challenges posed by distribution shifts, where models trained on specific datasets underperform on others from varying hospitals, or patient populations. To address this issue, researchers have been actively developing strategies to increase the adaptability of DL models, enabling their effective use in unfamiliar environments. This paper systematically reviews approaches that apply DL techniques to MedIA systems affected by distribution shifts. Rather than organizing existing methods by technical characteristics, we explicitly bridge real-world clinical constraints -- such as limited data accessibility, strict privacy requirements, and heterogeneous collaboration protocols -- with the technical paradigms able to address them. By establishing this connection between operational constraints and methodological evolution, we categorize existing works into Joint Training, Federated Learning, Fine-tuning, and Domain Generalization, each aligned with specific healthcare scenarios. Beyond this taxonomy, our empirical analysis suggests that, as domain information becomes progressively less accessible across these paradigms, performance improvements become increasingly constrained, and further uncovers a gradual shift in methodological focus from explicit distribution alignment toward uncertainty-aware modeling, ultimately pointing to the need for more deployability-aware design in real-world MedIA.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. This survey reviews deep learning methods for distribution shifts in medical image analysis (MedIA). It organizes approaches into four paradigms—Joint Training, Federated Learning, Fine-tuning, and Domain Generalization—explicitly mapped to clinical constraints such as data accessibility and privacy. Beyond the taxonomy, an empirical analysis of the reviewed literature claims that performance gains become progressively constrained and methodological emphasis shifts from explicit alignment to uncertainty-aware modeling as domain information accessibility decreases, advocating for deployability-aware design.

Significance. If the empirical trends prove robust, the work supplies a clinically grounded taxonomy that connects operational constraints to technical evolution in MedIA, a perspective that could usefully inform both survey literature and future method development focused on real-world deployment.

major comments (3)
  1. [Empirical analysis section] Empirical analysis section: No description is given of the literature search protocol, databases queried, inclusion/exclusion criteria, total papers screened, or the quantitative procedure used to measure 'performance improvements' and 'methodological focus' shifts. Without these, the central claim that performance becomes increasingly constrained across the four paradigms cannot be evaluated for selection bias or representativeness.
  2. [Taxonomy section] Taxonomy and ordering (abstract and § on paradigms): The progressive decrease in domain-information accessibility is asserted to order the four categories, yet the manuscript does not demonstrate that every paper assigned to a later category truly has strictly less accessible domain information than those in earlier categories, nor does it address papers that could fit multiple categories.
  3. [Empirical analysis section] Empirical analysis: The reported 'gradual shift' toward uncertainty-aware modeling is presented as an observed trend, but the manuscript supplies neither counts of papers per methodological focus within each paradigm nor explicit criteria for classifying a method as 'uncertainty-aware' versus 'explicit distribution alignment.'
minor comments (2)
  1. [Abstract] Abstract: The acronym 'MedIA' is introduced without expansion on first use.
  2. Figure or table summarizing the taxonomy: A visual mapping of clinical constraints to the four paradigms would improve readability; none appears to be referenced.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thorough review and constructive comments on our survey. We address each of the major comments below and will incorporate revisions to improve the clarity and rigor of the manuscript.

read point-by-point responses

  1. Referee: [Empirical analysis section] Empirical analysis section: No description is given of the literature search protocol, databases queried, inclusion/exclusion criteria, total papers screened, or the quantitative procedure used to measure 'performance improvements' and 'methodological focus' shifts. Without these, the central claim that performance becomes increasingly constrained across the four paradigms cannot be evaluated for selection bias or representativeness.

    Authors: We agree with this observation. The original manuscript omitted a detailed account of the review methodology. In the revised version, we will insert a new subsection detailing the literature search protocol, including the databases (PubMed, IEEE Xplore, arXiv, Google Scholar), search terms (combinations of 'distribution shift', 'domain adaptation', 'medical image analysis', 'deep learning'), inclusion criteria (studies on DL methods for distribution shifts in MedIA published 2015-2024), exclusion criteria (non-DL methods, non-imaging), number of papers screened (approximately 500) and included (around 150), and the quantitative procedure (manual extraction of reported performance deltas and categorization of methods based on their primary technical approach). This will allow readers to assess representativeness and potential biases. revision: yes

  2. Referee: [Taxonomy section] Taxonomy and ordering (abstract and § on paradigms): The progressive decrease in domain-information accessibility is asserted to order the four categories, yet the manuscript does not demonstrate that every paper assigned to a later category truly has strictly less accessible domain information than those in earlier categories, nor does it address papers that could fit multiple categories.

    Authors: The ordering is conceptual, reflecting the typical clinical constraints associated with each paradigm rather than a strict empirical ordering of individual papers. Joint Training assumes shared raw data access, Federated Learning limits to model updates, Fine-tuning involves limited target data, and Domain Generalization assumes no target domain info. We recognize overlaps exist. We will revise the taxonomy section to explicitly state the conceptual nature of the ordering, provide justification based on clinical scenarios, and add a paragraph discussing multi-category papers with examples of our assignment decisions based on the primary clinical scenario addressed. revision: partial

  3. Referee: [Empirical analysis section] Empirical analysis: The reported 'gradual shift' toward uncertainty-aware modeling is presented as an observed trend, but the manuscript supplies neither counts of papers per methodological focus within each paradigm nor explicit criteria for classifying a method as 'uncertainty-aware' versus 'explicit distribution alignment.'

    Authors: We will add explicit classification criteria in the revised empirical analysis: 'explicit distribution alignment' encompasses methods using adversarial learning, statistical distance minimization (e.g., MMD, CORAL), or style transfer; 'uncertainty-aware modeling' includes those employing probabilistic modeling, uncertainty estimation via dropout/ensembles, or evidential learning to handle shifts via uncertainty. We will also include quantitative counts, such as a table summarizing the number of papers in each paradigm focusing on alignment vs. uncertainty-aware approaches, to substantiate the gradual shift claim. revision: yes

Circularity Check

0 steps flagged

No circularity: survey paper with observational claims only

full rationale

This is a literature survey paper that organizes existing external works into four paradigms (Joint Training, Federated Learning, Fine-tuning, Domain Generalization) based on clinical constraints and then reports observed trends in performance and methodology. No equations, derivations, fitted parameters, predictions, or first-principles results are present. The central claim is an empirical observation drawn from reviewed literature rather than any self-referential construction or reduction to the paper's own inputs. While the representativeness of the sample is an assumption, this does not meet the criteria for circularity, which require explicit reductions such as self-definitional logic, fitted inputs renamed as predictions, or load-bearing self-citation chains. The paper is self-contained as a review against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Survey paper; no new mathematical derivations, free parameters, axioms, or invented entities are introduced beyond standard literature review practices.

pith-pipeline@v0.9.0 · 5770 in / 1102 out tokens · 22500 ms · 2026-05-23T17:59:45.849080+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

257 extracted references · 257 canonical work pages · 5 internal anchors

  1. [1]

    Asmaa Abbas, Mohammed M Abdelsamea, and Mohamed Medhat Gaber. 2020. Detrac: Transfer learning of class decomposed medical images in convolutional neural networks. IEEE Access 8 (2020), 74901–74913

    work page 2020

  2. [2]

    Mohammed Adnan, Shivam Kalra, Jesse C Cresswell, Graham W Taylor, and Hamid R Tizhoosh. 2022. Federated learning and differential privacy for medical image analysis. Scientific Reports 12, 1 (2022), 1953

    work page 2022

  3. [3]

    Manal AlAmir and Manal AlGhamdi. 2022. The Role of generative adversarial network in medical image analysis: An in-depth survey. ACM Computing Surveys (CSUR) 55, 5 (2022), 1–36

    work page 2022

  4. [4]

    Sidra Aleem, Fangyijie Wang, Mayug Maniparambil, Eric Arazo, Julia Dietlmeier, Kathleen Curran, Noel EO’ Connor, and Suzanne Little. 2024. Test-Time Adaptation with SaLIP: A Cascade of SAM and CLIP for Zero-shot Medical Image Segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 5184–5193

    work page 2024

  5. [5]

    Syed Muhammad Anwar, Muhammad Majid, Adnan Qayyum, Muhammad Awais, Majdi Alnowami, and Muham- mad Khurram Khan. 2018. Medical image analysis using convolutional neural networks: a review. Journal of Medical Systems 42 (2018), 1–13. 24 Zixian Su, Jingwei Guo, Xi Yang, Qiufeng Wang, Frans Coenen, and Kaizhu Huang

    work page 2018

  6. [6]

    Danilo Avola, Luigi Cinque, Alessio Fagioli, Gianluca Foresti, and Alessio Mecca. 2021. Ultrasound medical imaging techniques: a survey. ACM Computing Surveys (CSUR) 54, 3 (2021), 1–38

    work page 2021

  7. [7]

    Hritam Basak and Zhaozheng Yin. 2023. Semi-supervised domain adaptive medical image segmentation through consistency regularized disentangled contrastive learning. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 260–270

    work page 2023

  8. [8]

    Mathilde Bateson, Hoel Kervadec, Jose Dolz, Hervé Lombaert, and Ismail Ben Ayed. 2022. Source-free domain adaptation for image segmentation. Medical Image Analysis 82 (2022), 102617

    work page 2022

  9. [9]

    Mathilde Bateson, Herve Lombaert, and Ismail Ben Ayed. 2022. Test-time adaptation with shape moments for image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 736–745

    work page 2022

  10. [10]

    Róger Bermúdez-Chacón, Pablo Márquez-Neila, Mathieu Salzmann, and Pascal Fua. 2018. A domain-adaptive two- stream U-Net for electron microscopy image segmentation. In IEEE International Symposium on Biomedical Imaging . IEEE, 400–404

    work page 2018

  11. [11]

    Xuesheng Bian, Xiongbiao Luo, Cheng Wang, Weiquan Liu, and Xiuhong Lin. 2022. DDA-Net: Unsupervised cross- modality medical image segmentation via dual domain adaptation. Computer Methods and Programs in Biomedicine 213 (2022), 106531

    work page 2022

  12. [12]

    Alexander Bigalke, Lasse Hansen, Jasper Diesel, Carlotta Hennigs, Philipp Rostalski, and Mattias P Heinrich. 2023. Anatomy-guided domain adaptation for 3D in-bed human pose estimation. Medical Image Analysis 89 (2023), 102887

    work page 2023

  13. [13]

    Jinzheng Cai, Zizhao Zhang, Lei Cui, Yefeng Zheng, and Lin Yang. 2019. Towards cross-modal organ translation and segmentation: A cycle-and shape-consistent generative adversarial network. Medical Image Analysis 52 (2019), 174–184

    work page 2019

  14. [14]

    Maria G Baldeon Calisto and Susana K Lai-Yuen. 2022. C-MADA: unsupervised cross-modality adversarial domain adaptation framework for medical image segmentation. In Medical Imaging 2022: Image Processing , Vol. 12032. SPIE, 971–978

    work page 2022

  15. [15]

    Yiming Cao, Lizhen Cui, Lei Zhang, Fuqiang Yu, Zhen Li, and Yonghui Xu. 2023. MMTN: multi-modal memory transformer network for image-report consistent medical report generation. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 37. 277–285

    work page 2023

  16. [16]

    Karen Carrasco, Lenin Tomalá, Eileen Ramírez Meza, Doris Meza Bolaños, and Washington Ramírez Montalvan. 2024. Computational Techniques in PET/CT Image Processing for Breast Cancer: A Systematic Mapping Review. ACM Computing Surveys (CSUR) 56, 8 (2024), 1–38

    work page 2024

  17. [17]

    Alper Emin Cetinkaya, Murat Akin, and Seref Sagiroglu. 2021. Improving performance of federated learning based medical image analysis in non-IID settings using image augmentation. In International Conference on Information Security and Cryptology. IEEE, 69–74

    work page 2021

  18. [18]

    Cheng Chen, Qi Dou, Hao Chen, Jing Qin, and Pheng-Ann Heng. 2019. Synergistic image and feature adaptation: To- wards cross-modality domain adaptation for medical image segmentation. InAAAI Conference on Artificial Intelligence, Vol. 33. 865–872

    work page 2019

  19. [19]

    Cheng Chen, Qi Dou, Hao Chen, Jing Qin, and Pheng Ann Heng. 2020. Unsupervised bidirectional cross-modality adaptation via deeply synergistic image and feature alignment for medical image segmentation. IEEE Transactions on Medical Imaging 39, 7 (2020), 2494–2505

    work page 2020

  20. [20]

    Chen Chen, Kerstin Hammernik, Cheng Ouyang, Chen Qin, Wenjia Bai, and Daniel Rueckert. 2021. Cooperative training and latent space data augmentation for robust medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 149–159

    work page 2021

  21. [21]

    Chen Chen, Zeju Li, Cheng Ouyang, Matthew Sinclair, Wenjia Bai, and Daniel Rueckert. 2022. Maxstyle: Adversarial style composition for robust medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 151–161

    work page 2022

  22. [22]

    Cheng Chen, Quande Liu, Yueming Jin, Qi Dou, and Pheng-Ann Heng. 2021. Source-free domain adaptive fundus image segmentation with denoised pseudo-labeling. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 225–235

    work page 2021

  23. [23]

    Chonglin Chen and Gang Wang. 2021. IOSUDA: An unsupervised domain adaptation with input and output space alignment for joint optic disc and cup segmentation. Applied Intelligence 51 (2021), 3880–3898

    work page 2021

  24. [24]

    Jianguo Chen, Kenli Li, Zhaolei Zhang, Keqin Li, and Philip S Yu. 2021. A survey on applications of artificial intelligence in fighting against COVID-19. ACM Computing Surveys (CSUR) 54, 8 (2021), 1–32

    work page 2021

  25. [25]

    Sihong Chen, Kai Ma, and Yefeng Zheng. 2019. Med3d: Transfer learning for 3d medical image analysis.arXiv preprint arXiv:1904.00625 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  26. [26]

    Xu Chen, Tianshu Kuang, Hannah Deng, Steve H Fung, Jaime Gateno, James J Xia, and Pew-Thian Yap. 2022. Dual adversarial attention mechanism for unsupervised domain adaptive medical image segmentation. IEEE Transactions on Medical Imaging 41, 11 (2022), 3445–3453. Navigating Distribution Shifts in Medical Image Analysis: A Survey 25

    work page 2022

  27. [27]

    Zhen Chen, Meilu Zhu, Chen Yang, and Yixuan Yuan. 2021. Personalized retrogress-resilient framework for real- world medical federated learning. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 347–356

    work page 2021

  28. [28]

    Kuo-Sheng Cheng, Qiong-Wen Zhang, Hung-Wen Tsai, Nien-tsu Li, and Pau-Choo Chung. 2023. Domain-Centroid- Guided Progressive Teacher-based Knowledge Distillation for Source-Free Domain Adaptation of Histopathological Images. IEEE Transactions on Artificial Intelligence (2023)

    work page 2023

  29. [29]

    Boris Chidlovskii, Stephane Clinchant, and Gabriela Csurka. 2016. Domain adaptation in the absence of source domain data. In ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . 451–460

    work page 2016

  30. [30]

    Hyeonwoo Cho, Kazuya Nishimura, Kazuhide Watanabe, and Ryoma Bise. 2022. Effective pseudo-labeling based on heatmap for unsupervised domain adaptation in cell detection. Medical Image Analysis 79 (2022), 102436

    work page 2022

  31. [31]

    Vikash Chouhan, Sanjay Kumar Singh, Aditya Khamparia, Deepak Gupta, Prayag Tiwari, Catarina Moreira, Robertas Damaševičius, and Victor Hugo C De Albuquerque. 2020. A novel transfer learning based approach for pneumonia detection in chest X-ray images. Applied Sciences 10, 2 (2020), 559

    work page 2020

  32. [32]

    Liam Collins, Hamed Hassani, Aryan Mokhtari, and Sanjay Shakkottai. 2021. Exploiting shared representations for personalized federated learning. In International Conference on Machine Learning . PMLR, 2089–2099

    work page 2021

  33. [33]

    Onat Dalmaz, Usama Mirza, Gökberk Elmas, Muzaffer Özbey, Salman UH Dar, and Tolga Çukur. 2022. A specificity- preserving generative model for federated mri translation. In International Workshop on Distributed, Collaborative, and Federated Learning. Springer, 79–88

    work page 2022

  34. [34]

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. 2009. Imagenet: A large-scale hierarchical image database. In IEEE Conference on Computer Vision and Pattern Recognition . Ieee, 248–255

    work page 2009

  35. [35]

    Zhifang Deng, Dandan Li, Shi Tan, Ying Fu, Xueguang Yuan, Xiaohong Huang, Yong Zhang, and Guangwei Zhou

    work page

  36. [36]

    In International Conference on Medical Image Computing and Computer-Assisted Intervention

    FedGrav: An Adaptive Federated Aggregation Algorithm for Multi-institutional Medical Image Segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 170–180

    work page

  37. [37]

    Weiping Ding, Mohamed Abdel-Basset, Hossam Hawash, Mahardhika Pratama, and Witold Pedrycz. 2023. Generaliz- able segmentation of COVID-19 infection from multi-site tomography scans: a federated learning framework. IEEE Transactions on Emerging Topics in Computational Intelligence (2023)

    work page 2023

  38. [38]

    Jiahua Dong, Yang Cong, Gan Sun, Bineng Zhong, and Xiaowei Xu. 2020. What can be transferred: Unsupervised domain adaptation for endoscopic lesions segmentation. InIEEE Conference on Computer Vision and Pattern Recognition. 4023–4032

    work page 2020

  39. [39]

    Qi Dou, Daniel Coelho de Castro, Konstantinos Kamnitsas, and Ben Glocker. 2019. Domain generalization via model-agnostic learning of semantic features. Advances in Neural Information Processing Systems 32 (2019)

    work page 2019

  40. [40]

    Qi Dou, Cheng Ouyang, Cheng Chen, Hao Chen, Ben Glocker, Xiahai Zhuang, and Pheng-Ann Heng. 2019. PnP- AdaNet: Plug-and-play adversarial domain adaptation network at unpaired cross-modality cardiac segmentation. IEEE Access 7 (2019), 99065–99076

    work page 2019

  41. [41]

    Qi Dou, Tiffany Y So, Meirui Jiang, Quande Liu, Varut Vardhanabhuti, Georgios Kaissis, Zeju Li, Weixin Si, Heather HC Lee, Kevin Yu, et al. 2021. Federated deep learning for detecting COVID-19 lung abnormalities in CT: a privacy- preserving multinational validation study. NPJ Digital Medicine 4, 1 (2021), 60

    work page 2021

  42. [42]

    Xiuquan Du and Yueguo Liu. 2021. Constraint-based unsupervised domain adaptation network for multi-modality cardiac image segmentation. IEEE Journal of Biomedical and Health Informatics 26, 1 (2021), 67–78

    work page 2021

  43. [43]

    James S Duncan and Nicholas Ayache. 2000. Medical image analysis: Progress over two decades and the challenges ahead. IEEE Transactions on Pattern Analysis and Machine Intelligence 22, 1 (2000), 85–106

    work page 2000

  44. [44]

    Gokberk Elmas, Salman UH Dar, Yilmaz Korkmaz, Emir Ceyani, Burak Susam, Muzaffer Ozbey, Salman Avestimehr, and Tolga Çukur. 2022. Federated learning of generative image priors for MRI reconstruction. IEEE Transactions on Medical Imaging (2022)

    work page 2022

  45. [45]

    Yuqi Fang, Mingliang Wang, Guy G Potter, and Mingxia Liu. 2023. Unsupervised cross-domain functional MRI adaptation for automated major depressive disorder identification. Medical Image Analysis 84 (2023), 102707

    work page 2023

  46. [46]

    Ines Feki, Sourour Ammar, Yousri Kessentini, and Khan Muhammad. 2021. Federated learning for COVID-19 screening from Chest X-ray images. Applied Soft Computing 106 (2021), 107330

    work page 2021

  47. [47]

    Wei Feng, Lie Ju, Lin Wang, Kaimin Song, Xin Zhao, and Zongyuan Ge. 2023. Unsupervised domain adaptation for medical image segmentation by selective entropy constraints and adaptive semantic alignment. In AAAI Conference on Artificial Intelligence, Vol. 37. 623–631

    work page 2023

  48. [48]

    Yuanyi Feng, Yuemei Luo, and Jianfei Yang. 2023. Cross-platform privacy-preserving CT image COVID-19 diagnosis based on source-free domain adaptation. Knowledge-Based Systems 264 (2023), 110324

    work page 2023

  49. [49]

    Jingru Fu, Simone Bendazzoli, Örjan Smedby, and Rodrigo Moreno. 2024. Unsupervised Domain Adaptation for Pediatric Brain Tumor Segmentation. arXiv preprint arXiv:2406.16848 (2024)

    work page arXiv 2024

  50. [50]

    Jun Gao, Qicheng Lao, Qingbo Kang, Paul Liu, Le Zhang, and Kang Li. 2022. Unsupervised cross-disease domain adaptation by lesion scale matching. In International Conference on Medical Image Computing and Computer-Assisted 26 Zixian Su, Jingwei Guo, Xi Yang, Qiufeng Wang, Frans Coenen, and Kaizhu Huang Intervention. Springer, 660–670

    work page 2022

  51. [51]

    Negin Ghamsarian, Javier Gamazo Tejero, Pablo Márquez-Neila, Sebastian Wolf, Martin Zinkernagel, Klaus Schoeff- mann, and Raphael Sznitman. 2023. Domain adaptation for medical image segmentation using transformation- invariant self-training. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 331–341

    work page 2023

  52. [52]

    Camila González, Amin Ranem, Daniel Pinto dos Santos, Ahmed Othman, and Anirban Mukhopadhyay. 2023. Lifelong nnU-Net: a framework for standardized medical continual learning. Scientific Reports 13, 1 (2023), 9381

    work page 2023

  53. [53]

    Karthik Gopinath, Christian Desrosiers, and Herve Lombaert. 2020. Graph domain adaptation for alignment-invariant brain surface segmentation. In MICCAI Workshop on Uncertainty for Safe Utilization of Machine Learning in Medical Imaging. Springer, 152–163

    work page 2020

  54. [54]

    Ophir Gozes, Maayan Frid-Adar, Hayit Greenspan, Patrick D Browning, Huangqi Zhang, Wenbin Ji, Adam Bernheim, and Eliot Siegel. 2020. Rapid ai development cycle for the coronavirus (covid-19) pandemic: Initial results for automated detection & patient monitoring using deep learning ct image analysis. arXiv preprint arXiv:2003.05037 (2020)

    work page arXiv 2020

  55. [55]

    Hao Guan and Mingxia Liu. 2021. Domain adaptation for medical image analysis: a survey. IEEE Transactions on Biomedical Engineering 69, 3 (2021), 1173–1185

    work page 2021

  56. [56]

    Gozde N Gunesli, Mohsin Bilal, Shan E Ahmed Raza, and Nasir M Rajpoot. 2023. A Federated Learning Approach to Tumor Detection in Colon Histology Images. Journal of Medical Systems 47, 1 (2023), 99

    work page 2023

  57. [57]

    Jingwei Guo, Kaizhu Huang, Xinping Yi, and Rui Zhang. 2022. Learning disentangled graph convolutional networks locally and globally. IEEE Transactions on Neural Networks and Learning Systems 35, 3 (2022), 3640–3651

    work page 2022

  58. [58]

    Jingwei Guo, Kaizhu Huang, Rui Zhang, and Xinping Yi. 2024. ES-GNN: Generalizing graph neural networks beyond homophily with edge splitting. IEEE Transactions on Pattern Analysis and Machine Intelligence (2024)

    work page 2024

  59. [59]

    Jingwei Guo, Yikang Yang, Qingwei Wu, Jionglong Su, and Fei Ma. 2016. Adaptive active contour model based automatic tongue image segmentation. In 2016 9th International Congress on image and signal processing, BioMedical engineering and informatics (CISP-BMEI) . IEEE, 1386–1390

    work page 2016

  60. [60]

    Pengfei Guo, Puyang Wang, Jinyuan Zhou, Shanshan Jiang, and Vishal M Patel. 2021. Multi-institutional collaborations for improving deep learning-based magnetic resonance image reconstruction using federated learning. In IEEE Conference on Computer Vision and Pttern Recognition . 2423–2432

    work page 2021

  61. [61]

    Changhee Han, Leonardo Rundo, Kohei Murao, Tomoyuki Noguchi, Yuki Shimahara, Zoltán Ádám Milacski, Saori Koshino, Evis Sala, Hideki Nakayama, and Shin’ichi Satoh. 2021. MADGAN: Unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction. BMC Bioinformatics 22, 2 (2021), 1–20

    work page 2021

  62. [62]

    Yufan He, Aaron Carass, Lianrui Zuo, Blake E Dewey, and Jerry L Prince. 2021. Autoencoder based self-supervised test-time adaptation for medical image analysis. Medical Image Analysis 72 (2021), 102136

    work page 2021

  63. [63]

    Robert Hertel and Rachid Benlamri. 2023. Deep learning techniques for COVID-19 diagnosis and prognosis based on radiological imaging. ACM Computing Surveys (CSUR) 55, 12 (2023), 1–39

    work page 2023

  64. [64]

    Jin Hong, Simon Chun-Ho Yu, and Weitian Chen. 2022. Unsupervised domain adaptation for cross-modality liver segmentation via joint adversarial learning and self-learning. Applied Soft Computing 121 (2022), 108729

    work page 2022

  65. [65]

    Jin Hong, Yu-Dong Zhang, and Weitian Chen. 2022. Source-free unsupervised domain adaptation for cross-modality abdominal multi-organ segmentation. Knowledge-Based Systems 250 (2022), 109155

    work page 2022

  66. [66]

    Timothy Hospedales, Antreas Antoniou, Paul Micaelli, and Amos Storkey. 2021. Meta-learning in neural networks: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 9 (2021), 5149–5169

    work page 2021

  67. [67]

    Md Imran Hossain, Ghada Zamzmi, Peter R Mouton, Md Sirajus Salekin, Yu Sun, and Dmitry Goldgof. 2023. Explainable AI for Medical Data: Current Methods, Limitations, and Future Directions. ACM Computing Surveys (CSUR) (2023)

    work page 2023

  68. [68]

    S Maryam Hosseini, Milad Sikaroudi, Morteza Babaie, and HR Tizhoosh. 2023. Proportionally fair hospital collabora- tions in federated learning of histopathology images. IEEE Transactions on Medical Imaging (2023)

    work page 2023

  69. [69]

    Ehsan Hosseini-Asl, Robert Keynton, and Ayman El-Baz. 2016. Alzheimer’s disease diagnostics by adaptation of 3D convolutional network. In IEEE International Conference on Image Processing . IEEE, 126–130

    work page 2016

  70. [70]

    Minhao Hu, Tao Song, Yujun Gu, Xiangde Luo, Jieneng Chen, Yinan Chen, Ya Zhang, and Shaoting Zhang. 2021. Fully test-time adaptation for image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 251–260

    work page 2021

  71. [71]

    Qingqiao Hu, Hongwei Li, and Jianguo Zhang. 2022. Domain-adaptive 3D medical image synthesis: An efficient unsupervised approach. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 495–504

    work page 2022

  72. [72]

    Shishuai Hu, Zehui Liao, and Yong Xia. 2022. Domain specific convolution and high frequency reconstruction based unsupervised domain adaptation for medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, 650–659

    work page 2022

  73. [73]

    Shishuai Hu, Zehui Liao, and Yong Xia. 2022. Prosfda: Prompt learning based source-free domain adaptation for medical image segmentation. arXiv preprint arXiv:2211.11514 (2022). Navigating Distribution Shifts in Medical Image Analysis: A Survey 27

    work page arXiv 2022

  74. [74]

    Shishuai Hu, Zehui Liao, and Yong Xia. 2023. Devil is in channels: Contrastive single domain generalization for medical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention . Springer, 14–23

    work page 2023

  75. [75]

    Kaizhu Huang, Amir Hussain, Qiu-Feng Wang, and Rui Zhang. 2019. Deep learning: fundamentals, theory and applications. Vol. 2. Springer

    work page 2019

  76. [76]

    Xiaoqiong Huang, Xin Yang, Haoran Dou, Yuhao Huang, Li Zhang, Zhendong Liu, Zhongnuo Yan, Lian Liu, Yuxin Zou, Xindi Hu, et al. 2023. Test-time bi-directional adaptation between image and model for robust segmentation. Computer Methods and Programs in Biomedicine 233 (2023), 107477

    work page 2023

  77. [77]

    V Durga Prasad Jasti, Abu Sarwar Zamani, K Arumugam, Mohd Naved, Harikumar Pallathadka, F Sammy, Abhishek Raghuvanshi, and Karthikeyan Kaliyaperumal. 2022. Computational technique based on machine learning and image processing for medical image analysis of breast cancer diagnosis. Security and Communication Networks 2022, 1 (2022), 1918379

    work page 2022

  78. [78]

    Jue Jiang and Harini Veeraraghavan. 2020. Unified Cross-Modality Feature Disentangler for Unsupervised Multi- domain MRI Abdomen Organs Segmentation. In International Conference on Medical Image Computing and Computer- Assisted Intervention. 347–358

    work page 2020

  79. [79]

    Kaida Jiang, Li Quan, and Tao Gong. 2022. Disentangled representation and cross-modality image translation based unsupervised domain adaptation method for abdominal organ segmentation. International Journal of Computer Assisted Radiology and Surgery 17, 6 (2022), 1101–1113

    work page 2022

  80. [80]

    Meirui Jiang, Zirui Wang, and Qi Dou. 2022. Harmofl: Harmonizing local and global drifts in federated learning on heterogeneous medical images. In AAAI Conference on Artificial Intelligence, Vol. 36. 1087–1095

    work page 2022

Showing first 80 references.