mlsanityData to 13 Oct 2026

Breaking the Dilemma of Medical Image-to-image Translation

arXiv 2110.06465 ↗First posted 13 Oct 2021eess.IV cs.CVcited by 82

Not on the map. Why?

Abstract (arXiv)

Supervised Pix2Pix and unsupervised Cycle-consistency are two modes that dominate the field of medical image-to-image translation. However, neither modes are ideal. The Pix2Pix mode has excellent performance. But it requires paired and well pixel-wise aligned images, which may not always be achievable due to respiratory motion or anatomy change between times that paired images are acquired. The Cycle-consistency mode is less stringent with training data and works well on unpaired or misaligned images. But its performance may not be optimal. In order to break the dilemma of the existing modes, we propose a new unsupervised mode called RegGAN for medical image-to-image translation. It is based on the theory of "loss-correction". In RegGAN, the misaligned target images are considered as noisy labels and the generator is trained with an additional registration network to fit the misaligned noise distribution adaptively. The goal is to search for the common optimal solution to both image-to-image translation and registration tasks. We incorporated RegGAN into a few state-of-the-art image-to-image translation methods and demonstrated that RegGAN could be easily combined with these methods to improve their performances. Such as a simple CycleGAN in our mode surpasses latest NICEGAN even though using less network parameters. Based on our results, RegGAN outperformed both Pix2Pix on aligned data and Cycle-consistency on misaligned or unpaired data. RegGAN is insensitive to noises which makes it a better choice for a wide range of scenarios, especially for medical image-to-image translation tasks in which well pixel-wise aligned data are not available

Key facts. Cited by 82 papers in the map (data through 13 Oct 2026). Builds on 47 of its 73 references that are in the map. First posted on arXiv 13 Oct 2021.

Builds on

47 of 73 references.

References · 47

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,656
  2. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,516
  3. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  4. 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,079
  5. 2018Multimodal Unsupervised Image-to-Image Translationcited by 795
  6. 2020DivideMix: Learning with Noisy Labels as Semi-supervised Learningcited by 692
  7. 2016Making Deep Neural Networks Robust to Label Noise: a Loss Correction Approachcited by 663
  8. 2018VoxelMorph: A Learning Framework for Deformable Medical Image Registrationcited by 657
  9. 2017Unsupervised Image-to-Image Translation Networkscited by 602
  10. 2018Joint Optimization Framework for Learning with Noisy Labelscited by 293
  11. 2017Learning to Discover Cross-Domain Relations with Generative Adversarial Networkscited by 278
  12. 2019Are Anchor Points Really Indispensable in Label-Noise Learning?cited by 208
35 more
  1. 2019U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translationcited by 186
  2. 2017Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identificationcited by 168
  3. 2017Learning from Noisy Labels with Distillationcited by 160
  4. 2018Learning to Learn from Noisy Labeled Datacited by 160
  5. 2020Dual T: Reducing Estimation Error for Transition Matrix in Label-noise Learningcited by 142
  6. 2015A Light CNN for Deep Face Representation with Noisy Labelscited by 126
  7. 2019Learning From Noisy Labels By Regularized Estimation Of Annotator Confusioncited by 126
  8. 2020Self-Adaptive Training: beyond Empirical Risk Minimizationcited by 126
  9. 2019Learning a Deep ConvNet for Multi-label Classification with Partial Labelscited by 124
  10. 2019Deep Self-Learning From Noisy Labelscited by 122
  11. 2019Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentationcited by 121
  12. 2020Cross-domain Correspondence Learning for Exemplar-based Image Translationcited by 116
  13. 2017Learning From Noisy Large-Scale Datasets With Minimal Supervisioncited by 115
  14. 2020Fast Symmetric Diffeomorphic Image Registration with Convolutional Neural Networkscited by 107
  15. 2017Deep learning from crowdscited by 105
  16. 2019Cycle-Consistency for Robust Visual Question Answeringcited by 105
  17. 2020Error-Bounded Correction of Noisy Labelscited by 74
  18. 2020Unsupervised Multi-Modal Image Registration via Geometry Preserving Image-to-Image Translationcited by 66
  19. 2015Seeing through the Human Reporting Bias: Visual Classifiers from Noisy Human-Centric Labelscited by 64
  20. 2020Reusing Discriminators for Encoding: Towards Unsupervised Image-to-Image Translationcited by 60
  21. 2020Disentangling Human Error from the Ground Truth in Segmentation of Medical Imagescited by 58
  22. 2019Unsupervised Deformable Registration for Multi-Modal Images via Disentangled Representationscited by 52
  23. 2019Networks for Joint Affine and Non-parametric Image Registrationcited by 44
  24. 2020DeepFLASH: An Efficient Network for Learning-based Medical Image Registrationcited by 37
  25. 2019CANZSL: Cycle-Consistent Adversarial Networks for Zero-Shot Learning from Natural Languagecited by 23
  26. 2020Meta Soft Label Generation for Noisy Labelscited by 21
  27. 2019Cycle-consistent Conditional Adversarial Transfer Networkscited by 15
  28. 2018Hyperspectral Image Classification in the Presence of Noisy Labelscited by 12
  29. 2019Visualizing the Consequences of Climate Change Using Cycle-Consistent Adversarial Networkscited by 12
  30. 2020DeepI2I: Enabling Deep Hierarchical Image-to-Image Translation by Transferring from GANscited by 11
  31. 2018Unsupervised Image-to-Image Translation with Stacked Cycle-Consistent Adversarial Networkscited by 10
  32. 2019Photometric Transformer Networks and Label Adjustment for Breast Density Predictioncited by 10
  33. 2019Optimal Transport driven CycleGAN for Unsupervised Learning in Inverse Problemscited by 4
  34. 2019Cycle-Consistent Adversarial Networks for Realistic Pervasive Change Generation in Remote Sensing Imagerycited by 2
  35. 2020Multi-Cycle-Consistent Adversarial Networks for CT Image Denoisingcited by 1

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2022: 14; 2023: 15; 2024: 21; 2025: 20; 2026: 12.

82 papers.

Sort:

By area:

  1. 2023Adaptive Latent Diffusion Model for 3D Medical Image to Image Translation: Multi-modal Magnetic Resonance Imaging Studycited by 64
  2. 2022BCI: Breast Cancer Immunohistochemical Image Generation through Pyramid Pix2pixcited by 61
  3. 2023Zero-shot Medical Image Translation via Frequency-Guided Diffusion Modelscited by 40
  4. 2022Class-Aware Adversarial Transformers for Medical Image Segmentationcited by 33
  5. 2024Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image Generationcited by 33
  6. 2024Cross-conditioned Diffusion Model for Medical Image to Image Translationcited by 26
  7. 2022FedMed-GAN: Federated Domain Translation on Unsupervised Cross-Modality Brain Image Synthesiscited by 21
  8. 2023Low-field magnetic resonance image enhancement via stochastic image quality transfercited by 19
  9. 2022Unsupervised Multi-Modal Medical Image Registration via Discriminator-Free Image-to-Image Translationcited by 11
  10. 2024StainDiffuser: MultiTask Dual Diffusion Model for Virtual Stainingcited by 11
  11. 2023DDMM-Synth: A Denoising Diffusion Model for Cross-modal Medical Image Synthesis with Sparse-view Measurement Embeddingcited by 10
  12. 2024Virtual birefringence imaging and histological staining of amyloid deposits in label-free tissue using autofluorescence microscopy and deep learningcited by 8
38 more
  1. 2025Regression is all you need for medical image translationcited by 8
  2. 2022Cross-Modality Neuroimage Synthesis: A Surveycited by 7
  3. 2024HEMIT: H&E to Multiplex-immunohistochemistry Image Translation with Dual-Branch Pix2pix Generatorcited by 7
  4. 2022DC-cycleGAN: Bidirectional CT-to-MR Synthesis from Unpaired Datacited by 6
  5. 2022Learning Single Image Defocus Deblurring with Misaligned Training Pairscited by 6
  6. 2023FDDM: Unsupervised Medical Image Translation with a Frequency-Decoupled Diffusion Modelcited by 6
  7. 2024Generative Medical Segmentationcited by 6
  8. 2024Non-Invasive to Invasive: Enhancing FFA Synthesis from CFP with a Benchmark Dataset and a Novel Networkcited by 6
  9. 2022Deformation equivariant cross-modality image synthesis with paired non-aligned training datacited by 5
  10. 2023Indescribable Multi-modal Spatial Evaluatorcited by 5
  11. 2022Enhancing Low-Light Images in Real World via Cross-Image Disentanglementcited by 4
  12. 2025DVG-Diffusion: Dual-View Guided Diffusion Model for CT Reconstruction from X-Rayscited by 4
  13. 2023fRegGAN with K-space Loss Regularization for Medical Image Translationcited by 3
  14. 2024CT to PET Translation: A Large-scale Dataset and Domain-Knowledge-Guided Diffusion Approachcited by 3
  15. 2024Towards a Universal 3D Medical Multi-modality Generalization via Learning Personalized Invariant Representationcited by 3
  16. 2022FedMed-ATL: Misaligned Unpaired Brain Image Synthesis via Affine Transform Losscited by 2
  17. 2022Classical and learned MR to pseudo-CT mappings for accurate transcranial ultrasound simulationcited by 2
  18. 2023An Attentive-based Generative Model for Medical Image Synthesiscited by 2
  19. 2024ContourDiff: Unpaired Medical Image Translation with Structural Consistencycited by 2
  20. 2024CHAIN: Enhancing Generalization in Data-Efficient GANs via lipsCHitz continuity constrAIned Normalizationcited by 2
  21. 2024Similarity-aware Syncretic Latent Diffusion Model for Medical Image Translation with Representation Learningcited by 2
  22. 2024Multi-modal MRI Translation via Evidential Regression and Distribution Calibrationcited by 2
  23. 2024Boosting Medical Image Synthesis via Registration-guided Consistency and Disentanglement Learningcited by 2
  24. 2024Deformation-aware GAN for Medical Image Synthesis with Substantially Misaligned Pairscited by 2
  25. 2024CAS-GAN for Contrast-free Angiography Synthesiscited by 2
  26. 2025Digital Staining with Knowledge Distillation: A Unified Framework for Unpaired and Paired-But-Misaligned Datacited by 2
  27. 2025Generative AI for Misalignment-Resistant Virtual Staining to Accelerate Histopathology Workflowscited by 2
  28. 2023Translating Simulation Images to X-ray Images via Multi-Scale Semantic Matchingcited by 1
  29. 2023Vista-Morph: Unsupervised Image Registration of Visible-Thermal Facial Pairscited by 1
  30. 2023Jurassic World Remake: Bringing Ancient Fossils Back to Life via Zero-Shot Long Image-to-Image Translationcited by 1
  31. 2023AGMDT: Virtual Staining of Renal Histology Images with Adjacency-Guided Multi-Domain Transfercited by 1
  32. 2024UWAFA-GAN: Ultra-Wide-Angle Fluorescein Angiography Transformation via Multi-scale Generation and Registration Enhancementcited by 1
  33. 2024Reliable Multi-modal Medical Image-to-image Translation Independent of Pixel-wise Aligned Datacited by 1
  34. 2025Hyperbolic Cycle Alignment for Infrared-Visible Image Fusioncited by 1
  35. 2025Deep learning-enabled virtual multiplexed immunostaining of label-free tissue for vascular invasion assessmentcited by 1
  36. 2025Tumor Synthesis conditioned on Radiomicscited by 1
  37. 2025FS-RWKV: Leveraging Frequency Spatial-Aware RWKV for 3T-to-7T MRI Translationcited by 1
  38. 2025PADM: A Physics-aware Diffusion Model for Attenuation Correctioncited by 1

50 most cited of 82 All 82 (JSON)