Pith. sign in

REVIEW 5 major objections 6 minor 64 references

AURAD: Anatomy-Pathology Unified Radiology Synthesis with Progressive Representations

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read AURAD claims that generating disease masks before chest X-rays produces images that look real and train better detectors.

desk verdict Real contribution in a two-stage CXR synthesis pipeline, but the headline pathology-accuracy numbers rest on a circular filter/evaluator pair; the downstream results are the more trustworthy evidence. read the letter →

arxiv 2509.04819 v2 pith:YIZVXJGK submitted 2025-09-05 eess.IV cs.CV

classification eess.IVcs.CV
keywords chestX-raysynthesistext-to-maskgenerationdiffusionmodelsControlNetanatomy-pathologyconsistencysemanticsegmentationmasksmedicalimageaugmentationdownstreamdetection
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper proposes AURAD, a two-stage framework for synthesizing chest X-rays together with pixel-level disease masks, and argues that it is the first approach to explicitly model where pulmonary diseases sit relative to anatomical structures. The idea is that a clinical text prompt is first turned into a dense mask recording disease class, location, and severity, and that mask then guides a diffusion model to render the X-ray. Because the mask is produced before the image, the pipeline gains fine-grained spatial control and, as a side product, provides free training labels. The paper reports that augmenting real chest X-ray data with one round of AURAD images improves detection on three test sets and classification on a fourth, and that board-certified radiologists rated 78% of synthesized images as realistic. A careful reader would care because the claim, if correct, offers a scalable substitute for expensive manual annotation in medical imaging.

What carries the argument

The load-bearing object is the pathology-augmented semantic mask: a dense map that fuses organ anatomy with disease regions, annotated by class, location (six lung zones plus heart and mediastinum), and severity. Stage one adapts an instruction-tuned image-editing diffusion model to transform an organ-only mask into a pathology-augmented mask using a natural-language disease instruction, while a prompt tool derives standardized text from real masks by measuring overlap between disease and organ regions. Stage two injects that mask into a frozen domain-specific chest X-ray diffusion backbone through a ControlNet-style conditioning module with zero-convolution layers, enforcing pixel-level spatial control while preserving image quality. The self-assessment loop, which re-captions masks, checks realism with a discriminator, and verifies disease presence with a pretrained multi-label classifier, is the mechanism that tries to keep the progressive pipeline from accumulating errors.

What would settle it

Score AURAD's output with an independent pathology classifier trained on a different dataset, or have radiologists re-label the synthesized images; if pathology accuracy and the filter's acceptance rate drop to chance under independent scoring, the diagnostic claims are an artifact of the evaluation model. A second check: replace the learned masks with random organ-constrained blobs; if downstream gains match, the learned pathology–anatomy geometry is not the active ingredient.

Watch

Extended reading notes

Core claim

AURAD's central claim is that chest X-ray synthesis should be decomposed into two progressive representations: first a text-to-mask stage that predicts a dense, multi-disease segmentation mask conditioned on anatomical organ masks and a structured clinical prompt, then a mask-to-image stage that renders the radiograph from that mask using a mask-conditioned, domain-specific diffusion model. The masks are not random samples; they are learned with pathology–anatomy consistency, so a disease such as effusion appears in the lung base and cardiomegaly follows the heart silhouette. The same masks double as training labels, and the pipeline adds a self-assessment filter that discards samples failing prompt-match, realism, or diagnostic checks. If the paper is right, paired image-mask synthesis of this kind is what makes generated data useful for downstream detection and segmentation, not just visually plausible.

Load-bearing premise

The whole pipeline assumes the pretrained chest X-ray models used to filter and evaluate synthetic images judge them the way a clinician would; if those models are biased, the reported pathology accuracy could simply reflect that bias.

Editorial extensions

If this is right

  • Augmenting real training data with one round of AURAD synthetic images raises detection mAP@50 by +3.28 on ChestX-Det, +1.40 on VinDr-CXR, and +5.02 on ChestX-ray14, and classification AUC by +3.78 on MIMIC-CXR.
  • The jointly generated masks give detection and segmentation models free supervision, so the framework can reduce manual annotation cost.
  • Targeted generation for a long-tail disease lifts that class's mAP@50 from 43.60 to 48.08, so the method can address class imbalance.
  • Under extreme data scarcity (1% real data), adding 5x synthetic samples raises mAP@50 from 7.61% to 25.21%, a roughly three-fold improvement.
  • Because the intermediate mask is interpretable, each synthesized finding can be traced to a disease class, location, and severity, enabling counterfactual and prompt-controlled generation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper, an independent check of the pathology metric matters most: the same pretrained classifier that filters generated images also scores their pathology, so a reader should ask whether that classifier is rewarding images that fit its own biases rather than clinical truth.
  • Beyond the paper, the two-stage design implies that organ segmentation quality is the true bottleneck, since errors in the anatomical mask propagate into both the generated disease mask and the final image, as the paper's own failure cases suggest.
  • Beyond the paper, a plausible control experiment would compare AURAD masks against random disease blobs constrained to organ regions; if the control matches the downstream gains, the learned pathology–anatomy geometry is not the active ingredient.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper proposes AURAD, a two-stage chest X-ray synthesis framework that first generates dense multi-disease masks from clinical prompts conditioned on anatomical structures (text-to-mask with InstructPix2Pix), then renders images from these masks (mask-to-image with a ControlNet-equipped RoentGen model). The pipeline includes a self-assessment mechanism with prompt-match validation, a realism discriminator, and a pretrained diagnostic classifier. The authors claim this is the first framework to explicitly model spatial relationships between pulmonary diseases and anatomical structures, and they evaluate the method with image-quality metrics (FID, CLIP score, MS-SSIM, Path AUC), downstream detection/segmentation/classification augmentation experiments across five datasets, ablations, and a radiologist study reporting 78% realism and 41% clinically useful segmentation overlays.

Significance. If the results hold, AURAD is a useful contribution: it produces paired synthetic images and dense mask labels, supports controllable multi-disease generation, and shows non-trivial downstream gains under data scarcity and long-tail conditions. Strengths include the clearly described two-stage design, the use of anatomical priors, the breadth of downstream benchmarks (ChestX-Det, VinDr-CXR, ChestX-ray14, MIMIC-CXR), the low-data and long-tail experiments, the radiologist study, and the detailed appendices with implementation details, free-form prompt robustness, and comparisons of alternative mask-generation backbones. However, several load-bearing evaluation choices currently undermine the central claims: the Path AUC metric is computed with the same classifier family used for filtering, and the main quantitative tables lack variance estimates. These issues are fixable within a revision.

major comments (5)
  1. [§4.2, §5.1, Table 2] The Diagnostic Reliability filter in §4.2 uses a pretrained multi-label CXR classifier from TorchXRayVision [53], and the Path AUC metric in §5.1 and Table 2 is computed with the same TorchXRayVision pretrained model family [59]. Because the AURAD samples are filtered to retain images that this classifier recognizes as containing the requested disease, while the 'Real' baseline is unfiltered, the reported improvement of +1.67 in Path AUC may reflect selection bias rather than genuine pathological or anatomical fidelity. Please report Path AUC on unfiltered AURAD samples, on a real sample that has passed the same filter, and with an independent evaluator (e.g., a different pretrained classifier or a radiologist read) so that the filter's effect is separated from the synthesis quality.
  2. [Tables 2–5] No confidence intervals, error bars, or significance tests are reported for any of the quantitative comparisons. The per-disease results in Table 3 are visibly inconsistent (e.g., on VinDr-CXR the AURAD-augmented model is worse than real-only for Atelectasis and Effusion but better for Consolidation and Pneumothorax), so the headline overall gains (+3.28, +1.40, +5.02 mAP@50; +1.33 Dice; +3.78 AUC) cannot be distinguished from run-to-run noise without multiple seeds or significance testing. Please provide variance estimates, and specify the exact number of synthetic samples used for the 1× augmentation in each experiment.
  3. [§5.3, Tables 2–5] XReal [14] is introduced as a main comparison method and discussed in the qualitative evaluation, but it is absent from all quantitative tables (Tables 2–5). Since XReal is a spatial-conditioning baseline, its omission from the numeric comparisons weakens the claim that AURAD provides superior fine-grained spatial control. Please include XReal in at least the FID/Path AUC table and the downstream augmentation comparison, or justify its exclusion quantitatively.
  4. [§5.2] The segmentation-helpfulness task evaluates masks predicted by a downstream Mask R-CNN trained on synthetic data, not the masks generated by AURAD. The text states that the radiologists 'assessed these model-predicted masks rather than the raw synthetic masks generated by our framework.' Consequently, the 41% usefulness figure does not directly validate AURAD's generated masks or its anatomy–pathology consistency claim; it validates the downstream model's detections. Please add a radiologist read of the generated masks themselves, or explicitly reframe the claim as an evaluation of downstream predictions.
  5. [§3.2.1, §4.2] The Prompt-Match Validation re-captions generated masks with the same rule-based prompt tool (Algorithm 1) that was used to create the training prompts, so it verifies self-consistency with the prompt template rather than clinical plausibility or anatomical correctness. This is a weaker guarantee than the paper's wording suggests ('anatomical consistency' in §4.2). Please either provide an external validation of mask correctness (e.g., overlap with expert segmentations on a held-out set) or soften the claim to self-consistency.
minor comments (6)
  1. [Appendix H, Table 16] The text reports Fleiss' κ = 0.53 but Table 16 reports κ = 0.50; please reconcile these values.
  2. [Appendix H, Eq. (1)] Equation (1) appears to be the formula for ICC(1,1), not ICC(2,1) as stated; please verify the formula and the reported p-values.
  3. [Table 6] The header 'Real Mask Syn. Mask Expert filter' is ambiguous; clarify which rows correspond to ground-truth masks versus synthesized masks, and note whether the 'Real Mask' row uses ground-truth masks with no expert filter.
  4. [§5.1] The paper states that classification AUC is averaged over five diseases; please list explicitly which five diseases are used for Path AUC and confirm they match the five diseases in Tables 3–5.
  5. [Appendix D, Table 12] The free-form rewrites were generated by GPT and may not reflect true radiologist-authored reports; this should be stated in the main text as a limitation of the free-form comparison.
  6. [§5.2] The realism study used only 20 patients; please report the number of images per patient and any exclusion criteria so the reader can gauge the precision of the 78% realism estimate.

Circularity Check

2 steps flagged · score 6.0 of 10

Path AUC is scored by the same TorchXRayVision classifier used as the acceptance filter, and prompt-match validation reuses the prompt tool as its own judge.

  1. fitted input called prediction [Section 4.2 (Diagnostic Reliability filter) vs. Section 5.1 (Path. AUC metric); Appendix B]
    "Diagnostic Reliability: A pre-trained multi-label CXR classifier [53] verifies the presence and anatomical consistency of pathologies in the generated images."

    The acceptance filter and the reported pathology metric are implemented with the same TorchXRayVision pretrained classifier. Section 5.1 states that 'Diagnostic accuracy is evaluated through multi-label disease classification using a pretrained model from TorchXRayVision [59]', and Appendix B identifies the classification weights as 'densenet121-res224-all' from TorchXRayVision. Generated images are retained only when this classifier recognizes the requested disease; Table 2 then scores the retained set (Ours 59.31) against an unfiltered real baseline (57.64). The +1.67 Path AUC gain is therefore induced by the selection rule rather than by an independent measurement of pathology fidelity.

  2. self definitional [Section 4.2 Prompt-Match Validation; Section 4.1/Appendix C prompt tool]
    "Each generated mask is re-captioned using the prompt tool and compared against the original input. Mismatched samples are discarded and regenerated."

    The 'prompt tool' is the same rule-based converter (Algorithm 1) used in Section 4.1 to construct the training prompts from pathology-mask/organ-mask overlap areas. Re-captioning a generated mask with this tool and requiring a match selects exactly those masks whose overlap statistics reproduce the input template; this is self-consistency with the training objective by construction. It does not provide independent evidence of clinical or anatomical plausibility, yet the paper presents Prompt-Match Validation as part of the self-assessment mechanism ensuring clinically meaningful outputs.

full rationale

Two evaluation loops are closed. First, the Path AUC evidence in Table 2 is contaminated: the diagnostic filter (Section 4.2) and the pathology evaluator (Section 5.1) share the same TorchXRayVision 'densenet121-res224-all' model, and the synthesized-image score is computed on the classifier-selected set while the real baseline is unfiltered. Second, Prompt-Match Validation is circular by construction because the rule-based prompt tool is both the label generator and the validator. These issues affect the paper's specific pathology-accuracy and prompt-alignment claims, including the ablations in Tables 6 and 7 that use Path AUC. The downstream detection and segmentation gains (Tables 3-5), the data-efficiency and long-tail experiments, and the blinded radiologist study use separate, independent evaluation pipelines, so the circularity is partial rather than total. No load-bearing self-citation or imported uniqueness theorem was found; the core synthesis pipeline is externally grounded in RoentGen, InstructPix2Pix, and ControlNet.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim rests on external support: public chest X-ray masks (ChestX-Det, CANDID-PTX), pretrained anatomy segmentation and classification models from TorchXRayVision, the RoentGen backbone, and the InstructPix2Pix/ControlNet modules. None of these are shipped by the paper. The hand-chosen severity thresholds and filter acceptance criteria are the main free parameters. No new physical or conceptual entities are introduced. The most fragile external dependence is the pretrained classifier, which both filters and evaluates outputs.

free parameters (2)
  • Disease-specific severity thresholds in the prompt tool = Not reported; cardiomegaly uses heart-to-thorax ratio thresholds of 0.50 and 0.55, other diseases use overlap-area…
    Algorithm 1 converts overlap areas between pathology and organ masks into mild/moderate/severe tokens. These thresholds are hand-chosen heuristics that determine prompt text and therefore the whole synthesis pipeline.
  • Self-assessment filter acceptance criteria = Not reported
    Section 4.2 lists prompt-match, realism, and diagnostic checks but does not give numeric thresholds for accepting or rejecting a generated mask or image; these criteria affect which samples survive to evaluation.
assumptions (5)
  • domain assumption Source pathology masks (ChestX-Det, CANDID-PTX) are accurate and label-consistent enough to train the text-to-mask stage.
    The pipeline inherits any errors in those public masks; no quality control on source masks is reported.
  • domain assumption Pretrained TorchXRayVision organ segmentation provides reliable anatomical priors.
    Organ masks extracted with this model condition the text-to-mask stage and define locations in prompts; Appendix I acknowledges segmentation failures but does not quantify them.
  • domain assumption The pretrained TorchXRayVision multi-label classifier is a valid clinical proxy for disease presence in synthetic images.
    Used both as filter (Section 4.2) and as Path AUC evaluator (Section 5.1); if the classifier is wrong about synthetic images, the diagnostic realism claim is unsupported.
  • ad hoc to paper InstructPix2Pix can transfer from natural-image editing to medical mask editing after fine-tuning on about 4,000 chest X-rays.
    The entire text-to-mask stage depends on this transfer; Appendix E shows alternative editors fail, so the paper relies on this specific backbone.
  • domain assumption Disease labels are standardized enough across datasets for direct augmentation transfer.
    The conclusion lists 'assumes standardized labels across datasets' as a limitation; this means the reported cross-dataset gains may not hold when label schemas differ.

how reviews work

0 comments
Cite this review

Pith. "Pith review of AURAD: Anatomy-Pathology Unified Radiology Synthesis with Progressive Representations." pith.science (2026). https://pith.science/paper/YIZVXJGK

@misc{pith2026250904819,
  author       = {Pith},
  title        = {Pith review of: AURAD: Anatomy-Pathology Unified Radiology Synthesis with Progressive Representations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YIZVXJGK}},
  note         = {Machine review of arXiv:2509.04819}
}
read the original abstract

Medical image synthesis has become an essential strategy for augmenting datasets and improving model generalization in data-scarce clinical settings. However, fine-grained and controllable synthesis remains difficult due to limited high-quality annotations and domain shifts across datasets. Existing methods, often designed for natural images or well-defined tumors, struggle to generalize to chest radiographs, where disease patterns are morphologically diverse and tightly intertwined with anatomical structures. To address these challenges, we propose AURAD, a controllable radiology synthesis framework that jointly generates high-fidelity chest X-rays and pseudo semantic masks. Unlike prior approaches that rely on randomly sampled masks-limiting diversity, controllability, and clinical relevance-our method learns to generate masks that capture multi-pathology coexistence and anatomical-pathological consistency. It follows a progressive pipeline: pseudo masks are first generated from clinical prompts conditioned on anatomical structures, and then used to guide image synthesis. We also leverage pretrained expert medical models to filter outputs and ensure clinical plausibility. Beyond visual realism, the synthesized masks also serve as labels for downstream tasks such as detection and segmentation, bridging the gap between generative modeling and real-world clinical applications. Extensive experiments and blinded radiologist evaluations demonstrate the effectiveness and generalizability of our method across tasks and datasets. In particular, 78% of our synthesized images are classified as authentic by board-certified radiologists, and over 40% of predicted segmentation overlays are rated as clinically useful. All code, pre-trained models, and the synthesized dataset will be released upon publication.

Figures

Figures reproduced from arXiv: 2509.04819 by the authors.

Figure 1
Figure 1. Overview of CXR synthesis strategies. Text-driven methods lack spatial control and medical priors; conditional [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Comparison of synthesis capabilities and limitations across existing CXR generation methods. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The AURAD pipeline consists of three phases: data preparation, training and inference. Data preparation [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of CXR synthesis methods. Colors denote different disease types; arrows of the same [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Pathology-driven t-SNE visualization of ResNet-18 [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 7
Figure 7. Figure 7: Examples of annotated diseases in Chest X-ray datasets. [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Screenshot of the evaluation interface.(a) Realism options for real and synthesized images. (b) Segmentation [PITH_FULL_IMAGE:figures/full_fig_p020_8.png]
Figure 9
Figure 9. Figure 9: Examples of our synthesized data comparing with real data, Cheff and RoentGen. [PITH_FULL_IMAGE:figures/full_fig_p022_9.png]
Figure 10
Figure 10. Figure 10: Examples of failure cases in our generated results. [PITH_FULL_IMAGE:figures/full_fig_p023_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 37 canonical work pages

  1. [53]

    Torchxrayvision: A library of chest x-ray datasets and models

    Joseph Paul Cohen, Joseph D Viviano, Paul Bertin, Paul Morrison, Parsa Torabian, Matteo Guarrera, Matthew P Lungren, Akshay Chaudhari, Rupert Brooks, Mohammad Hashir, et al. Torchxrayvision: A library of chest x-ray datasets and models. InInternational Conference on Medical Imaging with Deep Learning, pages 231–249. PMLR, 2022

  2. [59]

    Viviano, Paul Bertin, Paul Morrison, Parsa Torabian, Matteo Guarrera, Matthew P Lungren, Akshay Chaudhari, Rupert Brooks, Mohammad Hashir, and Hadrien Bertrand

    Joseph Paul Cohen, Joseph D. Viviano, Paul Bertin, Paul Morrison, Parsa Torabian, Matteo Guarrera, Matthew P Lungren, Akshay Chaudhari, Rupert Brooks, Mohammad Hashir, and Hadrien Bertrand. TorchXRayVision: A library of chest X-ray datasets and models. InMedical Imaging with Deep Learning, 2022

  3. [14]

    Xreal: Realistic anatomy and pathology-aware x-ray generation via controllable diffusion model.arXiv preprint arXiv:2403.09240, 2024

    Anees Ur Rehman Hashmi, Ibrahim Almakky, Mohammad Areeb Qazi, Santosh Sanjeev, Vijay Ram Papineni, Jagalpathy Jagdish, and Mohammad Yaqub. Xreal: Realistic anatomy and pathology-aware x-ray generation via controllable diffusion model.arXiv preprint arXiv:2403.09240, 2024

  4. [1]

    Challenges related to artificial intelligence research in medical imaging and the importance of image analysis competitions.Radiology: Artificial Intelligence, 1(1):e180031, 2019

    Luciano M Prevedello, Safwan S Halabi, George Shih, Carol C Wu, Marc D Kohli, Falgun H Chokshi, Bradley J Erickson, Jayashree Kalpathy-Cramer, Katherine P Andriole, and Adam E Flanders. Challenges related to artificial intelligence research in medical imaging and the importance of image analysis competitions.Radiology: Artificial Intelligence, 1(1):e180031, 2019

  5. [2]

    Synthetic data in medical research.BMJ medicine, 1(1):e000167, 2022

    Theodora Kokosi and Katie Harron. Synthetic data in medical research.BMJ medicine, 1(1):e000167, 2022

  6. [3]

    Harnessing the power of synthetic data in healthcare: innovation, application, and privacy.NPJ digital medicine, 6(1):186, 2023

    Mauro Giuffrè and Dennis L Shung. Harnessing the power of synthetic data in healthcare: innovation, application, and privacy.NPJ digital medicine, 6(1):186, 2023

  7. [4]

    A vision–language foundation model for the generation of realistic chest x-ray images.Nature Biomedical Engineering, pages 1–13, 2024

    Christian Bluethgen, Pierre Chambon, Jean-Benoit Delbrouck, Rogier van der Sluijs, Małgorzata Połacin, Juan Manuel Zambrano Chaves, Tanishq Mathew Abraham, Shivanshu Purohit, Curtis P Langlotz, and Ak- shay S Chaudhari. A vision–language foundation model for the generation of realistic chest x-ray images.Nature Biomedical Engineering, pages 1–13, 2024

  8. [5]

    Chest-diffusion: a light-weight text-to-image model for report-to-cxr generation

    Peng Huang, Xue Gao, Lihong Huang, Jing Jiao, Xiaokang Li, Yuanyuan Wang, and Yi Guo. Chest-diffusion: a light-weight text-to-image model for report-to-cxr generation. In2024 IEEE International Symposium on Biomedical Imaging (ISBI), pages 1–5. IEEE, 2024

Show all 64 references
  1. [6]

    Biomedjourney: Counterfactual biomedical image generation by instruction-learning from multimodal patient journeys.arXiv preprint arXiv:2310.10765, 2023

    Yu Gu, Jianwei Yang, Naoto Usuyama, Chunyuan Li, Sheng Zhang, Matthew P Lungren, Jianfeng Gao, and Hoifung Poon. Biomedjourney: Counterfactual biomedical image generation by instruction-learning from multimodal patient journeys.arXiv preprint arXiv:2310.10765, 2023

  2. [7]

    Semi-supervised learning with generative adversarial networks for chest x-ray classification with ability of data domain adaptation

    Ali Madani, Mehdi Moradi, Alexandros Karargyris, and Tanveer Syeda-Mahmood. Semi-supervised learning with generative adversarial networks for chest x-ray classification with ability of data domain adaptation. In2018 IEEE 15th International symposium on biomedical imaging (ISBI...

  3. [8]

    Xraygan: Consistency-preserving generation of x-ray images from radiology reports.arXiv preprint arXiv:2006.10552, 2020

    Xingyi Yang, Nandiraju Gireesh, Eric Xing, and Pengtao Xie. Xraygan: Consistency-preserving generation of x-ray images from radiology reports.arXiv preprint arXiv:2006.10552, 2020

  4. [9]

    Efficient gan-based chest radiographs (cxr) augmentation to diagnose coronavirus disease pneumo- nia.International journal of medical sciences, 17(10):1439, 2020

    Saleh Albahli. Efficient gan-based chest radiographs (cxr) augmentation to diagnose coronavirus disease pneumo- nia.International journal of medical sciences, 17(10):1439, 2020

  5. [10]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

  6. [11]

    Diffusion models beat gans on image synthesis.Advances in neural information processing systems, 34:8780–8794, 2021

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis.Advances in neural information processing systems, 34:8780–8794, 2021

  7. [12]

    Cascaded latent diffusion models for high-resolution chest x-ray synthesis

    Tobias Weber, Michael Ingrisch, Bernd Bischl, and David Rügamer. Cascaded latent diffusion models for high-resolution chest x-ray synthesis. InPacific-Asia conference on knowledge discovery and data mining, pages 180–191. Springer, 2023

  8. [13]

    Radedit: stress-testing biomedical vision models via diffusion image editing

    Fernando Pérez-García, Sam Bond-Taylor, Pedro P Sanchez, Boris van Breugel, Daniel C Castro, Harshita Sharma, Valentina Salvatelli, Maria TA Wetscherek, Hannah Richardson, Matthew P Lungren, et al. Radedit: stress-testing biomedical vision models via diffusion image editing. I...

  9. [15]

    Label- free liver tumor segmentation

    Qixin Hu, Yixiong Chen, Junfei Xiao, Shuwen Sun, Jieneng Chen, Alan L Yuille, and Zongwei Zhou. Label- free liver tumor segmentation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7422–7432, 2023

  10. [16]

    Towards generalizable tumor synthesis

    Qi Chen, Xiaoxi Chen, Haorui Song, Zhiwei Xiong, Alan Yuille, Chen Wei, and Zongwei Zhou. Towards generalizable tumor synthesis. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11147–11158, 2024

  11. [17]

    Layoutdiffusion: Controllable diffusion model for layout-to-image generation

    Guangcong Zheng, Xianpan Zhou, Xuewei Li, Zhongang Qi, Ying Shan, and Xi Li. Layoutdiffusion: Controllable diffusion model for layout-to-image generation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22490–22499, 2023

  12. [18]

    Layoutllm-t2i: Eliciting layout guidance from llm for text-to-image generation

    Leigang Qu, Shengqiong Wu, Hao Fei, Liqiang Nie, and Tat-Seng Chua. Layoutllm-t2i: Eliciting layout guidance from llm for text-to-image generation. InProceedings of the 31st ACM International Conference on Multimedia, pages 643–654, 2023. 11

  13. [19]

    Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning.arXiv preprint arXiv:1711.05225, 2017

    Pranav Rajpurkar, Jeremy Irvin, Kaylie Zhu, Brandon Yang, Hershel Mehta, Tony Duan, Daisy Ding, Aarti Bagul, Curtis Langlotz, Katie Shpanskaya, et al. Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning.arXiv preprint arXiv:1711.05225, 2017

  14. [20]

    Chexpert: A large chest radiograph dataset with uncertainty labels and expert comparison

    Jeremy Irvin, Pranav Rajpurkar, Michael Ko, Yifan Yu, Silviana Ciurea-Ilcus, Chris Chute, Henrik Marklund, Behzad Haghgoo, Robyn Ball, Katie Shpanskaya, et al. Chexpert: A large chest radiograph dataset with uncertainty labels and expert comparison. InProceedings of the AAAI c...

  15. [21]

    A structure-aware relation network for thoracic diseases detection and segmentation.IEEE Transactions on Medical Imaging, 40(8):2042–2052, 2021

    Jie Lian, Jingyu Liu, Shu Zhang, Kai Gao, Xiaoqing Liu, Dingwen Zhang, and Yizhou Yu. A structure-aware relation network for thoracic diseases detection and segmentation.IEEE Transactions on Medical Imaging, 40(8):2042–2052, 2021

  16. [22]

    Abnormality detection and localization in chest x-rays using deep convolutional neural networks.arXiv preprint arXiv:1705.09850, 2017

    Mohammad Tariqul Islam, Md Abdul Aowal, Ahmed Tahseen Minhaz, and Khalid Ashraf. Abnormality detection and localization in chest x-rays using deep convolutional neural networks.arXiv preprint arXiv:1705.09850, 2017

  17. [23]

    Yolo-cxr: A novel detection network for locating multiple small lesions in chest x-ray images.IEEE Access, 2024

    Shengnan Hao, Xinlei Li, Wei Peng, Zhu Fan, Zhanlin Ji, and Ivan Ganchev. Yolo-cxr: A novel detection network for locating multiple small lesions in chest x-ray images.IEEE Access, 2024

  18. [24]

    Evaluating and improving the effectiveness of synthetic chest x-rays for medical image analysis.arXiv preprint arXiv:2411.18602, 2024

    Eva Prakash, Jeya Maria Jose Valanarasu, Zhihong Chen, Eduardo Pontes Reis, Andrew Johnston, Anuj Pareek, Christian Bluethgen, Sergios Gatidis, Cameron Olsen, Akshay Chaudhari, et al. Evaluating and improving the effectiveness of synthetic chest x-rays for medical image analys...

  19. [25]

    Mri cross-modality image-to- image translation.Scientific reports, 10(1):3753, 2020

    Qianye Yang, Nannan Li, Zixu Zhao, Xingyu Fan, Eric I-Chao Chang, and Yan Xu. Mri cross-modality image-to- image translation.Scientific reports, 10(1):3753, 2020

  20. [26]

    Translating and segmenting multimodal medical volumes with cycle-and shape-consistency generative adversarial network

    Zizhao Zhang, Lin Yang, and Yefeng Zheng. Translating and segmenting multimodal medical volumes with cycle-and shape-consistency generative adversarial network. InProceedings of the IEEE conference on computer vision and pattern Recognition, pages 9242–9251, 2018

  21. [27]

    A gan-based image synthesis method for skin lesion classification.Computer methods and programs in biomedicine, 195:105568, 2020

    Zhiwei Qin, Zhao Liu, Ping Zhu, and Yongbo Xue. A gan-based image synthesis method for skin lesion classification.Computer methods and programs in biomedicine, 195:105568, 2020

  22. [28]

    Gan-based data augmentation and anonymization for skin- lesion analysis: A critical review

    Alceu Bissoto, Eduardo Valle, and Sandra Avila. Gan-based data augmentation and anonymization for skin- lesion analysis: A critical review. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1847–1856, 2021

  23. [29]

    Gan-based synthetic medical image augmentation for increased cnn performance in liver lesion classification.Neurocomputing, 321:321–331, 2018

    Maayan Frid-Adar, Idit Diamant, Eyal Klang, Michal Amitai, Jacob Goldberger, and Hayit Greenspan. Gan-based synthetic medical image augmentation for increased cnn performance in liver lesion classification.Neurocomputing, 321:321–331, 2018

  24. [30]

    A review of medical image data augmentation techniques for deep learning applications.Journal of medical imaging and radiation oncology, 65(5):545–563, 2021

    Phillip Chlap, Hang Min, Nym Vandenberg, Jason Dowling, Lois Holloway, and Annette Haworth. A review of medical image data augmentation techniques for deep learning applications.Journal of medical imaging and radiation oncology, 65(5):545–563, 2021

  25. [31]

    Medgen3d: A deep generative framework for paired 3d image and mask generation

    Kun Han, Yifeng Xiong, Chenyu You, Pooya Khosravi, Shanlin Sun, Xiangyi Yan, James S Duncan, and Xiaohui Xie. Medgen3d: A deep generative framework for paired 3d image and mask generation. InInternational Conference on Medical Image Computing and Computer-Assisted Intervention...

  26. [32]

    Boosting dermatoscopic lesion segmentation via diffusion models with visual and textual prompts

    Shiyi Du, Xiaosong Wang, Yongyi Lu, Yuyin Zhou, Shaoting Zhang, Alan Yuille, Kang Li, and Zongwei Zhou. Boosting dermatoscopic lesion segmentation via diffusion models with visual and textual prompts. In2024 IEEE International Symposium on Biomedical Imaging (ISBI), pages 1–5....

  27. [33]

    Maskmedpaint: Masked medical image inpainting with diffusion models for mitigation of spurious correlations.arXiv preprint arXiv:2411.10686, 2024

    Qixuan Jin, Walter Gerych, and Marzyeh Ghassemi. Maskmedpaint: Masked medical image inpainting with diffusion models for mitigation of spurious correlations.arXiv preprint arXiv:2411.10686, 2024

  28. [34]

    Removing anomalies as noises for industrial defect localization

    Fanbin Lu, Xufeng Yao, Chi-Wing Fu, and Jiaya Jia. Removing anomalies as noises for industrial defect localization. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 16166–16175, 2023

  29. [35]

    Pixel-wise defect detection by cnns without manually labeled training data.Applied Artificial Intelligence, 33(6):548–566, 2019

    Matthias Haselmann and Dieter P Gruber. Pixel-wise defect detection by cnns without manually labeled training data.Applied Artificial Intelligence, 33(6):548–566, 2019

  30. [36]

    Learning from limited and imbalanced medical images with finer synthetic images from gans.Ieee Access, 10:91663–91677, 2022

    Xiaoli Qin, Francis Minhthang Bui, Ha H Nguyen, and Zhu Han. Learning from limited and imbalanced medical images with finer synthetic images from gans.Ieee Access, 10:91663–91677, 2022

  31. [37]

    Medical image synthesis via conditional gans: Application to segmenting brain tumours.Computers in Biology and Medicine, 170:107982, 2024

    Mohammad Hamghalam and Amber L Simpson. Medical image synthesis via conditional gans: Application to segmenting brain tumours.Computers in Biology and Medicine, 170:107982, 2024

  32. [38]

    Diffusion models for medical anomaly detection

    Julia Wolleb, Florentin Bieder, Robin Sandkühler, and Philippe C Cattin. Diffusion models for medical anomaly detection. InInternational Conference on Medical image computing and computer-assisted intervention, pages 35–45. Springer, 2022. 12

  33. [39]

    Diffboost: Enhancing medical image segmentation via text-guided diffusion model.IEEE Transactions on Medical Imaging, 2024

    Zheyuan Zhang, Lanhong Yao, Bin Wang, Debesh Jha, Gorkem Durak, Elif Keles, Alpay Medetalibeyoglu, and Ulas Bagci. Diffboost: Enhancing medical image segmentation via text-guided diffusion model.IEEE Transactions on Medical Imaging, 2024

  34. [40]

    Complex organ mask guided radiology report generation

    Tiancheng Gu, Dongnan Liu, Zhiyuan Li, and Weidong Cai. Complex organ mask guided radiology report generation. InProceedings of the IEEE/CVF winter conference on applications of computer vision, pages 7995–8004, 2024

  35. [41]

    A multi-stage gan for multi-organ chest x-ray image generation and segmentation.Mathematics, 9(22):2896, 2021

    Giorgio Ciano, Paolo Andreini, Tommaso Mazzierli, Monica Bianchini, and Franco Scarselli. A multi-stage gan for multi-organ chest x-ray image generation and segmentation.Mathematics, 9(22):2896, 2021

  36. [42]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervision. InInternational conference on machine learning, pag...

  37. [43]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis.arXiv preprint arXiv:2307.01952, 2023

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas Müller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis.arXiv preprint arXiv:2307.01952, 2023

  38. [44]

    Spot the fake lungs: Generating synthetic medical images using neural diffusion models

    Hazrat Ali, Shafaq Murad, and Zubair Shah. Spot the fake lungs: Generating synthetic medical images using neural diffusion models. InIrish Conference on Artificial Intelligence and Cognitive Science, pages 32–39. Springer, 2022

  39. [45]

    Adapting pretrained vision- language foundational models to medical imaging domains.arXiv preprint arXiv:2210.04133, 2022

    Pierre Chambon, Christian Bluethgen, Curtis P Langlotz, and Akshay Chaudhari. Adapting pretrained vision- language foundational models to medical imaging domains.arXiv preprint arXiv:2210.04133, 2022

  40. [46]

    Localization-adjusted diagnostic performance and assistance effect of a computer-aided detection system for pneumothorax and consolidation.npj Digital Medicine, 5(1):107, 2022

    Sun Yeop Lee, Sangwoo Ha, Min Gyeong Jeon, Hao Li, Hyunju Choi, Hwa Pyung Kim, Ye Ra Choi, Hoseok I, Yeon Joo Jeong, Yoon Ha Park, et al. Localization-adjusted diagnostic performance and assistance effect of a computer-aided detection system for pneumothorax and consolidation....

  41. [47]

    Medsegdiff: Medical image segmentation with diffusion probabilistic model

    Junde Wu, Rao Fu, Huihui Fang, Yu Zhang, Yehui Yang, Haoyi Xiong, Huiying Liu, and Yanwu Xu. Medsegdiff: Medical image segmentation with diffusion probabilistic model. InMedical Imaging with Deep Learning, pages 1623–1639. PMLR, 2024

  42. [48]

    Layoutdm: Transformer-based diffusion model for layout generation

    Shang Chai, Liansheng Zhuang, and Fengying Yan. Layoutdm: Transformer-based diffusion model for layout generation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18349–18358, 2023

  43. [49]

    Instructpix2pix: Learning to follow image editing instructions

    Tim Brooks, Aleksander Holynski, and Alexei A Efros. Instructpix2pix: Learning to follow image editing instructions. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 18392–18402, 2023

  44. [50]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. InProceedings of the IEEE/CVF international conference on computer vision, pages 3836–3847, 2023

  45. [51]

    Curation of the candid-ptx dataset with free-text reports.Radiology: Artificial Intelligence, 3(6):e210136, 2021

    Sijing Feng, Damian Azzollini, Ji Soo Kim, Cheng-Kai Jin, Simon P Gordon, Jason Yeoh, Eve Kim, Mina Han, Andrew Lee, Aakash Patel, et al. Curation of the candid-ptx dataset with free-text reports.Radiology: Artificial Intelligence, 3(6):e210136, 2021

  46. [52]

    Chestx-det10: Chest x-ray dataset on detection of thoracic abnormalities, 2020

    Jingyu Liu, Jie Lian, and Yizhou Yu. Chestx-det10: Chest x-ray dataset on detection of thoracic abnormalities, 2020

  47. [54]

    Alistair E. W. Johnson, Tom J. Pollard, Nathaniel R. Greenbaum, Matthew P. Lungren, Chih-ying Deng, Yifan Peng, Zhiyong Lu, Roger G. Mark, Seth J. Berkowitz, and Steven Horng. Mimic-cxr-jpg, a large publicly available database of labeled chest radiographs.arXiv preprint arXiv:...

  48. [55]

    Biomedclip: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs.arXiv preprint arXiv:2303.00915, 2023

    Sheng Zhang, Yanbo Xu, Naoto Usuyama, Hanwen Xu, Jaspreet Bagga, Robert Tinn, Sam Preston, Rajesh Rao, Mu Wei, Naveen Valluri, et al. Biomedclip: a multimodal biomedical foundation model pretrained from fifteen million scientific image-text pairs.arXiv preprint arXiv:2303.00915, 2023

  49. [56]

    Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases

    Xiaosong Wang, Yifan Peng, Le Lu, Zhiyong Lu, Mohammadhadi Bagheri, and Ronald M Summers. Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. InProceedings of the IEEE conference on com...

  50. [57]

    Nguyen, Khanh Lam, Linh T

    Ha Q. Nguyen, Khanh Lam, Linh T. Le, Hieu H. Pham, Dat Q. Tran, Dung B. Nguyen, Dung D. Le, Chi M. Pham, Hang T. T. Tong, Diep H. Dinh, Cuong D. Do, Luu T. Doan, Cuong N. Nguyen, Binh T. Nguyen, Que V . Nguyen, Au D. Hoang, Hien N. Phan, Anh T. Nguyen, Phuong H. Ho, Dat T. Ngo...

  51. [58]

    Gans trained by a two time-scale update rule converge to a local nash equilibrium, 2018

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilibrium, 2018

  52. [60]

    Learning transferable visual models from natural language supervision, 2021

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision, 2021

  53. [61]

    Mask r-cnn

    Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick. Mask r-cnn. InProceedings of the IEEE International Conference on Computer Vision (ICCV), pages 2961–2969, 2017

  54. [62]

    Feature pyramid networks for object detection.arXiv preprint arXiv:1612.03144, 2017

    Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. Feature pyramid networks for object detection.arXiv preprint arXiv:1612.03144, 2017

  55. [63]

    Diffedit: Diffusion-based semantic image editing with mask guidance.arXiv preprint arXiv:2210.11427, 2022

    Guillaume Couairon, Jakob Verbeek, Holger Schwenk, and Matthieu Cord. Diffedit: Diffusion-based semantic image editing with mask guidance.arXiv preprint arXiv:2210.11427, 2022

  56. [64]

    densenet121-res224-all

    Yoad Tewel, Rinon Gal, Dvir Samuel, Yuval Atzmon, Lior Wolf, and Gal Chechik. Add-it: Training-free object insertion in images with pretrained diffusion models.arXiv preprint arXiv:2411.07232, 2024. 14 Appendix A Datasets: Statistics and Annotation Examples We utilize five pub...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.