Pith. sign in

REVIEW 4 major objections 5 minor 2 cited by

EyeDiff: text-to-image diffusion model improves rare eye disease diagnosis

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Adding synthetic images from a text-to-image diffusion model significantly improves automated diagnosis of common and rare retinal diseases, with the largest gains on a 17-class rare-disease benchmark.

desk verdict Worth refereeing for the idea, but the headline result is undermined by an OCTDL/Retinal Image Bank train-test overlap. read the letter →

arxiv 2411.10004 v1 pith:PLG7QEVW submitted 2024-11-15 eess.IV cs.AIcs.CV

Ruoyu Chen , Weiyi Zhang , Bowen Liu , Xiaolan Chen , Pusheng Xu , Shunming Liu , Mingguang He , Danli Shi This is my paper · ORCID
classification eess.IVcs.AIcs.CV
keywords text-to-imagegenerationlatentdiffusionmodelmultimodalretinalimagingrareeyediseasedataaugmentationclassimbalancesyntheticmedicalimagesdiagnosis
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 introduces EyeDiff, a text-to-image diffusion model trained on 42,048 multimodal ophthalmic images with text descriptions covering 14 imaging modalities and over 80 ocular diseases. The authors aim to show that adding EyeDiff-generated images to real training images improves automated diagnosis of both common and rare retinal diseases, and does so more effectively than traditional random oversampling. On a 17-class rare-disease task, the reported AUROC rises from 0.871 to 0.919 and AUPR from 0.439 to 0.530 when synthetic images are added. If the claim holds, it would reduce the need for large annotated collections of rare eye disease images and enable sharing of training data without exposing real patient photos.

What carries the argument

EyeDiff is a latent diffusion model built on Stable Diffusion v1-5, where text prompts are encoded by a CLIP text encoder and injected into a time-conditional UNet through cross-attention, and the UNet denoises latents produced by a variational autoencoder to synthesize images. The generated images are then used as additional training examples alongside real images; the downstream classifier is a Vision Transformer initialized with RETFound weights that fuses real and synthetic embeddings through an attention-based feature fusion module before softmax classification. The load-bearing step is the text-conditioned generation of class-specific synthetic images for minority classes.

What would settle it

Retrain EyeDiff with OCTDL and the Retinal Image Bank held out from training (or hold out the specific test images), rerun the augmentation protocol on OCTDL and the rare-disease set, and check whether the AUROC gains over baseline and oversampling persist; if they disappear, the claimed benefit is an artifact of training/test overlap. A complementary check is to measure nearest-neighbor distances between generated images and the downstream test images to detect memorization.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a text-conditioned latent diffusion model can generate multimodal ophthalmic images whose lesion features align closely enough with natural-language prompts that, when fused with real images, they substantially improve a foundation model's classification of underrepresented and rare retinal diseases across ten external benchmarks. The authors report that the synthetic-image augmentation outperforms both the RETFound baseline and random oversampling, with statistically significant AUROC gains on most tasks and the largest gains on the rarest classes. The explanation offered is that generated images provide diverse, class-specific visual features for minority classes rather than merely duplicating existing examples.

Load-bearing premise

The downstream evaluation assumes that the datasets used to test EyeDiff-generated augmentation are disjoint from the data EyeDiff was trained on, but OCTDL appears in both the training set and the downstream validation set with the same image count, and the rare-disease validation set is drawn from the Retinal Image Bank that also trained EyeDiff.

Editorial extensions

If this is right

  • Synthetic images generated from text prompts can partially replace real annotated data for training classifiers of rare eye diseases.
  • The benefit over random oversampling indicates that the value of synthetic data lies not only in class rebalancing but in the diversity or fidelity of the generated examples.
  • Because synthetic images do not belong to real patients, the approach could lower privacy barriers to sharing ophthalmic training data across centers.
  • The augmentation protocol transfers across multiple imaging modalities and datasets from different countries, pointing to a general strategy for multimodal retinal disease diagnosis.
  • The largest AUROC improvements are on the rarest classes, such as optic nerve hypoplasia rising from 0.701 to 0.774, where data scarcity is most severe.

Reading between the lines

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

  • A strict test would retrain EyeDiff with OCTDL and the Retinal Image Bank excluded from training; the current numbers may overstate generalization because those datasets appear in both training and validation.
  • If the improvement survives that split, a natural follow-up is to measure feature diversity of generated versus oversampled minority images to isolate why synthetic data helps.
  • The approach likely extends to other medical imaging domains with text-rich image archives, provided the text prompts capture the relevant lesion morphology, an aspect the authors note is still simplified.
  • A head-to-head comparison with GAN-based augmentation under identical held-out conditions would clarify whether diffusion-model generation, rather than any synthetic data, drives the gain.
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

4 major / 5 minor

Summary. The paper introduces EyeDiff, a latent diffusion model fine-tuned on multimodal ophthalmic image–text pairs, and evaluates whether synthetic images generated by EyeDiff improve downstream retinal disease diagnosis. The authors report that augmenting original real images with EyeDiff-generated images improves AUROC and AUPR for common and rare diseases relative to a RETFound baseline and to traditional oversampling, with the headline rare-disease result an AUROC increase from 0.871 to 0.919 (Table 3, ImageBank). The manuscript also includes VQAScore alignment metrics and a small human evaluation of generated images.

Significance. If the diagnostic improvement claims were valid, EyeDiff would offer a practical tool for addressing class imbalance and data scarcity in ophthalmic imaging, and the multimodal text-to-image model itself could be a useful resource. The paper provides a substantial amount of empirical data and a clear experimental framework. However, the central claim is compromised by a likely train–test overlap: two of the downstream validation datasets, OCTDL and the Retinal Image Bank rare-disease set, also appear in EyeDiff's training data (Table 1). This overlap means the reported gains may reflect test-set memorization rather than genuine generalization of the synthetic augmentation. The manuscripts also contains internal inconsistencies between the narrative text and the tabulated statistical significance. Because the main contribution is the claimed diagnostic benefit, the significance of the work as presented is not established.

major comments (4)
  1. [Table 1; Methods – Datasets] The downstream evaluation is not independent of EyeDiff's training data. Table 1 lists OCTDL with the identical image count (2,064) in both the EyeDiff Development set and the Downstream Validation set, and the rare-disease downstream set (2,339 images) is described in Methods as 'collected from the Retinal Image Bank between 2019 and 2023,' the same repository that contributes 22,941 training images. No exclusion, temporal cut, hash-based deduplication, or patient-level split is described. If the generative model memorized near-duplicates of the test images, especially for rare classes with as few as 3–7 examples (Table 4), the synthetic images could be test-set replicas, and the reported AUROC gains in Table 3 (e.g., OCTDL 0.982→0.996, ImageBank 0.871→0.919) would be inflated. This directly undermines the paper's central claim that EyeDiff-generated images improve diagnosis on real-world data.
  2. [Results – EyeDiff improves multi-class disease diagnosis; Table 3] The text states that for JSIEC and Retina 'these differences were statistically significant,' but Table 3 reports a p-value of 0.082 for JSIEC EyeDiff versus RETFound, which is not significant at the conventional 0.05 level. Similarly, the earlier statement that DR and glaucoma improvements were 'statistically significant' is contradicted by Table 3 for APTOS2019 (p=0.211) and MESSIDOR2 (p=0.213), where EyeDiff does not improve AUROC, and for PAPILA, where EyeDiff's overall AUROC (0.814) is numerically lower than baseline (0.820). The paper's abstract and conclusion claim that EyeDiff 'significantly improved' diagnosis accuracy, but the tabulated results support that claim only for a subset of datasets. This discrepancy is load-bearing because the central contribution is the claimed diagnostic benefit.
  3. [Methods – Datasets for downstream tasks; Rare diseases diagnosis] The rare-disease evaluation protocol is critically underspecified. The manuscript does not describe how the 2,339 rare-disease images were split into training and test sets for the downstream classifier, how many synthetic images were generated per class, what prompts were used for each of the 17 classes, or whether any real images in the test set were also used to condition generation. Without this information, the reported AUPR improvement from 0.439 to 0.530 (Table 3, ImageBank) cannot be attributed to the augmentation method; it could be the result of the generative model having been trained on the same images that later appeared in the test set. This is not a minor omission but a fundamental experimental control that must be reported.
  4. [Abstract; Results] The abstract characterizes the downstream datasets as 'ten multi-country external datasets,' but Table 1 shows that OCTDL and the Retinal Image Bank are used both for training EyeDiff and for downstream validation. These are therefore not external in the standard sense of being disjoint from the training distribution. The term 'external validation' is misleading and overstates the generalizability of the reported performance. At minimum, the authors should either exclude all training-source datasets from downstream evaluation or explicitly describe the train/test splits within those datasets and confirm that no image-level overlap exists.
minor comments (5)
  1. [Table 4 footnote] The footnote states 'The results displayed in the table are validated on the test set,' but no description of how the test set was constructed is provided anywhere in Methods; this should be specified for every downstream dataset, not just implied.
  2. [Results – EyeDiff-generated images enhance disease classification in minority classes; Table 4] Several minority-class improvements reported in the text are accompanied by non-significant p-values in Table 4, such as OCTDL retinal artery occlusion (p=0.135) and JSIEC Bietti crystalline dystrophy (p=0.231); the text should not describe all listed improvements as significant.
  3. [Methods – Human evaluation] The human evaluation of generated images was performed by only two graders, both of whom are co-authors (R.C. and X.C.), on a sample of 50 images. The Turing test result that graders mistook 62–66% of generated images for real is close to chance (50%) and should be interpreted cautiously; the author involvement and small sample should be acknowledged as limitations.
  4. [Data and code availability] The code link points to a general HuggingFace Diffusers example (dreambooth) rather than to the actual EyeDiff implementation; the paper should provide the specific training and generation code, the exact text prompts used, and the hyperparameters so that the experiments are reproducible.
  5. [Keywords] The keywords list contains a stray double comma ('Stable Diffusion, , Rare disease'); also, the paper consistently misspells 'RETFound' as 'RetFound' in Table 3 and elsewhere, which should be corrected.

Circularity Check

1 steps flagged · score 6.0 of 10

Rare-disease and OCTDL downstream evaluations reuse EyeDiff training datasets; reported gains are confounded by train/test overlap.

  1. other [Methods – Datasets; Table 1; Methods – Datasets for downstream tasks]
    "EyeDiff Development 42,048(100.00%) ... RetinaImageBank ... 22,941(54.56%) ... OCTDL OCT AMD, DME, RVO, RAO, ERM, VID, Normal 2,064(4.91%) ... DownstreamValidation 14,530(100.00%) ... OCTDL ... 2,064(14.20%) ... RareDiseases ... 2,339(16.10%)"

    The downstream evaluation for OCTDL and for rare diseases reuses the same datasets used to train EyeDiff: OCTDL appears with the identical count (2,064) in both 'EyeDiff Development' and 'DownstreamValidation', and the rare-disease set is 'Images collected from the Retinal Image Bank between 2019 and 2023', which is EyeDiff's largest training source (22,941 images). No train/test split, exclusion, or deduplication is described. A diffusion model trained on these images can memorize and emit near-duplicates of the evaluation images, so the synthetic images added to the downstream classifier can carry test-set information.

full rationale

The paper's central claim is that EyeDiff-generated images improve diagnosis, especially for rare diseases. For the OCTDL and rare-disease downstream tasks, the evaluation is not independent of the generative model's training data: Table 1 lists OCTDL with the same image count in both the EyeDiff development set and the downstream validation set, and the rare-disease set is explicitly collected from the Retinal Image Bank, which is also EyeDiff's largest training source. Because no exclusion or split is documented, synthetic images used to augment the downstream classifier may be near-replicas of the test images, so the observed AUROC/AUPR improvements on those benchmarks do not establish genuine generalization. The remaining downstream datasets (IDRiD, APTOS, MESSIDOR-2, PAPILA, Glaucoma Fundus, JSIEC, Retina, OCTID) appear disjoint from the training sets and provide some independent evidence for the method, so the circularity is partial rather than total. No additional definitional circularity was found in the generative training or text-prompt design; the issue is specifically the overlap between EyeDiff's training distribution and the evaluation sets used for the headline rare-disease and OCTDL results.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim depends on several unstated or weakly supported premises: that short text prompts capture clinically relevant visual features, that VQAScore reflects clinical alignment, and especially that the downstream evaluation data are independent of EyeDiff's training data. The last premise is likely violated, which is the main circularity burden. No new physical entities are introduced; the model is a fine-tuned Stable Diffusion.

free parameters (3)
  • Number of synthetic images added per minority class = not reported
    The augmentation benefit likely depends on the synthetic-to-real ratio; the paper does not disclose how many EyeDiff images are added for each class, so this could have been tuned to favor EyeDiff.
  • Stable Diffusion guidance scale = not reported
    The classifier-free guidance scale controls text-image alignment but is not stated; this affects generated image quality and downstream utility.
  • Text prompt template = hand-crafted via custom dictionary
    The prompt construction (modality plus disease name plus severity) is an ad hoc design choice that determines what the model generates; the mapping is not fully released.
assumptions (4)
  • domain assumption Text prompts consisting of image modality and disease name are sufficient to condition generation of clinically valid images
    The whole method relies on the custom dictionary mapping diseases to short prompts; no evidence is given that these prompts capture the visual features needed for diagnosis, especially for rare diseases with low VQAScores (e.g., 0.493 for retinopathy of prematurity).
  • domain assumption VQAScore is a valid proxy for clinical image-text alignment
    The paper uses VQAScore as its main objective quality metric, but VQAScore is a general VQA model, not validated for ophthalmic pathology; human evaluation on only 50 images gives modest deception rates (62-66%).
  • domain assumption The downstream evaluation datasets are independent of EyeDiff training data
    This is the load-bearing assumption; it is violated by the apparent overlap of OCTDL and Retinal Image Bank between training and downstream evaluation.
  • standard math Standard latent diffusion training objective and CLIP text encoder behave as described
    The paper relies on established SDv1.5 components; this is unproblematic background.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EyeDiff: text-to-image diffusion model improves rare eye disease diagnosis." pith.science (2026). https://pith.science/paper/PLG7QEVW

@misc{pith2026241110004,
  author       = {Pith},
  title        = {Pith review of: EyeDiff: text-to-image diffusion model improves rare eye disease diagnosis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PLG7QEVW}},
  note         = {Machine review of arXiv:2411.10004}
}
read the original abstract

The rising prevalence of vision-threatening retinal diseases poses a significant burden on the global healthcare systems. Deep learning (DL) offers a promising solution for automatic disease screening but demands substantial data. Collecting and labeling large volumes of ophthalmic images across various modalities encounters several real-world challenges, especially for rare diseases. Here, we introduce EyeDiff, a text-to-image model designed to generate multimodal ophthalmic images from natural language prompts and evaluate its applicability in diagnosing common and rare diseases. EyeDiff is trained on eight large-scale datasets using the advanced latent diffusion model, covering 14 ophthalmic image modalities and over 80 ocular diseases, and is adapted to ten multi-country external datasets. The generated images accurately capture essential lesional characteristics, achieving high alignment with text prompts as evaluated by objective metrics and human experts. Furthermore, integrating generated images significantly enhances the accuracy of detecting minority classes and rare eye diseases, surpassing traditional oversampling methods in addressing data imbalance. EyeDiff effectively tackles the issue of data imbalance and insufficiency typically encountered in rare diseases and addresses the challenges of collecting large-scale annotated images, offering a transformative solution to enhance the development of expert-level diseases diagnosis models in ophthalmic field.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. LesionGen: A Concept-Guided Diffusion Model for Dermatology Image Synthesis

    eess.IV 2025-07 conditional novelty 5.0 of 10

    Concept-guided captions and prompt balancing improve synthetic skin lesion images enough that augmenting real data with them lifts rare-class classification, though synthetic-only training remains clearly worse than r...

  2. FFA Sora, video generation as fundus fluorescein angiography simulator

    cs.CV 2024-12 conditional novelty 5.0 of 10

    FFA Sora generates simulated fundus fluorescein angiography videos from text reports using an Open-Sora-based diffusion transformer, with moderate objective metrics and weak evidence for the privacy claim.

Reference graph

Works this paper leans on

42 extracted references · 36 canonical work pages · cited by 2 Pith papers

  1. [1]

    & Rosário, A

    Raimundo, R. & Rosário, A. The Impact of Artificial Intelligence on Data System Security: A Literature Review. Sensors (Basel, Switzerland) 21(2021)

    work page 2021

  2. [2]

    Severe macular complications in glaucoma: high-resolution multimodal imaging characteristics and review of the literature

    Lama, H., et al. Severe macular complications in glaucoma: high-resolution multimodal imaging characteristics and review of the literature. BMC ophthalmology 23, 318 (2023)

    work page 2023

  3. [3]

    Association of Diabetic Lesions and Retinal Nonperfusion Using Widefield Multimodal Imaging

    Stino, H., et al. Association of Diabetic Lesions and Retinal Nonperfusion Using Widefield Multimodal Imaging. Ophthalmology. Retina 7, 1042-1050 (2023)

    work page 2023

  4. [4]

    & Michaelides, M

    Rahman, N., Georgiou, M., Khan, K.N. & Michaelides, M. Macular dystrophies: clinical and imaging features, molecular genetics and therapeutic options. The British journal of ophthalmology 104, 451-460 (2020)

    work page 2020

  5. [5]

    Artificial intelligence and deep learning in ophthalmology

    Ting, D.S.W., et al. Artificial intelligence and deep learning in ophthalmology. The British journal of ophthalmology 103, 167-175 (2019)

    work page 2019

  6. [6]

    Artificial Intelligence for Screening of Multiple Retinal and Optic Nerve Diseases

    Dong, L., et al. Artificial Intelligence for Screening of Multiple Retinal and Optic Nerve Diseases. JAMA network open 5, e229960 (2022)

    work page 2022

  7. [7]

    Policy-Driven, Multimodal Deep Learning for Predicting Visual Fields from the Optic Disc and OCT Imaging

    Kihara, Y., et al. Policy-Driven, Multimodal Deep Learning for Predicting Visual Fields from the Optic Disc and OCT Imaging. Ophthalmology 129, 781-791 (2022)

    work page 2022

  8. [8]

    & Hardalaç, F

    Özdaş, M.B., Uysal, F. & Hardalaç, F. Classification of Retinal Diseases in Optical Coherence Tomography Images Using Artificial Intelligence and Firefly Algorithm. Diagnostics (Basel, Switzerland) 13(2023)

    work page 2023

Show all 42 references
  1. [9]

    Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks

    Cen, L.P., et al. Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks. Nature communications 12, 4828 (2021)

    2021

  2. [10]

    & Ting, D.S.W

    Aung, Y.Y.M., Wong, D.C.S. & Ting, D.S.W. The promise of artificial intelligence: a review of the opportunities and challenges of artificial intelligence in healthcare. British medical bulletin 139, 4-15 (2021)

    2021

  3. [11]

    AI recognition of patient race in medical imaging: a modelling study

    Gichoya, J.W., et al. AI recognition of patient race in medical imaging: a modelling study. The Lancet. Digital health 4, e406-e414 (2022)

    2022

  4. [12]

    & Khoshgoftaar, T.M.J.J.o.B.D

    Shorten, C. & Khoshgoftaar, T.M.J.J.o.B.D. A survey on Image Data Augmentation for Deep Learning. 6, 1-48 (2019)

    2019

  5. [13]

    Oversampling in Health Surveys: Why, When, and How? American journal of public health 107, 1214-1215 (2017)

    Vaughan, R. Oversampling in Health Surveys: Why, When, and How? American journal of public health 107, 1214-1215 (2017)

    2017

  6. [14]

    & Chandra, R

    Khan, A.A., Chaudhari, O. & Chandra, R. A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation. Expert Systems with Applications 244, 122778 (2024)

    2024

  7. [15]

    Translating color fundus photography to indocyanine green angiography using deep-learning for age-related macular degeneration screening

    Chen, R., et al. Translating color fundus photography to indocyanine green angiography using deep-learning for age-related macular degeneration screening. NPJ Digit Med 7, 34 (2024)

    2024

  8. [16]

    Translation of Color Fundus Photography into Fluorescein Angiography Using Deep Learning for Enhanced Diabetic Retinopathy Screening

    Shi, D., et al. Translation of Color Fundus Photography into Fluorescein Angiography Using Deep Learning for Enhanced Diabetic Retinopathy Screening. Ophthalmol Sci 3, 100401 (2023)

    2023

  9. [17]

    Data augmentation for patch-based OCT chorio-retinal segmentation using generative adversarial networks

    Kugelman, J., et al. Data augmentation for patch-based OCT chorio-retinal segmentation using generative adversarial networks. 33, 7393 - 7408 (2021)

    2021

  10. [18]

    & Kim, H.K

    Yoo, T.K., Choi, J.Y. & Kim, H.K. Feasibility study to improve deep learning in OCT diagnosis of rare retinal diseases with few-shot classification. Medical & biological engineering & computing 59, 401-415 (2021)

    2021

  11. [19]

    & Keane, P.A

    Sonmez, S.C., Sevgi, M., Antaki, F., Huemer, J. & Keane, P.A. Generative artificial intelligence in ophthalmology: current innovations, future applications and challenges. Br J Ophthalmol 108, 1335-1340 (2024)

    2024

  12. [20]

    A foundation model for generalizable disease detection from retinal images

    Zhou, Y., et al. A foundation model for generalizable disease detection from retinal images. Nature 622, 156-163 (2023)

    2023

  13. [21]

    High-Resolution Image Synthesis with Latent Diffusion Models

    Rombach, R., et al. High-Resolution Image Synthesis with Latent Diffusion Models. 10674-10685 (2021)

    2021

  14. [22]

    & Krishnan, D.J.A

    Tian, Y., Fan, L., Isola, P., Chang, H. & Krishnan, D.J.A. StableRep: Synthetic Images from Text-to-Image Models Make Strong Visual Representation Learners. abs/2306.00984(2023)

    2023 arXiv

  15. [23]

    Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs

    Gulshan, V., et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. Jama 316, 2402-2410 (2016)

    2016

  16. [24]

    OCTDL: Optical Coherence Tomography Dataset for Image-Based Deep Learning Methods

    Kulyabin, M., et al. OCTDL: Optical Coherence Tomography Dataset for Image-Based Deep Learning Methods. Scientific data 11, 365 (2024)

    2024

  17. [25]

    REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs

    Orlando, J.I., et al. REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs. Medical image analysis 59, 101570 (2020)

    2020

  18. [26]

    ORIGA(-light): an online retinal fundus image database for glaucoma analysis and research

    Zhang, Z., et al. ORIGA(-light): an online retinal fundus image database for glaucoma analysis and research. Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conferen...

    2010

  19. [27]

    & Gonzalez-Hernandez, M.J.t.I.S.o.C.-B.M.S

    Fumero, F., Alayón, S., Sánchez, J.L., Sigut, J.F. & Gonzalez-Hernandez, M.J.t.I.S.o.C.-B.M.S. RIM-ONE: An open retinal image database for optic nerve evaluation. 1-6 (2011)

    2011

  20. [28]

    Drishti-GS: Retinal image dataset for optic nerve head(ONH) segmentation

    Sivaswamy, J., et al. Drishti-GS: Retinal image dataset for optic nerve head(ONH) segmentation. 53-56 (2014)

    2014

  21. [29]

    GAMMA Challenge: Glaucoma grAding from Multi-Modality imAges

    Wu, J., et al. GAMMA Challenge: Glaucoma grAding from Multi-Modality imAges. 90, 102938 (2022)

    2022

  22. [30]

    Classifier-Free Diffusion Guidance

    Ho, J. Classifier-Free Diffusion Guidance. ArXiv abs/2207.12598(2022)

    2022 arXiv

  23. [31]

    Evaluating Text-to-Visual Generation with Image-to-Text Generation

    Lin, Z., et al. Evaluating Text-to-Visual Generation with Image-to-Text Generation. ArXiv abs/2404.01291(2024)

    2024 arXiv

  24. [32]

    ChatFFA: An ophthalmic chat system for unified vision-language understanding and question answering for fundus fluorescein angiography

    Chen, X., et al. ChatFFA: An ophthalmic chat system for unified vision-language understanding and question answering for fundus fluorescein angiography. iScience 27, 110021 (2024)

    2024

  25. [33]

    & Choi, Y

    Hessel, J., Holtzman, A., Forbes, M., Le Bras, R. & Choi, Y. CLIPScore: A Reference-free Evaluation Metric for Image Captioning. ArXiv abs/2104.08718(2021)

    2021 arXiv

  26. [34]

    Overcoming challenges in research and development of rare eye diseases

    Sharma, M. Overcoming challenges in research and development of rare eye diseases. Indian J Ophthalmol 70, 2214-2215 (2022)

    2022

  27. [35]

    EyeFound: A Multimodal Generalist Foundation Model for Ophthalmic Imaging

    Shi, D., et al. EyeFound: A Multimodal Generalist Foundation Model for Ophthalmic Imaging. ArXiv abs/2405.11338(2024)

    2024 arXiv

  28. [36]

    EyeCLIP: A visual-language foundation model for multi-modal ophthalmic image analysis

    Shi, D., et al. EyeCLIP: A visual-language foundation model for multi-modal ophthalmic image analysis. (2024)

    2024

  29. [37]

    Diffusion models in medical imaging: A comprehensive survey

    Kazerouni, A., et al. Diffusion models in medical imaging: A comprehensive survey. Medical image analysis 88, 102846 (2023)

    2023

  30. [38]

    Bridging the Camera Domain Gap With Image-to-Image Translation Improves Glaucoma Diagnosis

    He, S., et al. Bridging the Camera Domain Gap With Image-to-Image Translation Improves Glaucoma Diagnosis. Transl Vis Sci Technol 12, 20-20 (2023)

    2023

  31. [39]

    Song, F., Zhang, W., Zheng, Y., Shi, D. & He, M. A deep learning model for generating fundus autofluorescence images from color fundus photography. Adv Ophthalmol Pract Res 3, 192-198 (2023)

    2023

  32. [40]

    Shi, D., He, S., Yang, J., Zheng, Y. & He, M. One-shot Retinal Artery and Vein Segmentation via Cross-modality Pretraining. Ophthalmol Sci 4, 100363 (2024)

    2024

  33. [41]

    Fundus2Video: Cross-Modal Angiography Video Generation from Static Fundus Photography with Clinical Knowledge Guidance

    Zhang, W., et al. Fundus2Video: Cross-Modal Angiography Video Generation from Static Fundus Photography with Clinical Knowledge Guidance. in Medical Image Computing and Computer Assisted Intervention – MICCAI 689-699 (Springer Nature Switzerland, Morocco, 2024)

    2024

  34. [42]

    & Nichol, A.J.A

    Dhariwal, P. & Nichol, A.J.A. Diffusion Models Beat GANs on Image Synthesis. abs/2105.05233(2021). Figure legends Figure 1. Flow diagramof this study. A. Characteristics of training datasets. B. The architecture of the Text-to-image diffusion model. (i) Variational Autoencoder...

    2021 arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.