Robust Incomplete-Modality Alignment for Ophthalmic Disease Grading and Diagnosis via Labeled Optimal Transport

Cheng Chen; Jianyang Xie; Jun Cheng; Lijun Zhang; Liming Chen; Lu Liu; Qinkai Yu; Yalin Zheng; Yanda Meng; Yitian Zhao

REVIEW 4 major objections 5 minor 24 references

Robust Incomplete-Modality Alignment for Ophthalmic Disease Grading and Diagnosis via Labeled Optimal Transport

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A class-constrained optimal transport alignment between fundus and OCT images lets a model keep its diagnostic accuracy when one eye-imaging modality is entirely missing at test time.

desk verdict Plausible and well-motivated missing-modality framework with strong reported numbers, but the test-time protocol for a single modality is unspecified and the feature-wise OT equation doesn't type-check. read the letter →

arxiv 2507.04999 v1 pith:T5KTDTPH submitted 2025-07-07 cs.CV

Qinkai Yu , Jianyang Xie , Yitian Zhao , Cheng Chen , Lijun Zhang , Liming Chen , Jun Cheng , Lu Liu

show 2 more authors

Yalin Zheng Yanda Meng

This is my paper · ORCID

classification cs.CV

keywords missingmodalityoptimaltransportophthalmicdiseasegradingfundusimagingopticalcoherencetomographymultimodalalignmentasymmetricfusionGromov-Wasserstein

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

This paper argues that a multimodal ophthalmic diagnosis model can be made robust to a fully missing modality by aligning fundus and OCT features with a class-constrained optimal transport plan, rather than by imputing images or distilling from a teacher. The proposed framework learns soft same-class matches between modalities, builds class prototypes from those matches, and uses an asymmetric fusion strategy that treats the fundus as global and OCT as localized. On three Harvard-30k datasets, the authors report the best accuracy, AUC, and F1 in both complete-modality fusion and inter-modality missing settings, including gains of 4-5% AUC over the IMDR baseline when one modality is absent. The point of the work is that paired training data can be exploited so strongly that at test time one modality alone suffices.

What carries the argument

The central object is Labeled Optimal Transport, a Gromov-Wasserstein transport plan constrained by $T_{ij}>0 \Rightarrow y_i^f=y_j^o$, so mass only flows between features of the same class. It is solved with entropy-regularized Sinkhorn iterations and then drives two alignment levels: class-wise alignment, where repeated sampling from the plan's row probabilities gives a soft prototype $c_i^o = \sum_j p(j|i) e_j^o$ used in a cosine loss, and feature-wise alignment, where the plan produces an 'OT fundus' feature $T_v^\top e_o$ for distribution matching. On top of this, an asymmetric fusion path concatenates cross-modal, semantic, and backbone features for the fundus through an attention layer, while OCT uses only semantic and backbone features through MLPs to avoid overfitting noise. The machinery's job is to replace image-level imputation and strict one-to-one matching with a soft, class-aware correspondence that can be exploited when one modality is absent.

What would settle it

Train two ResNet-50 baselines, one on fundus only and one on OCT only, using the same Harvard-30k splits, and compare against the proposed model trained only on paired data then tested with one modality missing. If the paired-trained model does not beat the single-modality baselines on the respective single-modality test sets, the learned alignment is not transferring and the robustness claim fails.

Watch

Extended reading notes

Core claim

The central claim is that label-constrained optimal transport alignment is enough to carry ophthalmic diagnosis when one modality is absent. The transport plan is restricted so that mass moves only between same-class samples, $T_{ij}>0 \Rightarrow y_i^f=y_j^o$, and then used to build soft class prototypes and a feature-wise 'OT fundus' representation. The authors report that this framework beats the compared methods in complete-modality fusion and in inter-modality missing settings on Harvard-30k AMD, DR, and Glaucoma, and the ablation attributes the largest share of the gain to class-wise alignment, which costs 7-8% AUC on DR when removed.

Load-bearing premise

The method's robustness rests on the assumption that the class-constrained optimal transport alignment learned from fully paired training data transfers to a classifier at test time when an entire modality is absent, even though the inter-modality missing training protocol never shows the model a missing-modality input.

Editorial extensions

If this is right

A model trained once on paired fundus/OCT data could be deployed in clinics that have only a fundus camera or only an OCT scanner, without retraining on single-modality data.
Because the alignment is class-constrained and soft, it captures within-class variation in lesions rather than forcing one-to-one matches, which the ablation attributes 7-8% AUC on DR to when removed.
The asymmetry principle, global feature transport toward fundus and class-only transport toward OCT, offers a design rule for other paired modalities with different noise and coverage profiles.
Under proportional random missing rates from 0% to 100%, the framework's AUC and F1 stay relatively flat compared with the B-CNN baseline, so the method is intended for real clinics with partially missing data.

Reading between the lines

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

An implication the authors leave implicit is that the same asymmetric optimal-transport alignment could transfer to any paired-modality medical task where one view is global and the other is localized or noisy, such as MRI with pathology slides or CT with endoscopy.
A simpler representation-level imputer that learns a conditional mean of the target modality given the source could serve as a baseline to isolate what the optimal-transport constraint adds, since the soft prototype is essentially a class-conditioned expected match.
The class constraint depends on training labels, so label noise would corrupt the transport plan; testing under noisy or partial labels would reveal whether the claimed robustness degrades gracefully.
The paper does not report single-modality-only training baselines, so an external comparison against fundus-only and OCT-only ResNet-50 models would clarify whether the reported gains come from the alignment or from the shared training recipe.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

What you should know: this is a reasonable, well-motivated paper on handling missing OCT/fundus at test time, built around labeled optimal transport and an asymmetric fusion. The genuinely new bits are the combination of class-wise prototype alignment with feature-wise transport and the asymmetry between modalities; the labeled OT core is borrowed from Ryu et al. [11] and GWOT is standard. The reported numbers are consistently strong across three Harvard-30k subsets, and the ablation study shows the components matter. Code is promised.

The soft spots, in order of importance. First, the test-time protocol for inter-modality missing is unspecified. Section 3 says only 'Train on a complete training set and evaluate on test set with single modality.' The model as drawn has two encoders and a fusion block; when OCT is absent, the OCT stream has no input. The caption mentions a 'Missing Feature' but the text never says what role it plays. If the missing stream is zero-filled, or the cross-modal terms are dropped, the robustness could be coming from that choice, not from the OT alignment. This is load-bearing for the central claim.

Second, Eq. 3 and the 'OT fundus' definition don't type-check as printed. T_v is declared in R^{Do x Df}, then used with indices k,l that would have to be sample indices in the GW distance, and T_v^T e_o is a shape mismatch when e_o is N x Do. This may be a notation slip, but as written the feature-wise alignment cannot be implemented, and feature-wise alignment contributes a measurable chunk in Table 2.

Third, the results tables report no standard deviations or significance tests, and the baselines are taken wholesale from Liu et al. [7]. That's a real limitation but not fatal; the gap to IMDR is large enough that it probably holds.

On circularity: I'm with the reader — this is supervised self-referential training, not a circular evaluation. The class constraints use labels, but the method isn't relabeling fitted constants as predictions.

Who this is for: researchers working on missing-modality medical imaging, especially ophthalmology, and people building on labeled OT for cross-modal alignment. It deserves a serious referee; the empirical story is plausible and the released code will let reviewers resolve the test-time question. My recommendation: send to peer review, but the revision must specify the missing-modality input handling and fix Eq. 3, or the robustness claim stays unverifiable.

Referee Report

4 major / 5 minor

Summary. The manuscript proposes RIMA, a multimodal alignment and fusion framework for ophthalmic disease grading and diagnosis from color fundus images and OCT scans. The core idea is to use labeled optimal transport to align features across modalities at two scales: class-wise alignment via soft prototypes, and feature-wise alignment via a Gromov-Wasserstein transport plan, followed by an asymmetric fusion strategy that treats the two modalities differently because fundus features are global while OCT features are noisy and localized. The authors report experiments on three Harvard-30k datasets (AMD, DR, Glaucoma) under three settings: complete-modality fusion, inter-modality missing (one modality entirely absent at test time), and proportional random missing (partial missing during both training and test). They claim state-of-the-art performance in the first two settings, with the abstract highlighting a 4–5% AUC improvement over the IMDR baseline when one modality is missing.

Significance. If the reported results are reproducible, the paper addresses a clinically important problem: robust ophthalmic diagnosis when one of two imaging modalities is unavailable. The labeled-OT formulation with class constraints is a reasonable and interesting extension of standard GWOT, and the asymmetric fusion is motivated by domain-specific characteristics of fundus and OCT. The paper also makes a concrete effort to evaluate on three real-world ophthalmic datasets and reports improvements over several existing methods. However, the current manuscript has a load-bearing reproducibility gap: the test-time protocol for a fully missing modality is never specified, and the feature-wise alignment equation contains a dimensional inconsistency that prevents implementation as written. The lack of standard deviations or significance tests further weakens the empirical claims. These issues need to be resolved before the central claim of robustness under inter-modality incompleteness can be accepted.

major comments (4)

[Section 3, Inter-Modality Missing; Section 2.3] The test-time protocol for a single-modality input is unspecified. The model description in Section 2 assumes both encoders produce features e_f and e_o, and the asymmetric fusion in Section 2.3 concatenates backbone, class-prototype, and feature-transport features from both streams. When one modality is absent at test time, there is no e_o (or e_f), no transport plan T_v, and no soft prototype, so the fusion cannot be computed as written. The figure labels a 'Missing Feature' but no text defines its role. The central robustness claim—exceeding IMDR by 4–5% AUC when one modality is missing—cannot be reproduced or attributed to OT alignment without a clear statement of how the network handles the absent encoder (e.g., zero-padding its output, a learned imputation branch, or dropping cross-modal terms). Please specify the exact inference path and, if the model is trained only on complete pairs, explain how the fusion adapts to a single-modality input.
[Equation (3)] Equation (3) is not implementable as written. T_v is defined as belonging to R^{D_o × D_f}, but the objective sums over sample indices i,j,k,l and multiplies by T_of_{ci,j} T_v_{k,l}, which implies T_v is indexed by samples. Moreover, the expression 'OT fundus = T_v^T e_o' is dimensionally inconsistent when e_o ∈ R^{N × D_o} and T_v ∈ R^{D_o × D_f}. Please provide the correct dimensions and index conventions for T_v, and clarify how the feature-wise transport plan is used to produce the OT fundus feature. This is not a cosmetic issue; without a consistent definition, the feature-wise alignment mechanism cannot be implemented from the paper.
[Table 1] The empirical evaluation lacks measures of variability and statistical testing. The paper states that five-fold cross-validation was used, but Table 1 reports only point estimates. The baseline comparisons are taken from a prior paper (Liu et al. [7]) without any variance information, so the claimed improvements (e.g., exceeding IMDR by 4–5% AUC in the inter-modality missing setting) cannot be assessed for statistical significance. Please report per-fold means and standard deviations for all methods, and, where feasible, paired significance tests for the key comparisons.
[Section 3, Proportional Random Missing] The training protocol for the proportional random missing setting is not described. The text says 'Train on the training set with a partial proportion of missing modalities,' but it does not specify how the network is trained when a modality is missing (e.g., whether the same imputation/handling mechanism as in the inter-modality missing setting is used). This matters because the proportional random missing experiments are used to claim robustness in Figure 3, which only compares against B-CNN and does not include other state-of-the-art methods. Please clarify the training procedure for this setting and include comparisons with at least one more baseline.

minor comments (5)

[Section 2.1, Equation (1)] The notation is confusing: the feasible set T_{µ,ν} is defined earlier, then ̅T_{µ,ν} is introduced with the same symbol T used for both the set and the transport plan. Please use distinct notation to avoid ambiguity.
[Table 2] The ablation table is garbled; rows for 'w/o Class-wise Alignment' and 'w/o Feature-wise Alignment' do not clearly indicate which modality is being evaluated, and the numbers seem misaligned. Please reformat the table so that each row clearly indicates the modality and the metrics.
[Figure 1] The 'Missing Feature' label in Figure 1 is not defined anywhere in the text. Please either remove it or explain its role in the architecture.
[Section 3, Implementation Details] The OCT input is resized to 96×96×96, but the paper does not specify how the volume is sampled (number of B-scans, slice selection, or preprocessing). Please add these details for reproducibility.
[Equation (5)] There is a typo in the summation index: 'nX' should be 'N'. Also, the cosine-similarity loss should clarify whether the sum is over all samples in the batch.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the OT alignment losses are self-referential training objectives, but all reported predictions are evaluated on held-out test sets against external baselines.

full rationale

The paper's contribution is an empirical training framework, not a derivation of test results from first principles. The class-wise alignment loss in Eq. (5) is minimized against a soft prototype in Eq. (4) that is computed from the current encoders' outputs and the label-constrained OT plan of Eq. (2); in that narrow sense the target is a moving function of the features being aligned. This is a self-referential, self-training-style objective common in representation learning, but it is not a relabeled prediction: the disease-grade predictions come from a classifier trained with cross-entropy and evaluated on held-out folds. The comparison to IMDR [7] uses numbers from a prior paper with overlapping authors, but those are published benchmark results and the present model is run under the same experimental setting; no load-bearing argument reduces to an unverified self-citation. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported, and no claimed result is defined into existence. The main weakness is an underspecified test-time protocol for the inter-modality missing setting: the paper does not describe how the network behaves when one encoder is absent, which is a reproducibility concern rather than a circularity concern. Overall, the derivation chain is self-contained against external benchmarks and no significant circularity is found.

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

The framework introduces no new physical or categorical entities. It rests on standard deep feature extractors, a class-constrained GWOT solver, and a hand-chosen asymmetric alignment direction. The main unstated degrees of freedom are the OT solver hyperparameters and the test-time masking policy for missing modalities.

free parameters (3)

Entropy regularization epsilon for OT = not reported
Controls the smoothness of all Sinkhorn based transport plans in Equations 2 and 3; its value is not stated and is likely tuned.
Alignment loss weights = 1 (default)
All terms in L_total are weighted equally in Equation 6; whether equal weighting is optimal is not explored.
Number of Sinkhorn iterations = not reported
Affects convergence and stability of the GW-like OT solvers; not given in the text.

assumptions (3)

domain assumption Class-constrained transport, where T_ij > 0 only if y_i^f = y_j^o, adequately captures semantic correspondence between modalities.
Used in Equation 1 to define the feasible transport plan; relies on the premise that same-class samples are the only ones that should align.
standard math Gromov-Wasserstein distance between intra-modality sample distances is a valid alignment objective for cross-modal features.
Equations 2 and 3 assume GW distance behaves as a cost for aligning heterogeneous features, a standard OT assumption.
ad hoc to paper Feature-wise alignment is only beneficial in the OCT-to-fundus direction because fundus features are global and OCT features are noisy and localized.
Section 2.1 justifies the asymmetry with a heuristic claim about modality properties; no experiment demonstrates the reverse direction fails.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Robust Incomplete-Modality Alignment for Ophthalmic Disease Grading and Diagnosis via Labeled Optimal Transport." pith.science (2026). https://pith.science/paper/T5KTDTPH

@misc{pith2026250704999,
  author       = {Pith},
  title        = {Pith review of: Robust Incomplete-Modality Alignment for Ophthalmic Disease Grading and Diagnosis via Labeled Optimal Transport},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T5KTDTPH}},
  note         = {Machine review of arXiv:2507.04999}
}

read the original abstract

Multimodal ophthalmic imaging-based diagnosis integrates color fundus image with optical coherence tomography (OCT) to provide a comprehensive view of ocular pathologies. However, the uneven global distribution of healthcare resources often results in real-world clinical scenarios encountering incomplete multimodal data, which significantly compromises diagnostic accuracy. Existing commonly used pipelines, such as modality imputation and distillation methods, face notable limitations: 1)Imputation methods struggle with accurately reconstructing key lesion features, since OCT lesions are localized, while fundus images vary in style. 2)distillation methods rely heavily on fully paired multimodal training data. To address these challenges, we propose a novel multimodal alignment and fusion framework capable of robustly handling missing modalities in the task of ophthalmic diagnostics. By considering the distinctive feature characteristics of OCT and fundus images, we emphasize the alignment of semantic features within the same category and explicitly learn soft matching between modalities, allowing the missing modality to utilize existing modality information, achieving robust cross-modal feature alignment under the missing modality. Specifically, we leverage the Optimal Transport for multi-scale modality feature alignment: class-wise alignment through predicted class prototypes and feature-wise alignment via cross-modal shared feature transport. Furthermore, we propose an asymmetric fusion strategy that effectively exploits the distinct characteristics of OCT and fundus modalities. Extensive evaluations on three large ophthalmic multimodal datasets demonstrate our model's superior performance under various modality-incomplete scenarios, achieving Sota performance in both complete modality and inter-modality incompleteness conditions. Code is available at https://github.com/Qinkaiyu/RIMA

Figures

Figures reproduced from arXiv: 2507.04999 by the authors.

**Figure 2.** A t-SNE visualization of the feature distribution from B-CNN and our [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

**Figure 3.** Comparison of B-CNN and Ours in proportional random missing exper [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 16 canonical work pages

[11]

Cross-modality Matching and Prediction of Perturbation Responses with Labeled Gromov-Wasserstein Optimal Transport

Ryu, J., Bunne, C., Pinello, L., Regev, A., Lopez, R.: Cross-modality matching and prediction of perturbation responses with labeled gromov-wasserstein optimal transport. arXiv preprint arXiv:2405.00838 (2024)

work page Pith review arXiv 2024
[7]

In: Proceedings of the AAAI Conference on Artificial Intelli- gence

Liu, c., Huang, z., Chen, Z., Tang, F., Tian, Y., Xu, Z., Luo, Z., Zheng, Y., Meng, Y.: Incomplete modality disentangled representation for ophthalmic disease grad- ing and diagnosis. In: Proceedings of the AAAI Conference on Artificial Intelli- gence. vol. 38, pp. 6225–6233 (2025) 10 Q.Yu et al

work page 2025
[1]

Advances in Neural Information Processing Systems 35, 39090–39102 (2022)

Cao, Z., Xu, Q., Yang, Z., He, Y., Cao, X., Huang, Q.: Otkge: Multi-modal knowl- edge graph embeddings via optimal transport. Advances in Neural Information Processing Systems 35, 39090–39102 (2022)

work page 2022
[2]

International journal of retina and vitreous 1, 1–15 (2015)

De Carlo, T.E., Romano, A., Waheed, N.K., Duker, J.S.: A review of optical coher- ence tomography angiography (octa). International journal of retina and vitreous 1, 1–15 (2015)

work page 2015
[3]

In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

Duan, J., Chen, L., Tran, S., Yang, J., Xu, Y., Zeng, B., Chilimbi, T.: Multi- modal alignment using representation codebook. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 15651–15660 (2022)

work page 2022
[4]

Investigative ophthalmology & visual science57(9), OCT1–OCT13 (2016)

Fujimoto, J., Swanson, E.: The development, commercialization, and impact of optical coherence tomography. Investigative ophthalmology & visual science57(9), OCT1–OCT13 (2016)

work page 2016
[5]

He,K.,Zhang,X.,Ren,S.,Sun,J.:Deepresiduallearningforimagerecognition.In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 770–778 (2016)

2016
[6]

IEEE Journal of Biomedical and Health Informatics25(7), 2686–2697 (2020)

Hua, C.H., Kim, K., Huynh-The, T., You, J.I., Yu, S.Y., Le-Tien, T., Bae, S.H., Lee, S.: Convolutional network with twofold feature augmentation for diabetic retinopathy recognition from multi-modal images. IEEE Journal of Biomedical and Health Informatics25(7), 2686–2697 (2020)

work page 2020

Show all 24 references

[8]

arXiv preprint arXiv:2310.02492 (2023)

Luo, Y., Tian, Y., Shi, M., Elze, T., Wang, M.: Eye fairness: A large-scale 3d imaging dataset for equitable eye diseases screening and fair identity scaling. arXiv preprint arXiv:2310.02492 (2023)

2023 arXiv
[9]

In: International conference on machine learning

Peyré, G., Cuturi, M., Solomon, J.: Gromov-wasserstein averaging of kernel and distance matrices. In: International conference on machine learning. pp. 2664–2672. PMLR (2016)

2016
[10]

Foundations and Trends® in Machine Learning11(5-6), 355–607 (2019)

Peyré, G., Cuturi, M., et al.: Computational optimal transport: With applications to data science. Foundations and Trends® in Machine Learning11(5-6), 355–607 (2019)

2019
[12]

arXiv preprint arXiv:2402.05294 (2024)

Saha, P., Mishra, D., Wagner, F., Kamnitsas, K., Noble, J.A.: Examining modality incongruityinmultimodalfederatedlearningformedicalvisionandlanguage-based disease detection. arXiv preprint arXiv:2402.05294 (2024)

2024 arXiv
[13]

Pacific Journal of Mathematics21(2), 343–348 (1967)

Sinkhorn, R., Knopp, P.: Concerning nonnegative matrices and doubly stochastic matrices. Pacific Journal of Mathematics21(2), 343–348 (1967)

1967
[14]

Informa- tion Fusion110, 102454 (2024)

Sun,Y.,Liu,Z.,Sheng,Q.Z.,Chu,D.,Yu,J.,Sun,H.:Similarmodalitycompletion- based multimodal sentiment analysis under uncertain missing modalities. Informa- tion Fusion110, 102454 (2024)

2024
[15]

Advances in Neural Information Processing Systems (2017)

Vaswani, A.: Attention is all you need. Advances in Neural Information Processing Systems (2017)

2017
[16]

Advances in Neural Information Pro- cessing Systems 36 (2024)

Wang, D., Li, M., Liu, X., Xu, M., Chen, B., Zhang, H.: Tuning multi-mode token- level prompt alignment across modalities. Advances in Neural Information Pro- cessing Systems 36 (2024)

2024
[17]

In: 2021 IEEE/CVF InternationalConferenceonComputerVision(ICCV).pp.1793–1802.IEEE(2021)

Wang, Y., Zhang, T., Zhang, X., Cui, Z., Huang, Y., Shen, P., Li, S., Yang, J.: Wasserstein coupled graph learning for cross-modal retrieval. In: 2021 IEEE/CVF InternationalConferenceonComputerVision(ICCV).pp.1793–1802.IEEE(2021)

2021
[18]

In: Proceedings of the European conference on computer vision (ECCV)

Woo, S., Park, J., Lee, J.Y., Kweon, I.S.: Cbam: Convolutional block attention module. In: Proceedings of the European conference on computer vision (ECCV). pp. 3–19 (2018)

2018
[19]

Medical Image Analysis 90, 102938 (2023)

Wu, J., Fang, H., Li, F., Fu, H., Lin, F., Li, J., Huang, Y., Yu, Q., Song, S., Xu, X., et al.: Gamma challenge: glaucoma grading from multi-modality images. Medical Image Analysis 90, 102938 (2023)

2023
[20]

arXiv preprint arXiv:2409.07825 (2024)

Wu, R., Wang, H., Chen, H.T., Carneiro, G.: Deep multimodal learning with miss- ing modality: A survey. arXiv preprint arXiv:2409.07825 (2024)

2024 arXiv
[21]

In: Proceedings of the IEEE/CVF International Conference on Computer Vision

Xu, Y., Chen, H.: Multimodal optimal transport-based co-attention transformer with global structure consistency for survival prediction. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 21241–21251 (2023)

2023
[22]

In: Medical Imaging With Deep Learning

Yang, Y., Chen, H., Chang, Z., Xiang, Y., Ye, C., Ma, T.: Incomplete learning of multi-modalconnectomeforbraindisorderdiagnosisviamodal-mixupanddeepsu- pervision. In: Medical Imaging With Deep Learning. pp. 1006–1018. PMLR (2024)

2024
[23]

Computer Methods and Pro- grams in Biomedicine229, 107307 (2023)

Zheng, J., Liu, H., Feng, Y., Xu, J., Zhao, L.: Casf-net: Cross-attention and cross- scale fusion network for medical image segmentation. Computer Methods and Pro- grams in Biomedicine229, 107307 (2023)

2023
[24]

Medical Image Analysis 96, 103214 (2024)

Zou, K., Lin, T., Han, Z., Wang, M., Yuan, X., Chen, H., Zhang, C., Shen, X., Fu, H.: Confidence-aware multi-modality learning for eye disease screening. Medical Image Analysis 96, 103214 (2024)

2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [11]

Cross-modality Matching and Prediction of Perturbation Responses with Labeled Gromov-Wasserstein Optimal Transport

Ryu, J., Bunne, C., Pinello, L., Regev, A., Lopez, R.: Cross-modality matching and prediction of perturbation responses with labeled gromov-wasserstein optimal transport. arXiv preprint arXiv:2405.00838 (2024)

work page Pith review arXiv 2024

[2] [7]

In: Proceedings of the AAAI Conference on Artificial Intelli- gence

Liu, c., Huang, z., Chen, Z., Tang, F., Tian, Y., Xu, Z., Luo, Z., Zheng, Y., Meng, Y.: Incomplete modality disentangled representation for ophthalmic disease grad- ing and diagnosis. In: Proceedings of the AAAI Conference on Artificial Intelli- gence. vol. 38, pp. 6225–6233 (2025) 10 Q.Yu et al

work page 2025

[3] [1]

Advances in Neural Information Processing Systems 35, 39090–39102 (2022)

Cao, Z., Xu, Q., Yang, Z., He, Y., Cao, X., Huang, Q.: Otkge: Multi-modal knowl- edge graph embeddings via optimal transport. Advances in Neural Information Processing Systems 35, 39090–39102 (2022)

work page 2022

[4] [2]

International journal of retina and vitreous 1, 1–15 (2015)

De Carlo, T.E., Romano, A., Waheed, N.K., Duker, J.S.: A review of optical coher- ence tomography angiography (octa). International journal of retina and vitreous 1, 1–15 (2015)

work page 2015

[5] [3]

In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

Duan, J., Chen, L., Tran, S., Yang, J., Xu, Y., Zeng, B., Chilimbi, T.: Multi- modal alignment using representation codebook. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 15651–15660 (2022)

work page 2022

[6] [4]

Investigative ophthalmology & visual science57(9), OCT1–OCT13 (2016)

Fujimoto, J., Swanson, E.: The development, commercialization, and impact of optical coherence tomography. Investigative ophthalmology & visual science57(9), OCT1–OCT13 (2016)

work page 2016

[7] [5]

He,K.,Zhang,X.,Ren,S.,Sun,J.:Deepresiduallearningforimagerecognition.In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 770–778 (2016)

2016

[8] [6]

IEEE Journal of Biomedical and Health Informatics25(7), 2686–2697 (2020)

Hua, C.H., Kim, K., Huynh-The, T., You, J.I., Yu, S.Y., Le-Tien, T., Bae, S.H., Lee, S.: Convolutional network with twofold feature augmentation for diabetic retinopathy recognition from multi-modal images. IEEE Journal of Biomedical and Health Informatics25(7), 2686–2697 (2020)

work page 2020