Pith. sign in

REVIEW 6 major objections 6 minor 64 references

D-Cube: Exploiting Hyper-Features of Diffusion Model for Robust Medical Classification

T0 review · 6 major / 6 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 D-Cube claims that frozen diffusion-model features, selected by testing which layers deviate from Gaussian noise, set new high scores on pancreas CT, breast MRI, and COVID chest X-ray classification.

desk verdict The paper's own ablation undercuts its central claim: the largest gain comes from a fine-tuned ResNet sub-feature stream, not from diffusion hyper-features. read the letter →

arxiv 2411.11087 v1 pith:YKXSFBUX submitted 2024-11-17 cs.CV

Minhee Jang , Juheon Son , Thanaporn Viriyasaranon , Junho Kim , Jang-Hwan Choi This is my paper · ORCID
classification cs.CV
keywords diffusionmodelsmedicalimageclassificationfeatureselectionGaussianitymetriccontrastivelearningcancerdiagnosissyntheticdatageneration
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 sets out to show that a diffusion model can do more than generate synthetic medical images: its internal feature maps, once the noise-dominated layers are filtered out, carry the semantic information needed for cancer classification. The proposed D-Cube pipeline first trains a class-conditional diffusion model with an added contrastive loss, then freezes it and trains a classifier on a concatenation of selected diffusion features and fine-tuned CNN sub-features. On pancreas CT, breast MRI, and COVID chest X-ray, the paper reports accuracies of 93.61%, 77.98%, and 96.28%, beating the compared CNN, transformer, and diffusion baselines on almost all metrics. The ablation attributes the largest single gain to the CNN sub-features, with the Gaussianity-based layer selection and the tailored losses adding smaller but consistent improvements.

What carries the argument

The load-bearing object is the Gaussianity metric for layer selection: a Kolmogorov-Smirnov test that compares the empirical cumulative distribution of each diffusion layer's feature map, computed over a batch of 256 images at timestep t=100, with the CDF of a Gaussian. Layers with p-value above 0.05 are judged to be doing noise prediction and are excluded; layers with p-value below 0.05 are judged to preserve semantic image content and are kept. The argument is that a denoising network must contain both kinds of layers, and that the non-Gaussian ones are exactly the features worth feeding to a classifier. Around this selection, the pipeline adds a contrastive loss during diffusion training, a cycle loss that ties classifier predictions back to the frozen diffusion model's noise estimates, a consistency-regularization loss on flipped inputs, and a channel-expansion step that multiplies the single-channel diffusion features by fine-tuned CNN sub-features.

What would settle it

Run the layer-selection step again on a different random batch or a different timestep and check whether the same layers are flagged as non-Gaussian; if the selected layers change, the metric is not a stable property of the model. Also test a version where the selected diffusion features are replaced by the same number of randomly chosen layers: if random layers match the KS-selected accuracy, the Gaussianity criterion is not doing the work.

Watch

Extended reading notes

Core claim

D-Cube's central claim is that intermediate features of a class-conditional denoising diffusion model, trained with a contrastive loss and then frozen, are strong enough to drive medical image classification once the layers that merely predict noise are discarded. The paper identifies those layers with a Kolmogorov-Smirnov test: a feature map whose p-value against a Gaussian distribution is above 0.05 is treated as noise and dropped, while non-Gaussian maps are kept as semantic hyper-features. These selected features are concatenated, multiplied channel-wise by sub-features from a fine-tuned CNN, and classified by a small convolutional head trained with cross-entropy, a cycle loss that checks whether the predicted class reproduces the ground-truth diffusion noise, and a consistency-regularization loss on horizontal flips. The paper reports that this combination scores 93.61% accuracy on pancreas CT, 77.98% on breast MRI, and 96.28% on COVID X-ray, outperforming the compared baselines on almost all metrics and exceeding the diffusion baseline DiffMIC on the two cancer datasets.

Load-bearing premise

The method rests on the premise that layers whose feature maps deviate from a Gaussian distribution are the ones holding classification-relevant semantics, and that a single batch of 256 images at timestep t=100 is enough to tell which layers those are.

Editorial extensions

If this is right

  • If the central claim holds, a frozen diffusion model pretrained on medical data can be reused as a classification feature extractor, avoiding the need to fine-tune the generator for each new task.
  • The Gaussianity metric gives a training-free rule for choosing layers, replacing hand-picked layer and timestep choices that other diffusion-feature methods rely on.
  • The reported gains on the smaller breast MRI dataset suggest the method is most valuable when data are scarce and class boundaries are blurred.
  • Synthetic images produced by the contrastively trained diffusion model can be added to training sets, improving standard CNN and transformer classifiers on the pancreas dataset by up to 2-4% in recall and F1.
  • The ablation's large jump from the CNN sub-features implies that the diffusion features and the CNN features make complementary rather than redundant contributions.

Reading between the lines

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

  • A natural next test is whether the same KS-based layer selection transfers to other generative backbones, such as latent diffusion models, where the noise-dominated and semantic layers may be arranged differently.
  • Because the feature-selection p-values are computed on a single batch and a single timestep, a more robust version of the method would average the Gaussianity statistic over several batches and timesteps; if the selected layers shift across those repetitions, the selection criterion is less principled than the paper implies.
  • The ablation showing that CNN sub-features add 5.35 accuracy points suggests that a fair head-to-head should compare D-Cube against a classifier using only those sub-features with the same loss functions, to see how much of the final score is actually due to the diffusion hyper-features.
  • The method's claim of broad applicability could be probed on other modalities with similar imbalance, such as ultrasound or histopathology, where the Gaussianity pattern of diffusion layers may differ.
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.

Referee Report

6 major / 6 minor

Summary. The paper proposes D-Cube, a two-stage pipeline for medical image classification. In step 1, a class-conditional U-ViT diffusion model is trained with a contrastive auxiliary loss in addition to the standard denoising objective. In step 2, the diffusion model is frozen, feature maps from layers selected by a Kolmogorov-Smirnov Gaussianity test are fed into a small convolutional classifier, and training is regularized by a cycle loss and a consistency loss. An ImageNet-pretrained ResNet is fine-tuned to provide sub-features that are concatenated with the diffusion features. The method is evaluated on pancreas CT, breast MRI, and COVID chest X-ray datasets against CNN, transformer, and diffusion-based baselines, with additional ablation studies, synthetic-data augmentation experiments, and FID comparisons.

Significance. If the attribution were established, D-Cube would be a valuable demonstration that diffusion-model hyper-features, selected by a Gaussianity criterion, improve medical classification under small-sample and imbalanced conditions. The paper has notable strengths: it validates across three imaging modalities and populations, it includes a fairly broad baseline table, it releases code, and it provides an explicit feature-selection criterion together with FID evidence that the proposed generative loss improves image fidelity. The central problem is that the experiments do not isolate the diffusion contribution from the fine-tuned ResNet sub-features, and the headline claims are stronger than the data support, particularly on the COVID dataset where the main diffusion-based baseline scores higher. With additional controlled ablations and variance reporting, the contribution could be made convincing.

major comments (6)
  1. [Section IV-D, Table IV] The ablation does not include a diffusion-free baseline, so the paper's central attribution claim is not supported. The largest single accuracy gain in Table IV comes from adding the ResNet sub-features fsub (+5.35 over +LCls), and every full D-Cube model in Table II uses fsub. A controlled run that trains the same step-2 classifier with fsub only, and without any diffusion features, is necessary to determine whether the state-of-the-art results are driven by the diffusion hyper-features or by a fine-tuned ImageNet-initialized ResNet head. Without this baseline, the abstract and introduction's claim that diffusion features and feature selection are responsible for the improvements is confounded.
  2. [Section IV-C, Table II] The claim that D-Cube 'surpasses previous models in almost metrics across datasets' is contradicted on the COVID chest X-ray dataset: DiffMIC achieves 96.40 accuracy and 97.20 F1 versus D-Cube's 96.28 and 96.87. The text in Section IV-C also states that 'by utilizing sub-features, we outperformed DiffMIC,' which is not true for this dataset. Please revise the claim, discuss the failure, and report whether the comparison changes under multiple seeds or a paired statistical test.
  3. [Section IV, Tables II, IV, and VI] All reported numbers appear to be from single training runs, with no standard deviations, confidence intervals, or significance tests. This is particularly concerning on the Duke MRI dataset, which has only 707 test images, where the differences between the best and worst feature selections are small (Table VI) and the ablation increments are small (Table IV). I request that the main comparisons and ablations be repeated for at least 3-5 seeds and reported as mean +/- std, with a paired test such as McNemar's test or bootstrapping for the head-to-head comparisons.
  4. [Section III-B2a and Table VI] The proposed KS-based Gaussianity feature selection is not shown to be load-bearing. Replacing the 'best' layers with the 'worst' layers changes accuracy by only 0.38, 1.14, and 0.43 points on the pancreas, breast, and COVID datasets, respectively. The manuscript also does not specify how many layers were selected, which layer indices correspond to the yellow stars and red crosses in Fig. 2, or how stable the selection is across random batches, timesteps, or model initializations. Please provide this information and compare the KS-selected layers against a random selection of the same number of layers, or a fixed high-level/low-level layer choice, to establish that the criterion is more than a negligible detail.
  5. [Section IV-B2 and Eq. (7)] The method description is internally inconsistent about the classifier input. Section IV-B2 says step 2 uses 'the original image x0 without any noise at a random time step t as input,' while the cycle loss in Eq. (7) is defined in terms of xt and the ground-truth and predicted class labels, and Table VII explicitly varies between x0 and xt. Please clarify exactly what input the classifier receives, what role t plays in step 2, and how Eq. (7) is computed when the classifier sees x0 rather than xt.
  6. [Abstract and Section I] The abstract and introduction make unqualified claims of 'state-of-the-art diagnostic accuracy and efficiency' and 'superior performance compared to existing baseline models.' These statements should be qualified to the datasets and metrics where the comparisons actually hold, especially in light of the COVID result in Table II and the missing diffusion-free baseline in Table IV.
minor comments (6)
  1. [Eq. (4)] The contrastive loss in Eq. (4) is written for a batch but does not explain how pairs (x1_t, x2_t) are sampled: are they two augmented views of the same image, two different images from the same class, or random pairs? Please specify the pairing rule.
  2. [Section IV-D1] The text says 'features from layers marked with yellow star, corresponding to p-values less than 0.05, were considered the best selections,' but the feature-selection section says that p-values greater than 0.05 indicate Gaussian maps. The logic is consistent only if the reader understands that non-Gaussian maps are semantically informative; please state this explicitly and use consistent notation for L_Diff vs 'Ldiff.
  3. [Section IV-C] The sentence 'D-Cube surpasses previous models in almost metrics across datasets' should read 'almost all metrics,' and the surrounding paragraph should be rewritten to distinguish the datasets on which D-Cube is best from the one where it is second-best.
  4. [Table III] The table is titled 'Augmentation with synthetic data of D-Cube,' but its rows are ResNeXt-101 and ViT-B/16 augmented with synthetic images. Please retitle the table to indicate that it evaluates the effect of synthetic-data augmentation on baseline models, not on D-Cube.
  5. [Table VIII] The F1 value for ResNeXt-101 sub-features is reported as 89.6, whereas the corresponding full D-Cube entry in Table II reports 89.69. Please correct the inconsistency and report precision and recall with consistent decimal places.
  6. [Fig. 2 and Table VI] The caption of Fig. 2 does not define the axes, the exact layers, or how the 'best' and 'worst' combinations were formed. Please label axes, indicate the number of layers selected, and describe in the caption how the yellow-star and red-cross markers map to the entries in Table VI.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the pipeline is empirically benchmarked against external baselines, and the feature-selection criterion is label-independent; remaining attribution concerns are correctness risks, not circular reductions.

full rationale

This paper is an empirical pipeline rather than a derivation, so there is no equation-level equivalence between inputs and claimed outputs. The Gaussianity-based layer selection (Eq. 6) is defined by KS test p-values computed from feature maps, not from classification labels or test accuracy; the Best/Worst comparison in Table VI is an empirical check of that criterion, not a fitted prediction. The diffusion model is trained on the same datasets that are later classified, but the reported accuracy is an external benchmark, not an input to the method, so this is a data-reuse/attribution concern rather than circularity. The only self-citation is the authors' own CPT baseline (Ref. [6], 'CPT (ours)' in Table IX), and it is not load-bearing: the central comparison is against external baselines such as DiffMIC, ResNet, and ViT. Table IV shows the largest gain comes from adding fsub, and no fsub-only run is reported; this is a genuine attribution/correctness gap, but it does not make any claimed result equivalent to its own input by construction. The COVID X-ray result (Table II) also does not support the 'surpasses state-of-the-art' wording, but that is a factual-claim issue, not circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The uncharged premises are empirical assumptions about diffusion feature semantics, the reliability of a single-batch KS test, and the usefulness of the cycle and consistency losses. The main free parameters are loss weights and the number of sub-feature channels, several of which are set without sensitivity analysis.

free parameters (5)
  • Contrastive loss margin = 0.1
    Margin in Eq. (4) is set without sensitivity analysis and affects the geometry of the diffusion features.
  • Cycle loss weight lambda1 = 10 for pancreas, 100 for breast, unspecified for COVID
    Per-dataset tuning of Eq. (10); no ablation or selection criterion is given.
  • Consistency loss weight lambda2 = 0.1
    Set empirically across all datasets; affects the contribution of Eq. (8).
  • Sub-feature channel count = 3 channels
    The authors state it is 'arbitrarily set to three channels', an architectural choice not derived from the data.
  • Selected diffusion layers = Dataset-dependent, from KS test
    Which layers are concatenated for fmain is chosen from KS p-values; the selection rule is a per-dataset choice rather than a fixed rule.
assumptions (4)
  • domain assumption Feature maps that deviate from a Gaussian distribution preserve semantic information needed for classification.
    Central to the Gaussianity feature selection; stated in Section III-B2a but not proven or independently validated.
  • domain assumption The KS test applied to a single batch of feature maps (n=256, t=100) reliably identifies layer semantics.
    The selection is based on average p-values of a single batch; no analysis of variance across batches or timesteps is provided.
  • domain assumption Cycle loss gradients through the frozen diffusion model usefully guide the classifier.
    Eq. (7) assumes that matching diffusion noise predictions for predicted and true labels improves classification, but no gradient analysis or convergence guarantee is offered.
  • domain assumption Horizontal flip is a label-preserving transformation for these medical classification tasks.
    Consistency regularization Eq. (8) assumes flip invariance, which may be violated for lateralized findings in chest X-ray or laterality-dependent MRI features.

how reviews work

0 comments
Cite this review

Pith. "Pith review of D-Cube: Exploiting Hyper-Features of Diffusion Model for Robust Medical Classification." pith.science (2026). https://pith.science/paper/YKXSFBUX

@misc{pith2026241111087,
  author       = {Pith},
  title        = {Pith review of: D-Cube: Exploiting Hyper-Features of Diffusion Model for Robust Medical Classification},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YKXSFBUX}},
  note         = {Machine review of arXiv:2411.11087}
}
read the original abstract

The integration of deep learning technologies in medical imaging aims to enhance the efficiency and accuracy of cancer diagnosis, particularly for pancreatic and breast cancers, which present significant diagnostic challenges due to their high mortality rates and complex imaging characteristics. This paper introduces Diffusion-Driven Diagnosis (D-Cube), a novel approach that leverages hyper-features from a diffusion model combined with contrastive learning to improve cancer diagnosis. D-Cube employs advanced feature selection techniques that utilize the robust representational capabilities of diffusion models, enhancing classification performance on medical datasets under challenging conditions such as data imbalance and limited sample availability. The feature selection process optimizes the extraction of clinically relevant features, significantly improving classification accuracy and demonstrating resilience in imbalanced and limited data scenarios. Experimental results validate the effectiveness of D-Cube across multiple medical imaging modalities, including CT, MRI, and X-ray, showing superior performance compared to existing baseline models. D-Cube represents a new strategy in cancer detection, employing advanced deep learning techniques to achieve state-of-the-art diagnostic accuracy and efficiency.

Figures

Figures reproduced from arXiv: 2411.11087 by the authors.

Figure 1
Figure 1. Overall Architecture: Step 1 involves the process of generating diffusion features that enhance the performance of D-Cube, while step 2 entails training a D-Cube for cancer diagnosis using the features generated from frozen diffusion model. xt represents an original image x0 with noise at random time step t. For more detailed information, refer to the method in Section III and the analysis in [PITH_FULL_IMAGE:figur… view at source ↗
Figure 2
Figure 2. Gaussianity Test of Each Datasets This graph presents the p-value analysis for pancreas cacner CT, breast cancer MRI, and COVID chest X-ray. Yellow star denote the best combination used, while red cross marks represent the worst combination. The p-value is calculated based on the average of features with a batch size of 256 when t=100. D. Ablation Study 1) Model Ablation: Our quantitative results confirm that featur… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 57 canonical work pages

  1. [1]

    U-net: Convolutional networks for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in Proc. MICCAI, 2015

    2015

  2. [2]

    Synthetic CT generation from CBCT images via deep learning,

    L. Chen, X. Liang, C. Shen, S. Jiang, and J. Wang, “Synthetic CT generation from CBCT images via deep learning,” Med. Phys., 2020

    work page 2020

  3. [3]

    Denoising diffusion probabilistic models,

    J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic models,” NeurIPS, 2020

    2020

  4. [4]

    Applications of artificial intelligence in pancreatic and biliary diseases,

    P.-T. Chen, D. Chang, T. Wu, M.-S. Wu, W. Wang, and W.-C. Liao, “Applications of artificial intelligence in pancreatic and biliary diseases,” J. Gastroenterol. Hepatol., 2021

    work page 2021

  5. [5]

    Pancreatic cancer detection on CT scans with deep learning: a nationwide population-based study,

    P.-T. Chen, T. Wu, P. Wang, D. Chang, K.-L. Liu, M.-S. Wu, H. R. Roth, P.-C. Lee, W.-C. Liao, and W. Wang, “Pancreatic cancer detection on CT scans with deep learning: a nationwide population-based study,” Radiology, 2023

    work page 2023

  6. [6]

    Unsupervised Visual Representation Learning Based on Segmentation of Geometric Pseudo- Shapes for Transformer-Based Medical Tasks,

    T. Viriyasaranon, S. M. Woo, and J.-H. Choi, “Unsupervised Visual Representation Learning Based on Segmentation of Geometric Pseudo- Shapes for Transformer-Based Medical Tasks,” IEEE J. Biomed. Health Inform., 2023

    work page 2023

  7. [7]

    "A robust and efficient AI assistant for breast tumor segmentation from DCE-MRI via a spatial

    Zhang, Jiadong, Zhiming Cui, Zhenwei Shi, Yingjia Jiang, Zhiliang Zhang, Xiaoting Dai, Zhenlu Yang, Yuning Gu, Lei Zhou, Chu Han, et al. "A robust and efficient AI assistant for breast tumor segmentation from DCE-MRI via a spatial

    work page

  8. [8]

    Synthetic data from diffusion models improves imagenet classification,

    S. Azizi, S. Kornblith, C. Saharia, M. Norouzi, and D. J. Fleet, “Synthetic data from diffusion models improves imagenet classification,” arXiv preprint arXiv:2304.08466, 2023

    arXiv 2023

Show all 64 references
  1. [9]

    All are worth words: A vit backbone for diffusion models,

    F. Bao, S. Nie, K. Xue, Y . Cao, C. Li, H. Su, and J. Zhu, “All are worth words: A vit backbone for diffusion models,” in Proc. CVPR, 2023

    2023

  2. [10]

    Zero-shot text-to-image generation,

    A. Ramesh, M. Pavlov, G. Goh, S. Gray, C. V oss, A. Radford, M. Chen, and I. Sutskever, “Zero-shot text-to-image generation,” in Proc. ICML, 2021

    2021

  3. [11]

    Stablevideo: Text-driven consistency-aware diffusion video editing,

    W. Chai, X. Guo, G. Wang, and Y . Lu, “Stablevideo: Text-driven consistency-aware diffusion video editing,” in Proc. ICCV , 2023

    2023

  4. [12]

    Deep residual learning for image recognition,

    K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. CVPR, 2016

    2016

  5. [13]

    Aggregated residual transformations for deep neural networks,

    S. Xie, R. Girshick, P. Dollár, Z. Tu, and K. He, “Aggregated residual transformations for deep neural networks,” in Proc. CVPR, 2017

    2017

  6. [14]

    ResNeSt: Split-Attention Networks

    H. Zhang, C. Wu, Z. Zhang, Y . Zhu, H. Lin, Z. Zhang, Y . Sun, T. He, J. Mueller, R. Manmatha, and others, “ResNeSt: Split-Attention Networks” in Proc. CVPR, 2022

    2022

  7. [15]

    Shufflenet v2: Practical guidelines for efficient cnn architecture design,

    N. Ma, X. Zhang, H.-T. Zheng, and J. Sun, “Shufflenet v2: Practical guidelines for efficient cnn architecture design,” in Proc. ECCV , 2018

    2018

  8. [16]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, and N. Houlsby, “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” in Proc. ICLR, 2021

    2021

  9. [17]

    Tokens-to-token vit: Training vision transformers from scratch on imagenet,

    L. Yuan, Y . Chen, T. Wang, W. Yu, Y . Shi, Z.-H. Jiang, F. E. Tay, J. Feng, and S. Yan, “Tokens-to-token vit: Training vision transformers from scratch on imagenet,” in Proc. ICCV , 2021

    2021

  10. [18]

    Cvt: Introducing convolutions to vision transformers,

    H. Wu, B. Xiao, N. Codella, M. Liu, X. Dai, L. Yuan, and L. Zhang, “Cvt: Introducing convolutions to vision transformers,” in Proc. ICCV , 2021

    2021

  11. [19]

    Rethinking spatial dimensions of vision transformers,

    B. Heo, S. Yun, D. Han, S. Chun, J. Choe, and S. J. Oh, “Rethinking spatial dimensions of vision transformers,” in Proc. ICCV , 2021

    2021

  12. [20]

    Pvt v2: Improved baselines with pyramid vision transformer,

    W. Wang, E. Xie, X. Li, D.-P. Fan, K. Song, D. Liang, T. Lu, P. Luo, and L. Shao, “Pvt v2: Improved baselines with pyramid vision transformer,” Computational Visual Media, 2022

    2022

  13. [21]

    Pyramid vision transformer: A versatile backbone for dense prediction without convolutions,

    W. Wang, E. Xie, X. Li, D.-P. Fan, K. Song, D. Liang, T. Lu, P. Luo, and L. Shao, “Pyramid vision transformer: A versatile backbone for dense prediction without convolutions,” in Proc. ICCV , 2021

    2021

  14. [22]

    On Aliased Resizing and Surprising Subtleties in GAN Evaluation,

    G. Parmar, R. Zhang, and J.-Y . Zhu, “On Aliased Resizing and Surprising Subtleties in GAN Evaluation,” in Proc. CVPR, 2022

    2022

  15. [23]

    SegFormer: Simple and efficient design for semantic segmentation with transformers,

    E. Xie, W. Wang, Z. Yu, A. Anandkumar, J. M. Alvarez, and P. Luo, “SegFormer: Simple and efficient design for semantic segmentation with transformers,” NeurIPS, 2021

    2021

  16. [24]

    An overview of deep learning in medical imaging focusing on MRI,

    A. S. Lundervold and A. Lundervold, “An overview of deep learning in medical imaging focusing on MRI,” Zeitschrift für Medizinische Physik, 2019

    2019

  17. [25]

    U-GAT-IT: Unsupervised Gen- erative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation,

    J. Kim, M. Kim, H. Kang, and K. H. Lee, “U-GAT-IT: Unsupervised Gen- erative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation,” in Proc. ICLR, 2019

    2019

  18. [26]

    Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI,

    N. M. Braman, M. Etesami, P. Prasanna, C. Dubchuk, H. Gilmore, P. Tiwari, D. Plecha, and A. Madabhushi, “Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI,” Breast Cancer...

    2017

  19. [27]

    Convolutional neural network of multiparametric MRI accurately detects axillary lymph node metastasis in breast cancer patients with pre neoadjuvant chemotherapy,

    T. Ren, S. Lin, P. Huang, and T. Q. Duong, “Convolutional neural network of multiparametric MRI accurately detects axillary lymph node metastasis in breast cancer patients with pre neoadjuvant chemotherapy,” Clin. Breast Cancer, 2022

    2022

  20. [28]

    Deep learning for identifying radiogenomic associations in breast cancer,

    Z. Zhu, E. Albadawy, A. Saha, J. Zhang, M. R. Harowicz, and M. A. Mazurowski, “Deep learning for identifying radiogenomic associations in breast cancer,” Comput. Biol. Med., 2019

    2019

  21. [29]

    I., Schönlieb, C

    Yang, Y ., Fu, H., Aviles-Rivero, A. I., Schönlieb, C. B., & Zhu, L. (2023, October). Diffmic: Dual-guidance diffusion network for medical image classification. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 95-105). Cham: Spring...

    2023

  22. [30]

    Can AI help in screening Viral and COVID-19 pneumonia?

    M.E.H. Chowdhury, T. Rahman, A. Khandakar, R. Mazhar, M.A. Kadir, Z.B. Mahbub, K.R. Islam, M.S. Khan, A. Iqbal, N. Al-Emadi, M.B.I. Reaz, M. T. Islam, “Can AI help in screening Viral and COVID-19 pneumonia?” IEEE Access, V ol. 8, 2020, pp. 132665 - 132676

    2020

  23. [31]

    and Chowdhury, M.E., 2020

    Rahman, T., Khandakar, A., Qiblawey, Y ., Tahir, A., Kiranyaz, S., Kashem, S.B.A., Islam, M.T., Maadeed, S.A., Zughaier, S.M., Khan, M.S. and Chowdhury, M.E., 2020. Exploring the Effect of Image Enhancement Techniques on COVID-19 Detection using Chest X-ray Images. arXiv prepr...

    2020 arXiv

  24. [32]

    and Mazurowski, M.A., 2018

    Saha, A., Harowicz, M.R., Grimm, L.J., Kim, C.E., Ghate, S.V ., Walsh, R. and Mazurowski, M.A., 2018. A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE- MRI features. British journal of cancer, 119(4), pp.508-516

    2018

  25. [33]

    C., Pareek, A., Jensen, M., Lungren, M

    Huang, S. C., Pareek, A., Jensen, M., Lungren, M. P., Yeung, S., & Chaudhari, A. S. (2023). Self-supervised learning for medical image classification: a systematic review and implementation guidelines. NPJ Digital Medicine, 6(1), 74

    2023

  26. [34]

    & Wang, Y

    Feng, Z., Wen, L., Xiao, J., Xu, Y ., Wu, X., Zhou, J., ... & Wang, Y . (2023). Diffusion-based Radiotherapy Dose Prediction Guided by Inter-slice Aware Structure Encoding. arXiv preprint arXiv:2311.02991

    2023 arXiv

  27. [35]

    Diffusion deep learning for brain age prediction and longitudinal tracking in children through adulthood

    Zapaishchykova, Anna, et al. "Diffusion deep learning for brain age prediction and longitudinal tracking in children through adulthood." Imaging Neuroscience 2 (2024): 1-14

    2024

  28. [36]

    Fu, L., Li, X., Cai, X., Wang, Y ., Wang, X., Yao, Y ., & Shen, Y . (2023). SP-DiffDose: A Conditional Diffusion Model for Radiation Dose Prediction Based on Multi-Scale Fusion of Anatomical Structures, Guided by SwinTransformer and Projector. arXiv preprint arXiv:2312.06187

    2023 arXiv

  29. [37]

    (2023, October)

    Lv, T., Liu, Y ., Miao, K., Li, L., & Pan, X. (2023, October). Diffusion Kinetic Model for Breast Cancer Segmentation in Incomplete DCE-MRI. MICCAI (pp. 100-109). Cham: Springer Nature Switzerland

    2023

  30. [38]

    Hardy, R., Klepich, J., Mitchell, R., Hall, S., Villareal, J., & Ilin, C. (2023). Improving nonalcoholic fatty liver disease classification performance with latent diffusion models. Scientific Reports, 13(1), 21619

    2023

  31. [39]

    Ling, Y ., Wang, Y ., Dai, W., Yu, J., Liang, P., & Kong, D. (2023). MTANet: Multi-Task Attention Network for Automatic Medical Image Segmentation and Classification. IEEE Transactions on Medical Imaging

    2023

  32. [40]

    A., Tselykh, A., Muthanna, M

    Ashurov, A., Chelloug, S. A., Tselykh, A., Muthanna, M. S. A., Muthanna, A., & Al-Gaashani, M. S. (2023). Improved breast Cancer classification through combining transfer learning and attention mechanism. Life, 13(9), 1945

    2023

  33. [41]

    Data variation-aware medical image segmen- tation

    Dushatskiy, Arkadiy, et al. "Data variation-aware medical image segmen- tation." Medical Imaging 2022: Image Processing. V ol. 12032. SPIE, 2022

    2022

  34. [42]

    Medsegdiff: Medical image segmentation with diffusion probabilistic model

    Wu, Junde, et al. "Medsegdiff: Medical image segmentation with diffusion probabilistic model." Medical Imaging with Deep Learning. PMLR, 2024

    2024

  35. [43]

    Diffusiondet: Diffusion model for object detection

    Chen, Shoufa, et al. "Diffusiondet: Diffusion model for object detection." Proceedings of the IEEE/CVF ICCV . 2023

    2023

  36. [44]

    M., Xie, J., Khan, F

    Wang, H., Cao, J., Anwer, R. M., Xie, J., Khan, F. S., & Pang, Y . (2023). Dformer: Diffusion-guided transformer for universal image segmentation. arXiv preprint arXiv:2306.03437

    2023 arXiv

  37. [45]

    Your diffusion model is secretly a zero-shot classifier

    Li, Alexander C., et al. "Your diffusion model is secretly a zero-shot classifier." Proceedings of the IEEE/CVF International Conference on Computer Vision. 2023

    2023

  38. [46]

    Diffusion hyperfeatures: Searching through time and space for semantic correspondence

    Luo, Grace, et al. "Diffusion hyperfeatures: Searching through time and space for semantic correspondence." Advances in Neural Information Processing Systems 36 (2024)

    2024

  39. [47]

    Label-Efficient Semantic Segmentation with Diffusion Models

    Baranchuk, D., Rubachev, I., V oynov, A., Khrulkov, V .,& Babenko, A. "Label-Efficient Semantic Segmentation with Diffusion Models." In ICCV, 2022. 9

    2022

  40. [48]

    Freedom: Training-free energy-guided conditional diffusion model

    Yu, Jiwen, et al. "Freedom: Training-free energy-guided conditional diffusion model." Proceedings of the IEEE/CVF ICCV . 2023

    2023

  41. [49]

    Learning transferable visual models from natural language supervision

    Radford, Alec, et al. "Learning transferable visual models from natural language supervision." ICML. PMLR, 2021

    2021

  42. [50]

    Caron, M., Touvron, H., Misra, I., Jégou, H., Mairal, J., Bojanowski, P., & Joulin, A. (2021). Emerging properties in self-supervised vision transformers. In Proceedings of the IEEE/CVF ICCV

    2021

  43. [51]

    J., Li, K., & Fei-Fei, L

    Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009, June). Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on CVPR (pp. 248-255). Ieee

    2009

  44. [52]

    How do vision transformers work?

    Park, Namuk, and Songkuk Kim. "How do vision transformers work?." arXiv preprint arXiv:2202.06709 (2022)

    2022 arXiv

  45. [53]

    AM- RADIO: Agglomerative Visual Foundation Model – Reduce All Domains Into One

    Ranzinger, Mike, Greg Heinrich, Jan Kautz, and Pavlo Molchanov. "AM- RADIO: Agglomerative Visual Foundation Model – Reduce All Domains Into One." In CVPR, 2024

    2024

  46. [54]

    Kolesnikov, A., Beyer, L., Zhai, X., Puigcerver, J., Yung, J., Gelly, S., & Houlsby, N. (2020). Big transfer (bit): General visual representation learning. In Computer Vision–ECCV 2020: 16th European Conference, 2020

    2020

  47. [55]

    Does robustness on imagenet transfer to downstream tasks?

    Yamada, Yutaro, and Mayu Otani. "Does robustness on imagenet transfer to downstream tasks?." Proceedings of CVPR. 2022

    2022

  48. [56]

    LoRA: Low-Rank Adaptation of Large Language Models

    Hu, Edward J., et al. "LoRA: Low-Rank Adaptation of Large Language Models." ICLR. 2022

    2022

  49. [57]

    Wang, Z., Wu, Z., Agarwal, D., & Sun, J. (2022). Medclip: Con- trastive learning from unpaired medical images and text. arXiv preprint arXiv:2210.10163

    2022 arXiv

  50. [58]

    K., Heidari, M., Azad, R., Fayyaz, M., Hacihaliloglu, I., & Merhof, D

    Kazerouni, A., Aghdam, E. K., Heidari, M., Azad, R., Fayyaz, M., Hacihaliloglu, I., & Merhof, D. (2023). Diffusion models in medical imaging: A comprehensive survey. Medical Image Analysis, 102846

    2023

  51. [59]

    T., Parekh, Z., Pham, H.,

    Jia, C., Yang, Y ., Xia, Y ., Chen, Y . T., Parekh, Z., Pham, H., ... & Duerig, T. (2021, July). Scaling up visual and vision-language representation learning with noisy text supervision. ICML. PMLR

    2021

  52. [60]

    Sora: Creating video from text

    OpenAI. “Sora: Creating video from text.” Accessed 2024. https://openai. com/sora

    2024

  53. [61]

    E., Setio, A

    Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., ... & Sánchez, C. I. (2017). A survey on deep learning in medical image analysis. Medical image analysis

    2017

  54. [62]

    C., Roth, H

    Shin, H. C., Roth, H. R., Gao, M., Lu, L., Xu, Z., Nogues, I., ... & Summers, R. M. (2016). Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE transactions on medical imaging, 35(5), 1285- 1298

    2016

  55. [63]

    Wongvorachan, T., He, S., & Bulut, O. (2023). A comparison of undersampling, oversampling, and SMOTE methods for dealing with imbalanced classification in educational data mining. Information, 14(1), 54

    2023

  56. [64]

    Qin, Y ., Zheng, H., Yao, J., Zhou, M., & Zhang, Y . (2023). Class- balancing diffusion models. CVPR. TABLE IX THE NETWORK CONFIGURATION FOR BENCHMARKING Method Learning Rate Weight Decay Epochs (#) Optimizer ResNet-101 1e-3 1e-4 50 SGD ResNeXt-101 1e-3 1e-4 50 SGD ShuffleNet ...

    2023

Pith tools

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