Pith. sign in

REVIEW 5 major objections 7 minor 32 references

DADU: Dual Attention-based Deep Supervised UNet for Automated Semantic Segmentation of Cardiac Images

T0 review · 5 major / 7 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read This paper claims that its DADU model, a U-Net with dual attention, edge-based skip connections, and deep supervision, segments cardiac MRI structures with 98% Dice and lower Hausdorff distances than leading methods.

desk verdict An incremental U-Net variant whose architecture is plausible but whose headline comparison to prior work is not supported by the experiments as reported. read the letter →

arxiv 2504.13415 v1 pith:7PSVY7HY submitted 2025-04-18 eess.IV cs.AIcs.CVcs.LG

Racheal Mukisa , Arvind K. Bansal This is my paper · ORCID
classification eess.IVcs.AIcs.CVcs.LG
keywords cardiacMRIsegmentationU-NetdualattentionchannelspatialdeepsupervisionDicesimilaritycoefficientHausdorffdistance
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 U-Net architecture augmented with DenseNet-style dense blocks, a dual attention block combining channel and spatial attention, edge-detection-based skip connections, and deep supervision can automatically segment the left ventricle, right ventricle, and myocardium from cardiac cine MRI more accurately than existing methods. It reports a mean Dice similarity score up to 0.98 and lower Hausdorff distances than the compared techniques, with the clearest gains on boundary-sensitive contour error. A sympathetic reader would care because reliable automated contours of these heart structures are the first step toward clinical measurements such as ventricular volumes and ejection fraction. The paper presents DADU as both more accurate and light enough to run quickly on modest hardware, which would matter for deployment in clinical settings.

What carries the argument

The load-bearing object is the Dual Attention Block (DAB), installed at each skip connection between encoder and decoder feature maps. DAB runs a Channel Attention Module first: it average-pools and max-pools each channel of the encoder map and the decoder map separately, adds the pooled vectors within each side, adds encoder and decoder summaries, and applies a sigmoid to produce channel weights. Those weights multiply the input feature map. Then a Spatial Attention Module average-pools, max-pools, and 1×1-convolves the refined map, concatenates the three spatial descriptors, passes them through a 7×7 convolution, and applies a sigmoid to produce a spatial weight map. The DAB output is the spatially reweighted feature map, and real extracted edges are also fused into the skip connection. Around this core, dense blocks in the encoder reuse features and auxiliary decoder paths contribute to the total loss, implementing deep supervision.

What would settle it

Re-run U-Net, UNet++, and Attention U-Net on the same 100-exam training and 50-exam test split, the same PNG/resizing preprocessing, and the same DSC and Hausdorff-distance computation used for DADU, and check whether DADU's Table 3 margins persist. If any baseline reaches or exceeds 0.98 DSC and matches or beats DADU's Hausdorff distances under identical conditions, the claimed superiority would not survive.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central discovery is that combining four mechanisms in one network—dense feature reuse in the encoder, channel and spatial attention at every skip connection, real edge information carried into the decoder, and deep supervision from auxiliary decoder paths—produces segmentation that the authors report as 0.98 Dice for the left ventricle at end-diastole, 0.97 for the right ventricle, and 0.93 for the myocardium, with Hausdorff distances lower than all compared models, especially at end-systole. The paper attributes the edge-quality improvement to the edge-based skip connections and attention-guided refinement, and the training stability to deep supervision. It concludes that DADU outperforms U-Net, FCN, CNN-based methods, MRF, UNet++, and Attention U-Net in both Dice score and Hausdorff distance.

Load-bearing premise

The load-bearing premise is that the comparison numbers in Table 3 for U-Net, FCN, CNN, MRF, UNet++, and Attention U-Net were obtained under the same training split, preprocessing, and evaluation protocol as DADU; the paper does not state this.

Editorial extensions

If this is right

  • At the reported accuracy, automated LV, RV, and myocardium contours from cine MRI reach near-expert overlap—around 0.98 Dice for the LV at end-diastole—with boundary errors low enough to support downstream volume and ejection-fraction estimates.
  • The edge-based skip connection recovers boundary detail lost in downsampling, so other U-Net-style segmentation models could adopt this mechanism without changing imaging hardware or acquisition protocols.
  • Deep supervision plus dense blocks allow the model to train in about 2.2 hours on 4 GB GPU memory and to segment a full phase in 0.2 seconds at test time, which makes clinical deployment plausible.
  • Because end-diastolic segmentation is consistently more accurate than end-systolic in the reported epochs, the main remaining risk to ejection-fraction accuracy is systolic contour quality, which the low systolic Hausdorff distances are meant to address.

Reading between the lines

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

  • Inference: Table 3's baseline numbers appear to be quoted from earlier papers rather than re-run under DADU's 100/50 split and preprocessing; until the comparison is repeated under identical conditions, the claimed margin over Attention U-Net remains unverified.
  • Inference: No ablation is reported, so the individual contributions of edge skip connections, dual attention, and deep supervision to the Hausdorff-distance gain are untested; an ablation study would isolate which mechanism carries the improvement.
  • Inference: Because DADU processes 2D slices independently, enforcing through-plane or cardiac-phase consistency could further sharpen end-systolic contours and is a natural next test.
  • Inference: The same recipe—encoder-decoder with channel-spatial attention and edge-aware skip connections—could transfer to other boundary-sensitive segmentation problems, such as echocardiographic chamber tracing or vessel-wall segmentation.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 7 minor

Summary. The paper proposes DADU, a U-Net variant combining dense blocks, channel and spatial attention (CBAM-style), edge-based skip connections, and deep supervision, for semantic segmentation of the left ventricle (LV), right ventricle (RV), and myocardium in the ACDC cardiac MRI dataset. The authors report DSC values up to 0.98 (LV, end-diastole) and Hausdorff distances lower than those of six compared methods in Table 3, and claim that DADU outperforms other leading techniques in both DSC and HD. The manuscript includes architecture diagrams, pseudo-code for the attention modules, implementation details, and a qualitative comparison of segmentation outputs.

Significance. If the reported results are reproducible and the comparisons in Table 3 are apples-to-apples, DADU would be a useful incremental contribution to cardiac MRI segmentation, particularly for the improved Hausdorff distance. The paper evaluates on the external ACDC benchmark and reports quantitative metrics for end-diastolic and end-systolic phases, which is a strength. However, the paper provides no code, trained models, or evaluation scripts, and the central comparative claim rests on baseline numbers whose provenance and protocol are not established. The claimed edge-based skip connection, one of the two stated contributions, is not actually described in the manuscript. As written, the significance of the reported results is therefore not yet established.

major comments (5)
  1. [Section 6, Table 3] The claim that DADU outperforms all other methods in DSC and HD depends entirely on Table 3, but the manuscript does not state whether the U-Net, FCN, CNN, MRF, UNET++, and AUN rows were recomputed under the same protocol or quoted from the cited papers. Section 5 states that fivefold cross-validation was used, while Section 5.1 describes a fixed 100-examination training / 50-examination testing split; the baseline rows are not tied to either regime. Moreover, the U-Net row cites [7], which is Avendi et al.'s right-ventricle CNN/autoencoder paper, and the AUN row cites [19], a pediatric echocardiography segmentation paper, rather than the Attention U-Net reference [17]. Without a controlled re-run under the same training, preprocessing, and evaluation protocol, the outperformance claim is unsupported.
  2. [Section 4.3, Eqs. (7)-(8)] Equations (7) and (8) are identical: both define M_{c in C}^{SpE} as f^{7x7}([F_{c in C}^{ChE-avg} • F_{c in C}^{ChE-1x1} • F_{c in C}^{ChE-max}]). One of these should presumably define the decoder map M_{c in C}^{SpD}. As written, the equations do not provide the decoder branch, and Eq. (9) then sums M_{c in C}^{SpE} with itself, so the spatial attention module is not well defined and cannot be implemented from the text.
  3. [Sections 4 and 4.1] The abstract and contributions list 'edge-detection based skip-connection' and 'real edge extraction' as central components, and Section 4 says skip connections include derived edge information. However, no equation, algorithm, or textual description specifies how edges are extracted (e.g., Sobel or Canny), how edge maps are fused with the DAB feature maps, or whether the edge branch is trained with a separate loss. This missing technical core makes the method unreproducible and prevents assessment of one of the two stated contributions.
  4. [Sections 5 and 5.1] The evaluation protocol is internally inconsistent: Section 5 states 'We employed a fivefold cross-validation approach,' while Section 5.1 states 'We divided the dataset into a training set comprising 100 examinations and a testing set comprising 50 examinations.' If the reported DSC/HD values are from the 50-examination test set, then the cross-validation sentence is misleading; if they are averaged over folds, the fixed split is misleading. Either way, Tables 1 and 3 cannot be interpreted without clarification of which regime produced the numbers.
  5. [Section 6, Tables 1-3] All reported metrics are single point estimates without standard deviations or confidence intervals. For a dataset with 150 examinations and two phases per patient, per-structure performance varies considerably across patients and slices; without variance or a paired significance test, the small HD improvements claimed in Table 3 (e.g., LV ED HD of 1.9 mm vs. 2.9 mm for AUN) cannot be distinguished from noise. The paper also does not specify whether HD is the maximum or 95th-percentile Hausdorff distance used in the ACDC challenge.
minor comments (7)
  1. [Section 5.3, Eq. (13)] The Hausdorff distance is defined as a max-min distance, but the ACDC challenge typically uses the 95th-percentile Hausdorff distance; please state explicitly which definition was used.
  2. [Abstract] The abstract refers to 'myocardium scar tissue,' but the ACDC dataset provides myocardium labels, not scar labels, and the paper does not segment scar tissue; please correct the terminology.
  3. [Tables 1 and 3] The LMyo end-diastolic HD value is 8.3 mm in Table 1 but 8.2 mm in Table 3 for DADU; please reconcile the rounding or identify which table is correct.
  4. [Algorithm 3] In the spatial attention loop, F_i^{SpD-max} = maxPool(F_i^{ChE}) uses the encoder feature for the decoder branch; this should presumably read F_i^{ChD}.
  5. [References] References [15] and [32] are the same CMS-UNet paper and should be consolidated; also several in-text citations in Section 3 and Table 3 do not match the cited works.
  6. [Figures 1 and 7-8] Figure 1 is low-resolution and the components described in the text (ED, DAB, deep supervision) are difficult to identify; Figures 7 and 8 would benefit from more detailed captions specifying what each column shows.
  7. [Table 2] The irregular behavior of HD around epoch 40 is mentioned but not explained; please provide an explanation or remove the anomaly from the discussion.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DADU is an empirical segmentation method evaluated against external ground truth; matching training loss and evaluation metric is standard, not a derivation loop.

full rationale

This paper makes no first-principles derivation whose conclusion is equivalent to its assumptions. The central claim is empirical: a U-Net variant with dual attention, edge-based skip connections, and deep supervision achieves certain DSC/HD values on the ACDC dataset. The reported DSC and HD are measured against manual ground-truth annotations on a held-out test set of 50 examinations, so the performance numbers are not constructed from the model's own outputs or definitions. The fact that Section 5.3 says "We used DSC to compute the loss during the experiments" and then reports DSC as the evaluation metric is objective alignment, not circularity: the loss is a per-sample training objective, and the reported metric is averaged on unseen test data against expert labels. There is no fitted parameter being relabeled as a prediction, and no self-citation chain is load-bearing. Reference [3] is a prior paper by the same authors but is cited only as general context on deep learning for cardiac imaging, not as evidence for DADU's superiority. The external citations [24], [29] support the CBAM and deep-supervision components and do not define the target result. The most serious weakness is comparison validity: Table 3's baseline rows appear to be quoted from heterogeneous sources without specifying whether U-Net, FCN, CNN, MRF, UNet++, and AUN were evaluated under the same split, preprocessing, and HD definition as DADU, and some citations are mismatched (e.g., U-Net row cites Avendi et al. [7], an RV CNN/autoencoder paper). That is a correctness and reproducibility concern, not a circularity concern, because the baseline values are not used to derive DADU's outputs. No circular step can be exhibited by reducing an equation or claim to its own input, so the appropriate finding is no circularity.

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

The central claim rests on the integrity of the ACDC ground truth, the chosen train/test split, and the untested assumption that the reported baseline numbers are comparable. The free parameters are standard training hyperparameters, mostly not tuned and some not reported (auxiliary loss weights). No new physical entities are introduced. The main unstated premise is that the baseline methods in Table 3 were evaluated under the same protocol as DADU.

free parameters (4)
  • learning rate = 0.001
    ADAM optimizer learning rate, set by hand, not tuned or justified.
  • batch size = 10
    Set by hand for minimum batch gradient descent.
  • training epochs = 100
    Training stops at 100 epochs with best model retained; no early stopping criterion described.
  • auxiliary loss weights eta_s = not reported
    Equation (11) includes balancing weights eta_s for auxiliary paths, but their values or update rule are never given.
assumptions (4)
  • domain assumption ACDC manual annotations are accurate ground truth
    Section 5.1 uses expert annotations for LV, RV, LMyo as ground truth; any annotation error directly affects DSC and HD values reported in Tables 1-3.
  • domain assumption Converting 3D volumes to 2D slices and segmenting each slice independently preserves clinically relevant anatomy
    Section 5.2 describes converting NIFTI volumes to 256x256 2D slices with no explicit handling of through-plane continuity; the model never sees the full 3D context.
  • domain assumption Baseline results in Table 3 are comparable to DADU results
    Section 6 reports competitor DSC/HD values without stating whether they were recomputed on the same 100/50 split and preprocessing pipeline; the claim that DADU outperforms them depends on this assumption.
  • domain assumption Dense blocks and attention modules reduce information loss and vanishing gradients as intended
    Section 4 asserts these benefits, but no ablation study isolates the contribution of each component; the architecture relies on this premise for its reported performance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DADU: Dual Attention-based Deep Supervised UNet for Automated Semantic Segmentation of Cardiac Images." pith.science (2026). https://pith.science/paper/7PSVY7HY

@misc{pith2026250413415,
  author       = {Pith},
  title        = {Pith review of: DADU: Dual Attention-based Deep Supervised UNet for Automated Semantic Segmentation of Cardiac Images},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7PSVY7HY}},
  note         = {Machine review of arXiv:2504.13415}
}
read the original abstract

We propose an enhanced deep learning-based model for image segmentation of the left and right ventricles and myocardium scar tissue from cardiac magnetic resonance (CMR) images. The proposed technique integrates UNet, channel and spatial attention, edge-detection based skip-connection and deep supervised learning to improve the accuracy of the CMR image-segmentation. Images are processed using multiple channels to generate multiple feature-maps. We built a dual attention-based model to integrate channel and spatial attention. The use of extracted edges in skip connection improves the reconstructed images from feature-maps. The use of deep supervision reduces vanishing gradient problems inherent in classification based on deep neural networks. The algorithms for dual attention-based model, corresponding implementation and performance results are described. The performance results show that this approach has attained high accuracy: 98% Dice Similarity Score (DSC) and significantly lower Hausdorff Distance (HD). The performance results outperform other leading techniques both in DSC and HD.

Figures

Figures reproduced from arXiv: 2504.13415 by the authors.

Figure 6
Figure 6. Sample segmentation output from the DADU model. Column a) and c) show the original images, Column b) and d) show the segmented image. All rows from top-to-bottom show the sequential arrangement of images from apex to the base [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 30 canonical work pages

  1. [7]

    R., Kheradvar, A., Jafarkhani, H.: Automatic Segmentation of the Right Ventricle from Cardiac MRI using a Learning‐based Approach

    Avendi, M. R., Kheradvar, A., Jafarkhani, H.: Automatic Segmentation of the Right Ventricle from Cardiac MRI using a Learning‐based Approach. Magnetic Resonance in Medicine, 78(6): 2439-2448 (2017)

    work page 2017

  2. [19]

    F., Qin, J., et el.: Dual Attention Enhancement Feature Fusion Network for Segmentation and Quantitative Analysis of Pediatric Echocardiography

    Gun, L., Lei, B., Chen, W., Du, J., Frangi, A. F., Qin, J., et el.: Dual Attention Enhancement Feature Fusion Network for Segmentation and Quantitative Analysis of Pediatric Echocardiography. Medical Image Analysis, 71, 102042 (2021)

    work page 2021

  3. [17]

    L., Lee, M., Heinrich, M., Misawa, K., et el.: Attention U-net: Learning Where to Look for the Pancreas

    Oktay, O., Schlemper, J., Folgoc, L. L., Lee, M., Heinrich, M., Misawa, K., et el.: Attention U-net: Learning Where to Look for the Pancreas. arXiv preprint arXiv:1804.03999 (2018)

    arXiv 2018

  4. [1]

    World Health Organization Facts Sheet, https://www.who.int/news-room/fact-sheets/detail/ cardiovascular-diseases-(cvds), last accessed 2024/04/25

    work page 2024

  5. [2]

    H., Razavi, R., Taylor, A

    Anderson, R. H., Razavi, R., Taylor, A. M.: Cardiac anatomy revisited. Journal of Anatomy, 205(3), 159−177 (2004)

    work page 2004

  6. [3]

    K., Mukisa, R.: Applying deep learning techniques in automated analysis of echocardiograms, CMRs and phonocardiograms for the detection and localization of cardiac diseases

    Bansal, A. K., Mukisa, R.: Applying deep learning techniques in automated analysis of echocardiograms, CMRs and phonocardiograms for the detection and localization of cardiac diseases. In: Proceedings of the Eleventh International Conference on Artificial Intelligence and Applications (INTELLI), pp. 36−42. IARIA Press, Venice, Italy (2022)

    work page 2022

  7. [4]

    C.: Cardiac MRI evaluation of myocardial disease

    Captur, G., Manisty, C., Moon, J. C.: Cardiac MRI evaluation of myocardial disease. Heart, 102 (18), 1429−1435 (2016)

    work page 2016

  8. [5]

    Circulation

    La Gerche, A., Claessen, G., Van de Bruaene, A., Pattyn, N., Van Cleemput, J., Gewillig, M., et el.: Cardiac MRI: A New Gold Standard for Ventricular Volume Quantification during High-intensity Exercise. Circulation. Cardiovascular Imaging, 6(2), 329-338 (2012)

    work page 2012

Show all 32 references
  1. [6]

    Radiographics, 32(5), 1381−1398 (2012)

    Jeung, M -Y., Germain, P., Croisille, P., El ghannudi, S., Roy, C., Gangi, A.: Myocardial tagging with MR imaging: overview of normal and pathologic findings. Radiographics, 32(5), 1381−1398 (2012)

    2012

  2. [8]

    Frontiers in Cardiovascular Medicine, 7, Article 25 (2020)

    Chen, C., Qin, C., Qiu, H., Tarroni, G., Duan, J., Bai, W., et al.: Deep Learning for Cardiac Image Segmentation: A Review. Frontiers in Cardiovascular Medicine, 7, Article 25 (2020)

    2020

  3. [9]

    In: Nawab, N., Hornegger, J., Wells, W

    Ronneberger, O., Fischer, P., Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Nawab, N., Hornegger, J., Wells, W. M., Frangi, A. F. (eds.) 18th International Conference on Medical Image Computing and Computer -Assisted Intervention (MICCAI) 2015,...

    2015

  4. [10]

    Zhou, Y., Shi, W.-R., Chen, W., Chen, Y.-l., Li, Y., Tan, L.-W., et al.: Active contours driven by localizing region and edge -based intensity fitting energy with application to the segmentation of the left ventricles in CT images, Neurocomputing, 156, 199−210 (2015)

    2015

  5. [11]

    A., Krishnamurthi, G.: Fully Convolutional Multi-Scale Residual DenseNets for Cardiac Segmentation and Automated Cardiac Diagnosis using Ensemble of Classifiers

    Khened, M., Kollerathu, V. A., Krishnamurthi, G.: Fully Convolutional Multi-Scale Residual DenseNets for Cardiac Segmentation and Automated Cardiac Diagnosis using Ensemble of Classifiers. Medical Image Analysis, 51, 21-45 (2019)

    2019

  6. [12]

    -M.: Cardiac Segmentation with Strong Anatomical Guarantees

    Painchaud, N., Skandarani, Y., Judge, T., Bernard, O., Lalande, A., Jodoin, P. -M.: Cardiac Segmentation with Strong Anatomical Guarantees. IEEE Transactions on Medical Imaging, 39(11), 3703−3713 (2020)

    2020

  7. [13]

    Zotti, C., Luo, Z., Lalande, A., Jodoin, P. -M. : Convolutional Neural Network with Shape Prior applied to Cardiac MRI Segmentation. IEEE Journal of Biomedical and Health Informatics, 23(3), 1119−1128 (2018)

    2018

  8. [14]

    In: Pop, M., Sermesant M., Jodoin, P.-M., Lalande, A., et al

    Grinias, E., Tziritas, G.: Fast fully automatic cardiac segmentation in MRI using MRF model optimization, substructures tracking and B -spline smoothing. In: Pop, M., Sermesant M., Jodoin, P.-M., Lalande, A., et al. (eds.) 8th International Workshop on Statistical Atlases and ...

    2018

  9. [16]

    Journal of Advances in Applied and Computational Mathematics, 9, 76-85 (2022)

    Zhao, C., Bober, R., Tang, H., Tang, J., Dong, M., Chaoyang, Z., et el.: Semantic Segmentation to Extract Coronary Arteries in Invasive Coronary Angiograms . Journal of Advances in Applied and Computational Mathematics, 9, 76-85 (2022)

    2022

  10. [18]

    -M.-R., Tajbakhsh, N., Liang, J.: UNET++ : Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

    Zhou, Z., Siddique, M. -M.-R., Tajbakhsh, N., Liang, J.: UNET++ : Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation. IEEE Transactions on Medical Imaging, 39(6), 1856-1867 (2020)

    2020

  11. [20]

    K., Horowitz, T

    Evans, K. K., Horowitz, T. S., Howe, P., Pedersini, R., Reijnen, E., Pinto, Y., et el.: Visual Attention. Wiley Interdisciplinary Reviews: Cognitive Science, 2(5), 503-514 (2011)

    2011

  12. [21]

    IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(1), 185-207 (2013)

    Borji, A., Itti, L.: State of the Art in Visual Attention Modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(1), 185-207 (2013)

    2013

  13. [22]

    V., Csiszer, E

    Jones, A., Ward, E. V., Csiszer, E. L., & Szymczak, J.: Temporal Expectation Improves Recognition Memory for Spatially Attended Objects. Journal of Cognitive Neuroscience, 34(9), 1616-1629 (2022)

    2022

  14. [23]

    N., et al.: Attention is all you need

    Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Aidan, N., Gomez, A. N., et al.: Attention is all you need. In: Guyon, I., Von Luxburg, U., Bengio, S., Wallach, H., et al. (eds.) 31st International Conference on Advances in Neural Information Processing Systems...

    2017

  15. [24]

    S.: CBAM: convolutional block attention module

    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), 2018, pp. 3-19, open access, available at https://openaccess.thecvf.com/content_ECCV_2018/papers/ Sanghyun_Woo_Convoluti...

    2018

  16. [25]

    7132-7141

    Hu, J., Shen, L., Sun, G.: Squeeze-and-Excitation Networks, In: IEEE/CVF Conference on Computer Vision and Pattern Recognition , IEEE Conference Publishing Services, Los Alamitos (2018), pp. 7132-7141

    2018

  17. [26]

    In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp

    Li, X., Wang, W., Hu, X., & Yang, J.: Selective Kernel Networks. In: IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 510 -519, open access, available at https://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Selective_Kernel_Networks_ CVPR_2019_paper.pdf

    2019

  18. [27]

    Mathematics, 10(10), 1750 (2022)

    Liu, T., Luo, R., Xu, L., Feng, D., Cao, L., Liu, S., et el.: Spatial Channel Attention for Deep Convolutional Neural Networks. Mathematics, 10(10), 1750 (2022)

    2022

  19. [28]

    S.: Bam: Bottleneck Attention Module." arXiv preprint arXiv:1807.06514 (2018)

    Park, J., Woo, S., Lee, J.-Y., Kweon, I. S.: Bam: Bottleneck Attention Module." arXiv preprint arXiv:1807.06514 (2018)

    2018 arXiv

  20. [29]

    Y., Xie, S., Gallagher, P., Zhang, Z., & Tu, Z.: Deeply Supervised Nets

    Lee, C. Y., Xie, S., Gallagher, P., Zhang, Z., & Tu, Z.: Deeply Supervised Nets. Journal of Machine Learning Research, 38, 562-570 (2015)

    2015

  21. [30]

    Bernard, O., Lalande, A., Zotti, C., Cervenansky, F., et al. Deep Learning Techniques for Automatic MRI Cardiac Multi -structures Segmentation and Diagnosis: Is the Problem Solved ? IEEE Transactions on Medical Imaging, 37(11), 2514-2525 (2018)

    2018

  22. [31]

    Automatic Cardiac Dataset Challenge (ACDC), available at https://www.creatis.insa- lyon.fr/Challenge/acdc/databases.html

  23. [32]

    In: Zhuang, X., Li, L

    Li, W., Wang, L., Qin, S.: CMS -UNet: Cardiac Multi -task Segmentation in MRI with a U - Shaped Network. In: Zhuang, X., Li, L. (eds) Myocardial Pathology Segmentation Combining Multi-Sequence Cardiac Magnetic Resonance Images. MyoPS 2020. LNCS, vol 12554. Springer, Cham. (2020)

    2020

  24. [33]

    Nature Communications, 12(1):715, doi: 10.1038/s41467-021-20966-2

    Zeleznik, R., Foldyna, B., Parastou Eslami, P., Jakob Weiss, J., Alexander, I., Jana Taron, J., et el.: Deep Convolutional Neural Networks to Predict Cardiovascular Risk from Computed Tomography. Nature Communications, 12(1):715, doi: 10.1038/s41467-021-20966-2

    doi

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.