Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation

Balamurali Murugesan; Jayaraj Joseph; Kaushik Sarveswaran; Keerthi Ram; Mohanasankar Sivaprakasam; Sharath M Shankaranarayana

REVIEW 5 major objections 5 minor 12 references

Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation

T0 review · 5 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A plug-in module that learns mask, contour, and distance together improves medical segmentation across four base networks.

desk verdict A plausible, low-cost multi-task module with honest ablations, but the headline 'any architecture' claim rests on a single split and post-hoc distance-map selection. read the letter →

arxiv 1908.05311 v1 pith:3RCV6DTR submitted 2019-08-14 cs.CV

Balamurali Murugesan , Kaushik Sarveswaran , Sharath M Shankaranarayana , Keerthi Ram , Jayaraj Joseph , Mohanasankar Sivaprakasam This is my paper · ORCID

classification cs.CV

keywords Multi-tasklearningMedicalimagesegmentationContourextractionDistancemapPlug-and-playmoduleConvolutionalneuralnetworkDicescorePolyp

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 claims that a small plug-and-play module, Conv-MCD, improves any medical image segmentation network by making it learn three related tasks at once: mask prediction, contour extraction, and distance-map estimation. The module attaches to the end of an existing encoder-decoder network and adds only a few thousand parameters, yet the authors report higher Dice and Jaccard scores and lower Hausdorff distances on polyp segmentation for UNet, SegNet, LinkNet, and UNet16. The point of the claim is that structural information already present in ground-truth masks, namely contours and distance transforms, can regularize segmentation at no extra annotation cost, replacing heavier dual-decoder multi-task designs. If the claim holds, it gives a cheap, architecture-agnostic way to tighten boundaries, suppress outliers, and reduce overfitting in medical segmentation.

What carries the argument

The load-bearing object is the Conv-MCD module itself: a set of three parallel 3x3 convolutional filters placed at the end of any segmentation network, with two filters for the mask and contour classification outputs and one for the distance-map regression output. The contour target comes from dilating ground-truth boundaries by a disk of radius 5, and the distance targets come from Euclidean or signed distance transforms of the mask or contour; the paper tests three variants and reports the best. The total loss is $\lambda_1 L_{mask} + \lambda_2 L_{contour} + \lambda_3 L_{distance}$ with negative log-likelihood for the two classification tasks and mean squared error for the regression task. This machinery forces the shared encoder and decoder to represent structural sub-features that a vanilla cross-entropy-trained network does not, which is what the ablations and feature-map visualizations are meant to show.

What would settle it

Run the module on several independent medical segmentation datasets with a fixed, pre-registered distance-map type and multiple random splits; if the Dice gain over each base network is not consistently positive or is within split-to-split noise, the generic plug-and-play claim is falsified.

Watch

Extended reading notes

Core claim

The central claim is that the Conv-MCD module, three parallel convolutional filters that output a mask, a contour map, and a distance map from the same feature representation, yields better segmentation than the base network alone and than existing dual-decoder multi-task networks, while keeping parameter count and inference time close to the plain base network. On the GIANA polyp dataset, adding Conv-MCD to UNet16 raises Dice from 0.8441 to 0.9124 and Jaccard from 0.7676 to 0.8559, and similar improvements appear for SegNet, UNet, and LinkNet34. The paper attributes the gain to joint learning of a classification task for the mask, another classification task for the contour, and a regression task for the distance map, which imposes shape and smoothness priors, handles foreground-background class imbalance, and removes small outliers that a distance-only dual-decoder design tends to discard in multi-instance cases.

Load-bearing premise

The claim of universal improvement rests on results from one randomly split polyp dataset, evaluated without cross-validation or significance tests, with the best of three distance-map variants chosen after seeing the test numbers.

Editorial extensions

If this is right

Adding Conv-MCD to a segmentation network should consistently improve Dice, Jaccard, Hausdorff distance, and boundary F-score over the same network without the module, as observed for all four tested architectures.
Because the auxiliary targets are derived from ground-truth masks, the improvement comes at no additional annotation cost and with only a few thousand extra parameters.
The module should reduce outliers and improve boundary smoothness, including in multi-instance cases where a distance-map-only dual-decoder network removes small objects.
Learning mask, contour, and distance in parallel acts as a regularizer, so validation loss should plateau higher and overfitting should appear later than in base-network training.

Reading between the lines

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

Beyond what the paper tests, the 'any architecture' claim would be stronger with evidence on more modalities; the benefit may be largest for blob-like objects with fuzzy boundaries similar to polyps.
The reported sensitivity to distance-map type suggests that distance transform choice could be tuned per organ or imaging modality, a cheap validation-set hyperparameter worth testing.
A natural testable extension is semi-supervised use, where predicted masks on unlabeled images generate contour and distance targets for consistency training.

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.

Desk Editor's Note

read the letter

The new thing here is the module design: three parallel 3x3 convolutional heads on top of a single decoder, producing mask, contour, and distance map simultaneously. That is a legitimately cheap alternative to the two-decoder designs in DCAN and DMTN, and the paper shows it on four architectures with consistent gains on the GIANA polyp dataset. The ablation comparing Conv-MC, Conv-MD, and full Conv-MCD is the right kind of experiment, and the feature-map visualization supports the single-decoder argument qualitatively. Credit is due for the clean experimental setup across architectures and for releasing code.

That said, the central claim—that the module improves "any state-of-the-art base network"—is under-supported. Table 2 is a single random 70/30 split with no error bars, no significance tests, and no cross-validation. Worse, Section 3.4 states the reported results come from "the best performing distance map," meaning the headline numbers are selected maxima over three design choices, chosen after seeing results. That is a real correctness risk, not a nitpick. The ORIGA generalization check is relegated to a footnote and supplementary material, so the main text never actually demonstrates transfer to a second dataset. Loss weights lambda_1 through lambda_3 are also left unspecified, which makes reproducing the exact numbers harder.

The within-dataset comparison against DCAN and DMTN is useful, and the UNet16 gain is striking, but I would not put much trust in the magnitude until I see seeds and multiple datasets. The module is plausible and the overhead is tiny, so it may well be a practical addition—but the evidence as presented cannot separate real gains from split luck and selection effects.

If I were editing, I would send this to peer review rather than desk-reject it. The idea is sound, the experiments are mostly well designed, and the flaws are fixable with more rigorous evaluation. The authors need to add multi-seed runs, report standard deviations, pre-specify or at least cross-validate the distance-map choice, and put the ORIGA results in the main text. The paper is for people working on medical image segmentation who might want a cheap multi-task boost; readers outside that niche will not gain much. I would not cite it in my own work yet, but I would point to it as an example of a plausible module that needs stronger validation.

Referee Report

5 major / 5 minor

Summary. The paper proposes Conv-MCD, a plug-and-play module that appends three parallel convolutional heads to a segmentation network, predicting the segmentation mask, a dilated contour map, and a distance map. The total loss combines NLL losses for mask and contour with MSE loss for the distance map. The authors claim that the module can be added to any state-of-the-art segmentation network with minimal parameter overhead and improves Dice, Jaccard, Hausdorff distance, and boundary F-score. Experiments on the GIANA polyp segmentation dataset compare four base networks (SegNet, UNet, UNet16, LinkNet34) with and without the module, and ablations compare the full module against single-auxiliary-task variants (Conv-MC, Conv-MD).

Significance. If the empirical claims hold, the module is a practically useful contribution: it provides a simple way to incorporate multi-task structural supervision without duplicating decoders, requires only a few thousand extra parameters, and includes code and ablations. The idea of using contour and distance maps as auxiliary tasks is not new, but the module design that decouples these tasks from the base architecture is convenient. However, the evidence for the central 'any architecture' claim is currently weak: the headline results come from a single random split with no error bars or significance tests, and the best distance-map variant was selected after seeing test results. These issues need to be addressed before the broad generalization claim is justified.

major comments (5)

[Section 3.4, Table 2] The central claim that adding Conv-MCD improves any state-of-the-art network is supported only by results on a single random 70/30 split of the GIANA dataset (912 images), with no cross-validation, no multiple random seeds, and no significance tests. The reported gains, such as UNet16 Dice rising from 0.8441 to 0.9124, could be partly due to split luck. Please add results over multiple seeds or cross-validation folds with error bars and, where appropriate, paired significance tests.
[Section 3.4, sentence beginning 'The results reported in this paper...'] The paper states that the reported results are from the best performing distance map among D1, D2, and D3. Because this choice was made after seeing test-set performance, the Conv-MCD rows in Tables 1 and 2 are selected maxima, which inflates the apparent improvement over the base networks and makes the comparison unfair. Please either pre-specify the distance map, select it on a validation split, or report results for all three distance-map variants in the main text.
[Equations (2) and (3)] The NLL losses are written as positive log-likelihoods, but the total loss in Eq. (1) is minimized by gradient descent. As written, minimizing L_total would maximize the likelihood terms, which is inconsistent with standard NLL loss. Please correct the sign (introduce a negative sign) or clarify that the losses are to be maximized, and ensure the accompanying text uses the corrected formulation.
[Footnote 4] The claim that the approach generalises beyond GIANA is supported only by a pointer to supplementary material; the main text does not contain any results on the ORIGA dataset. Since the central claim is 'any state-of-the-art base network' and generalisability across datasets, the ORIGA results should either be moved into the main text or the claim should be tempered to reflect that only single-dataset evidence is presented.
[Table 1] The ablation study gives mixed evidence for the contribution of the individual auxiliary tasks: Conv-MC yields a small Dice improvement over the base (0.8149 vs 0.8125) but decreases MF (0.6083 vs 0.6144), while Conv-MD improves Dice but lowers MF compared to the base. This does not clearly support the statement that each auxiliary task individually helps, and it complicates the attribution of the full module's gains. Please discuss this and provide repeated-run statistics or at least error bars.

minor comments (5)

[Section 2.3, Eq. (1)] The values of the loss scaling factors λ1, λ2, λ3 are not reported; please state them explicitly for reproducibility.
[Section 2.2] The contour dilation disk radius of 5 is said to be empirically optimal for 256x256 images, but no supporting experiment is shown; please clarify whether this was tuned on the training split or chosen post hoc.
[Section 2.3, Eq. (4)] The distance map D3 is a signed distance transform and can take negative values, but the predicted distance map is passed through a sigmoid activation, which outputs values in [0,1]. Please clarify how negative distances are represented, e.g., whether the signed map is normalized or the sigmoid is used only for unsigned variants.
[Section 3.4, Figures 2a and 3a] The notation '1Enc 1Dec M', '1Enc 2Dec MC', and '1Enc 2Dec MD' is used in the text and Table 1 but is not defined before first use; please define these terms in the table caption or in a short notation paragraph.
[Abstract and Section 3.4] The abstract says 'significant increase in performance' but the paper reports no statistical significance tests; please replace 'significant' with 'consistent' or add actual significance testing.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: Conv-MCD's claims are empirical comparisons, not derivations that reduce to their inputs; best-distance-map selection is model selection, not logical circularity.

full rationale

The paper proposes a multi-task module and validates it empirically, so there is no derivation chain whose output is equivalent to its input by construction. The loss function combines mask NLL, contour NLL, and distance-map MSE, with the auxiliary targets computed directly from ground-truth masks; this is a standard training setup rather than a hidden reuse of the evaluation quantity. The reported Dice/Jaccard/Hausdorff improvements are test-set measurements against baseline networks, so they are not forced by construction. The statement in Section 3.4 that 'The results reported in this paper are from the best performing distance map' indicates selection over auxiliary-task designs on the same benchmark; this raises concerns about generalization and significance, but it is a statistical reporting issue, not logical circularity. There are no load-bearing self-citations (the only multi-task learning citation is Caruana 1993), no imported uniqueness theorem, and no ansatz smuggled in through citation. The paper is self-contained as an empirical study, and its central claim, while possibly under-supported by a single dataset and split, is not circular.

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

The central claim rests on three auxiliary supervision sources computed from ground-truth masks (mask, contour, distance), on unspecified loss weights, on the hand-tuned dilation radius, and on the assumption that one polyp dataset split once is enough to support a general claim. No new entities are invented.

free parameters (3)

Loss scaling factors lambda_1, lambda_2, lambda_3 = Not stated in main text
Equation (1) combines three losses with scaling factors; no values or tuning procedure are given, so the reported results depend on unspecified weights.
Contour dilation disk radius = 5 pixels
Section 2.2 says radius 5 was "empirically found... optimal" for 256x256 images, a hand-tuned preprocessing choice.
Distance map variant D1/D2/D3 = Best-performing variant used
Section 3.4 says results are from the best performing distance map, chosen after experiments; this selection can inflate apparent gains.

assumptions (3)

domain assumption Multi-task learning with auxiliary contour and distance objectives improves the primary segmentation task.
The paper relies on this as its core inductive bias (Section 2), citing prior work rather than deriving it.
domain assumption The GIANA polyp dataset, split once 70/30, is representative enough to support a general claim about medical image segmentation.
Used without cross-validation or external validation in the main results; ORIGA is only in supplementary.
domain assumption Standard neural network training assumptions (e.g., Adam, fixed learning rate, 150 epochs) are adequate for fair comparison.
Implementation details in Section 3.2 are fixed, but no sensitivity analysis is reported.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation." pith.science (2026). https://pith.science/paper/3RCV6DTR

@misc{pith2026190805311,
  author       = {Pith},
  title        = {Pith review of: Conv-MCD: A Plug-and-Play Multi-task Module for Medical Image Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3RCV6DTR}},
  note         = {Machine review of arXiv:1908.05311}
}

read the original abstract

For the task of medical image segmentation, fully convolutional network (FCN) based architectures have been extensively used with various modifications. A rising trend in these architectures is to employ joint-learning of the target region with an auxiliary task, a method commonly known as multi-task learning. These approaches help impose smoothness and shape priors, which vanilla FCN approaches do not necessarily incorporate. In this paper, we propose a novel plug-and-play module, which we term as Conv-MCD, which exploits structural information in two ways - i) using the contour map and ii) using the distance map, both of which can be obtained from ground truth segmentation maps with no additional annotation costs. The key benefit of our module is the ease of its addition to any state-of-the-art architecture, resulting in a significant improvement in performance with a minimal increase in parameters. To substantiate the above claim, we conduct extensive experiments using 4 state-of-the-art architectures across various evaluation metrics, and report a significant increase in performance in relation to the base networks. In addition to the aforementioned experiments, we also perform ablative studies and visualization of feature maps to further elucidate our approach.

Figures

Figures reproduced from arXiv: 1908.05311 by the authors.

**Figure 2.** Left (a): Four sample cases, from left to right: Image, Ground truth, 1Enc [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗

**Figure 3.** Left (a): From row 1 to 4: SegNet, UNet, UNet16 and LinkNet34. In each [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 10 canonical work pages

[1]

IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481--2495 (2016)

Badrinarayanan, V., Kendall, A., Cipolla, R.: Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481--2495 (2016)

work page 2016
[2]

In: ICML (1993)

Caruana, R.: Multitask learning: A knowledge-based source of inductive bias. In: ICML (1993)

work page 1993
[3]

2017 IEEE Visual Communications and Image Processing (VCIP) pp

Chaurasia, A., Culurciello, E.: Linknet: Exploiting encoder representations for efficient semantic segmentation. 2017 IEEE Visual Communications and Image Processing (VCIP) pp. 1--4 (2017)

work page 2017
[4]

In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Chen, H., Qi, X., Yu, L., Heng, P.: DCAN: Deep Contour-Aware Networks for Accurate Gland Segmentation . In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 2487--2496 (2016)

work page 2016
[5]

Csurka, G., Larlus, D.: What is a good evaluation measure for semantic segmentation? In: Proceedings of the British Machine Vision Conference (2013)

work page 2013
[6]

a henb\

Kr\" a henb\" u hl, P., Koltun, V.: Efficient inference in fully connected crfs with gaussian edge potentials. In: Advances in Neural Information Processing Systems 24, pp. 109--117 (2011)

work page 2011
[7]

In: Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015

Ronneberger, O., Fischer, P., Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation . In: Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015. pp. 234--241 (2015)

work page 2015
[8]

In: 3rd International Conference on Learning Representations ICLR (2015)

Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: 3rd International Conference on Learning Representations ICLR (2015)

work page 2015

Show all 12 references

[9]

In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)

Tan, C., Zhao, L., Yan, Z., et al.: Deep multi-task and task-specific feature learning network for robust shape preserved organ segmentation . In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). pp. 1221--1224 (2018)

2018
[10]

Journal of healthcare engineering 2017 (2017)

V \'a zquez, D., Bernal, J., S \'a nchez, F.J., et al.: A benchmark for endoluminal scene segmentation of colonoscopy images. Journal of healthcare engineering 2017 (2017)

2017
[11]

, " * write output.state after.block = add.period write

ENTRY address author booktitle chapter doi edition editor eid howpublished institution journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION in...
[12]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481--2495 (2016)

Badrinarayanan, V., Kendall, A., Cipolla, R.: Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 39, 2481--2495 (2016)

work page 2016

[2] [2]

In: ICML (1993)

Caruana, R.: Multitask learning: A knowledge-based source of inductive bias. In: ICML (1993)

work page 1993

[3] [3]

2017 IEEE Visual Communications and Image Processing (VCIP) pp

Chaurasia, A., Culurciello, E.: Linknet: Exploiting encoder representations for efficient semantic segmentation. 2017 IEEE Visual Communications and Image Processing (VCIP) pp. 1--4 (2017)

work page 2017

[4] [4]

In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Chen, H., Qi, X., Yu, L., Heng, P.: DCAN: Deep Contour-Aware Networks for Accurate Gland Segmentation . In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 2487--2496 (2016)

work page 2016

[5] [5]

Csurka, G., Larlus, D.: What is a good evaluation measure for semantic segmentation? In: Proceedings of the British Machine Vision Conference (2013)

work page 2013

[6] [6]

a henb\

Kr\" a henb\" u hl, P., Koltun, V.: Efficient inference in fully connected crfs with gaussian edge potentials. In: Advances in Neural Information Processing Systems 24, pp. 109--117 (2011)

work page 2011

[7] [7]

In: Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015

Ronneberger, O., Fischer, P., Brox, T.: U-Net: Convolutional Networks for Biomedical Image Segmentation . In: Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015. pp. 234--241 (2015)

work page 2015

[8] [8]

In: 3rd International Conference on Learning Representations ICLR (2015)

Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. In: 3rd International Conference on Learning Representations ICLR (2015)

work page 2015