Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

Aly Khalifa; Chris McIntosh; Sangwook Kim; Thomas G. Purdie

REVIEW 4 major objections 5 minor 27 references

Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy

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

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

Pith's one-line read This paper argues that automated contouring and dose prediction in radiotherapy should be trained jointly, reporting DVH error reductions of 19.82% (prostate) and 16.33% (head and neck) versus a sequential automated pipeline.

desk verdict A useful sequential-vs-MTL comparison in automated radiotherapy planning, but capacity is unmatched and the headline gains are overstated. read the letter →

arxiv 2411.18767 v1 pith:W63XQK63 submitted 2024-11-27 physics.med-ph cs.CVcs.LG

Sangwook Kim , Aly Khalifa , Thomas G. Purdie , Chris McIntosh This is my paper · ORCID

classification physics.med-phcs.CVcs.LG

keywords multi-tasklearningautomatedcontouringdosepredictionradiotherapytreatmentplanningvolumehistogramdeepprostatecancerheadandneck

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

Automated radiotherapy planning today runs two deep-learning jobs in series: first a model contours the organs, then a second model uses those contours to predict the radiation dose distribution. This paper argues that the two jobs should be trained together in one multi-task network that takes only a CT scan and outputs contours and a dose map at the same time. On an in-house prostate dataset and the public OpenKBP head-and-neck dataset, the integrated model reduced the mean absolute error of dose-volume histogram metrics by 19.82% and 16.33%, respectively, compared with the sequential automated pipeline, while matching or improving contouring accuracy. If the finding holds, integrated training could make fully automated treatment planning faster and less sensitive to imperfect input contours.

What carries the argument

The load-bearing mechanism is a cross-task attention network: one shared encoder extracts common CT features, while separate output branches predict contours and dose, and specialized attention modules within the encoder let each task weigh the shared features differently. Training balances a mean-absolute-error loss for dose with a combo loss (dice plus cross-entropy) for contours, and a dynamic weight average adjusts the relative loss weights during training. This mechanism is what lets the model output contours and dose simultaneously from CT alone, and it is the component the paper credits for the reported DVH-MAE improvements.

What would settle it

Train a single-task dose-prediction model with the same encoder capacity, pre-trained backbone, and attention machinery as Multi-Automated but without the contouring output branch; if its mean absolute error of the dose-volume histogram (DVH) metrics matches the reported 3.528 Gy and 10.109 Gy values, the improvement is due to model capacity rather than to joint contouring and dose training.

Watch

Extended reading notes

Core claim

The paper's central claim is that multi-task learning, rather than a sequential contour-then-plan workflow, is the right way to structure automated radiotherapy planning. In the proposed Multi-Automated model, a single shared encoder processes the CT image and feeds two task-specific output branches, one for organ contours and one for voxel-wise dose, linked by cross-task attention. The authors report that this integrated model outperforms both sequential baselines: it achieves a DVH-MAE of 3.528 Gy versus 4.301 Gy for sequential automated planning on prostate and 10.109 Gy versus 11.650 Gy on head and neck, and it improves average contouring Dice coefficients from 0.818 to 0.824 on prostate and from 0.674 to 0.716 on head and neck. The authors interpret this as evidence that sharing features between contouring and dose prediction lets the model learn anatomical relationships that a dose-only or contour-then-dose model cannot exploit.

Load-bearing premise

The central comparison assumes that the dose-prediction gain is caused by multi-task learning itself, but the multi-task model is a larger, more complex network than the single-task baselines, and no capacity-matched single-task model is tested.

Editorial extensions

If this is right

A fully automated planning pipeline no longer needs a separate contouring model or pre-computed contours at inference time: CT goes in, contours and dose come out.
Dose prediction becomes less sensitive to contour quality, because the model is not forced to trust imperfect input contours.
The reported gains on prostate and head-and-neck datasets suggest the benefit is not site-specific and may transfer to other treatment sites.
Integrated training can slightly improve contouring as well as dose prediction, so automation need not trade one task against the other.

Reading between the lines

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

The paper does not compare Multi-Automated against a single-task dose predictor with the same parameter count or the same pre-trained encoder; until such a comparison is run, part of the reported gain could be attributed to model capacity rather than to multi-task learning.
Because the OpenKBP CT-only baseline already beat both contour-based sequential models, the benefit of the integrated model there may partly come from its ability to ignore noisy contour labels; corrupting input contours in a controlled experiment would test this.
The method is evaluated on 2D slices and a limited set of organs; translating it to full 3D plans with more anatomical structures is the natural next test of clinical value.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Dear [Colleague],

The useful thing here is a head-to-head comparison: an integrated multi-task network that segments and predicts dose in one pass versus the standard sequential pipeline (automated contouring, then dose prediction) on prostate and OpenKBP head-and-neck data. The integrated model beats the sequential automated pipeline on average DVH-MAE at both sites, and the authors are honest enough to report that a CT-only single-task baseline beats them on OpenKBP. That is the kind of result worth discussing.

What is new: prior work from this group introduced the cross-task attention architecture [25], and Jiao et al. did MTL for dose prediction. This paper's contribution is the explicit comparison to sequential human and automated pipelines, plus the observation that sequential automated contouring can degrade dose prediction relative to a CT-only baseline. The contouring results are also presented cleanly, with dice and Hausdorff per structure.

The soft spots are real. The MTL model has a shared encoder, two task-specific decoders, and cross-task attention, while the sequential and baseline models are single-decoder U-Nets. No parameter-matched ablation is run, so the improvement could be a capacity effect, not multi-task learning. On OpenKBP the simple CT-only baseline (9.436 Gy) actually beats the MTL model (10.109 Gy), which alone undercuts the claim that MTL is the best way to predict dose.

The abstract's headline numbers are also misleading: 19.82% and 16.33% are improvements relative to the baseline model, not against the Seq-Automated pipeline. Against Seq-Automated the gains are 17.97% and 13.23%—still positive but smaller. Reporting lacks error bars, seeds, and full p-values, and no code or data are released. The prostate D99 is worse with MTL, so the average gain is carried by other metrics.

None of this sinks the paper's value as an empirical comparison. The fix is straightforward: add a parameter-matched single-task dose predictor (or a capacity-matched MTL) and report the Seq-Automated-relative numbers correctly. I would send this to peer review, but with the expectation of major revisions. The question is important and the comparison is useful, but the attribution to multi-task learning is not yet established.

It is a maybe for our reading group—useful for a discussion on confounds in MTL papers, but not a must-read.

Best, [Your name]

Referee Report

4 major / 5 minor

Summary. The paper proposes a multi-task learning (MTL) framework that simultaneously performs automated contouring and voxel-based dose prediction for radiotherapy, using a shared encoder, two task-specific decoders, and cross-task attention taken from the authors' prior work [25]. The method is evaluated on an in-house prostate dataset (110 patients) and the public OpenKBP head-and-neck dataset (318 patients after exclusions), and is compared against a CT-only baseline, a sequential model using human contours (Seq-Human), and a sequential model using automated contours (Seq-Automated). The authors report average DVH-MAE improvements of Multi-Automated over Seq-Automated on both datasets, together with contouring Dice and Hausdorff distance results, and conclude that MTL can integrate contouring and dose prediction without compromising contouring accuracy.

Significance. If the central attribution to multi-task learning were established, this would be a practically relevant contribution: an integrated model that takes only CT as input and outputs both contours and dose could simplify the radiotherapy planning workflow and reduce sensitivity to contour quality. The use of the public OpenKBP dataset, the inclusion of both human-contour and automated-contour sequential baselines, and the reporting of two anatomically distinct sites are strengths. However, the manuscript as written does not establish that multi-task learning, rather than increased model capacity, drives the observed improvements; the abstract's headline percentages are not supported by the tables; and the statistical reporting is incomplete. The paper is a reasonable empirical study but requires substantial revision before the central claim can be accepted.

major comments (4)

[Abstract; Section 3.1, Tables 2 and 3] The headline percentages do not follow from the reported averages. From Tables 2 and 3, Multi-Automated improves over Seq-Automated by (4.301−3.528)/4.301 = 17.97% on prostate and (11.650−10.109)/11.650 = 13.23% on OpenKBP, not the 19.82% and 16.33% stated in the abstract. The relative differences listed in the tables are computed against the Baseline column, not against Seq-Automated, so the abstract appears to use a different denominator. Because these percentages are the paper's primary quantitative claim, they must be corrected and the denominator explicitly defined.
[Section 2.2.2-2.2.3; Tables 2 and 3] The comparison does not control for model capacity. Multi-Automated uses a shared encoder plus two task-specific decoders and cross-task attention, whereas Seq-Automated and the baseline are single-decoder U-Nets. No parameter-matched single-task dose predictor or MTL ablation is reported, so the observed DVH-MAE gains over Seq-Automated could be due to the larger model rather than to multi-task learning. The OpenKBP result is especially telling: the CT-only single-task Baseline achieves 9.436 Gy, beating Multi-Automated's 10.109 Gy on average, so the integrated model is not universally superior to a strong single-task predictor. A matched-capacity single-task dose model, or an MTL variant without cross-task attention, is required before the central MTL attribution can be accepted.
[Section 2.3.3; Tables 2-5] Statistical support is underreported. The tables only mark significance with daggers, without reporting p-values, standard deviations, confidence intervals, number of random seeds, or training epochs. With test sizes of 10 (prostate) and 78 (OpenKBP) and many DVH metrics, the absence of uncertainty measures makes it impossible to assess the stability of the reported average improvements. In addition, the caption of Table 3 states 'p < 0.5' instead of 'p < 0.05', which appears to be a typo but should be corrected.
[Section 3.2, Table 4] The claim that contouring accuracy is 'maintained or sometimes even improved' is only partially supported. On the prostate dataset, Multi-Automated's average Hausdorff distance is worse than the baseline (12.049 vs 7.549), and the right femur Hausdorff distance deteriorates significantly (36.441 vs 9.669), while Dice is essentially unchanged. On OpenKBP, Dice improves significantly for several ROIs, but the spinal cord Hausdorff distance worsens. The discussion should report these trade-offs explicitly rather than emphasizing the improvements only.

minor comments (5)

[Section 2.1] The dataset arithmetic is inconsistent: the text says 328 patients were collected and 22 omitted, but the stated split of 200/40/78 sums to 318, whereas 328−22=306. Please reconcile these numbers and state how the omitted cases are distributed across the splits.
[Throughout] There are several typographical errors: 'able able' in the abstract, 'respectivley' in the Figure 1 caption, 'radiotherpay' in Table 1, 'coeffcient' in Tables 4 and 5, and the 'p < 0.5' captions in Tables 3 and 5. These should be corrected.
[Section 2.2.4] Training details are incomplete: the number of epochs, learning-rate schedule, dynamic weight average hyperparameters, and whether results come from a single run or multiple seeds are not reported. A code or data availability statement would also improve reproducibility.
[Section 1] The paper says the framework is novel 'to the best of our knowledge,' but the architecture is from the authors' prior paper [25] and another multi-task dose-prediction/contouring study [24] is cited. Please state explicitly which components are new in this work, e.g., the systematic sequential-model comparison and dual-site validation, to avoid overclaiming novelty.
[Section 2.3.1, Table 1] The description of D0.1cc is awkward: it is written as 'Maximum dose received by the smallest volume of 0.1 cc.' Standard D0.1cc is the dose received by 0.1 cc of the structure, not a maximum-dose definition; please rephrase for clarity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central MTL-vs-sequential claim is an empirical held-out comparison, not a construction that encodes its own conclusion.

full rationale

The paper's main claim is that Multi-Automated reduces DVH-MAE relative to sequential pipeline models. This is assessed on held-out test patients (10 prostate, 78 OpenKBP) with DVH-MAE computed from predicted versus clinical dose distributions. There is no fitted parameter that by construction equals the reported improvement; the loss functions (MAE, combo loss) and dynamic weight averaging are standard training objectives, not definitions of the evaluation metric. The self-citations to [25] supply the cross-task attention architecture and the baseline model, but the performance comparison is not derived from that citation: it is a direct experimental result, and the OpenKBP evaluation is externally benchmarked. The capacity-mismatch concern (Multi-Automated has an extra decoder and attention versus single-decoder baselines) is a legitimate experimental-design limitation, but it is not an instance of self-definition, fitted-input-as-prediction, or renaming. The abstract's 19.82%/16.33% figures use the CT-only baseline as denominator rather than Seq-Automated, and the paper's Tables 2-3 report different relative differences; this is a reporting issue, not a circular derivation. No load-bearing step reduces, by the paper's own equations or by self-citation, to its own inputs.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central empirical claim rests on standard deep-learning training assumptions, the validity of the two datasets as ground truth, and the assumption that the sequential baselines are fair and comparable. No new physical entities are introduced. The main uncharged assumptions are that 2D slices and the chosen loss weights are adequate and that the comparison isolates the effect of multi-task learning.

free parameters (6)

combo loss weight for dice loss = 0.3
Chosen by hand in Sec. 2.2.4; affects contouring and hence the shared representation.
combo loss weight for cross-entropy loss = 0.7
Chosen by hand in Sec. 2.2.4; paired with the dice weight.
dynamic weight average parameters = not specified
Used to balance task losses in Sec. 2.2.4, but the exact schedule and parameters are not given.
batch size = 32 (Prostate), 8 (OpenKBP)
Hyperparameters chosen in Sec. 2.2.4 with no sensitivity analysis.
learning rate = 1e-4
Adam optimizer setting in Sec. 2.2.4.
weight decay = 1e-5
Adam optimizer setting in Sec. 2.2.4.

assumptions (5)

domain assumption 2D axial CT slices are sufficient for dose prediction and contouring accuracy.
All models process 2D slices rather than full 3D volumes (Sec. 2.2.1).
domain assumption The four OpenKBP ROIs are consistently and correctly labeled across patients, and the excluded cases do not bias training.
Authors excluded 22 OpenKBP cases due to incomplete contours (Sec. 2.1), assuming the remaining data are reliable.
domain assumption DVH metrics computed using the clinical contours from the datasets are valid ground truth for dose evaluation.
Evaluation uses clinical contours and DVH metrics (Sec. 2.3.1).
domain assumption The cross-task attention network from [25] is a valid and correctly instantiated MTL architecture.
The paper refers to [25] for the architecture details in Sec. 2.2.3.
domain assumption Training with the same loss functions and encoder is fair across baseline, sequential, and multi-task models.
Fairness of comparison is assumed; no parameter-matched control is provided.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy." pith.science (2026). https://pith.science/paper/W63XQK63

@misc{pith2026241118767,
  author       = {Pith},
  title        = {Pith review of: Multi-Task Learning for Integrated Automated Contouring and Voxel-Based Dose Prediction in Radiotherapy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W63XQK63}},
  note         = {Machine review of arXiv:2411.18767}
}

read the original abstract

Deep learning-based automated contouring and treatment planning has been proven to improve the efficiency and accuracy of radiotherapy. However, conventional radiotherapy treatment planning process has the automated contouring and treatment planning as separate tasks. Moreover in deep learning (DL), the contouring and dose prediction tasks for automated treatment planning are done independently. In this study, we applied the multi-task learning (MTL) approach in order to seamlessly integrate automated contouring and voxel-based dose prediction tasks, as MTL can leverage common information between the two tasks and be able able to increase the efficiency of the automated tasks. We developed our MTL framework using the two datasets: in-house prostate cancer dataset and the publicly available head and neck cancer dataset, OpenKBP. Compared to the sequential DL contouring and treatment planning tasks, our proposed method using MTL improved the mean absolute difference of dose volume histogram metrics of prostate and head and neck sites by 19.82% and 16.33%, respectively. Our MTL model for automated contouring and dose prediction tasks demonstrated enhanced dose prediction performance while maintaining or sometimes even improving the contouring accuracy. Compared to the baseline automated contouring model with the dice score coefficients of 0.818 for prostate and 0.674 for head and neck datasets, our MTL approach achieved average scores of 0.824 and 0.716 for these datasets, respectively. Our study highlights the potential of the proposed automated contouring and planning using MTL to support the development of efficient and accurate automated treatment planning for radiotherapy.

Figures

Figures reproduced from arXiv: 2411.18767 by the authors.

**Figure 1.** Architectures, inputs, and outputs of the (i) sequential human contouring and planning (Seq-Human), (ii) sequential automated contouring and planning (SeqAutomated), and proposed framework (iii) automated contouring and planning using multi-task learning (Multi-Automated). In Seq-Human and Seq-Automated, the input includes clinician labeled ground-truth (GT) and deep learning (DL) regions of interest (ROI), respect… view at source ↗

**Figure 2.** This figure illustrates qualitative results of Prostate dataset, showing input CT scans, ground truth clinical dose distributions (Clinical Dose), and the dose distributions generated by sequential automated contouring and planning model (Seq-Automated) and the automated contouring and planning model using multi-task learning (Multi-Automated). The voxel-wise Mean Absolute Error (MAE) for both model outputs are prov… view at source ↗

**Figure 3.** In this figure, we present the results of automated DL contouring using Prostate dataset. The image showcases the ground truth labels (GT), and the DL contouring outputs from the baseline model (DL-Baseline) and the automated multi-task contouring and planning model (Multi-Automated). The figure includes representative examples, displaying best, median, and worst cases from our test dataset. deliverable plans, a cap… view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: This figure presents qualitative results from the OpenKBP dataset, focusing on the head and neck cancer, showing input CT scans, ground truth clinical dose distributions (Clinical Dose), and the dose distributions generated by sequential automated contouring and planni…

**Figure 5.** Figure 5: In this figure, we present the results of DL automated contouring using OpenKBP dataset. The image showcases the ground truth labels (GT), and the automated contouring outputs from the baseline (DL-Baseline) and the automated contouring and planning model using multi-t…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 23 canonical work pages

[25]

Cross-task attention network: Improving multi-task learning for medical imaging applications

Sangwook Kim, Thomas G Purdie, and Chris McIntosh. Cross-task attention network: Improving multi-task learning for medical imaging applications. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 119–128. Springer, 2023

work page 2023
[1]

Chris McIntosh, Leigh Conroy, Michael C. Tjong, Tim Craig, Andrew Bayley, Charles Catton, Mary Gospodarowicz, Joelle Helou, Naghmeh Isfahanian, Vickie Kong, Tony Lam, Srinivas Raman, Padraig Warde, Peter Chung, Alejandro Berlin, and Thomas G. Purdie. Clinical integration of machine learning for curative-intent radiation treatment of patients with prostate...

work page 2021
[2]

U-net: Convolutional networks for biomedical image segmentation

Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18 , pages 234–241. Springer, 2015

2015
[3]

nnu-net: a self-configuring method for deep learning-based biomedical image segmentation

Fabian Isensee, Paul F Jaeger, Simon AA Kohl, Jens Petersen, and Klaus H Maier- Hein. nnu-net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2):203–211, 2021

2021
[4]

Geometric evaluations of ct and mri based deep learning segmentation for brain oars in radiotherapy

Nouf Alzahrani, Ann Henry, Anna Clark, Louise Murray, Michael Nix, and Bashar Al-Qaisieh. Geometric evaluations of ct and mri based deep learning segmentation for brain oars in radiotherapy. Physics in Medicine & Biology, 68(17):175035, 2023

work page 2023
[5]

Generaliz- ability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy

Ruiyan Ni, Kathy Han, Benjamin Haibe-Kains, and Alexandra Rink. Generaliz- ability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy. Radiotherapy and Oncology, 197:110332, 2024

work page 2024
[6]

A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning

Dan Nguyen, Troy Long, Xun Jia, Weiguo Lu, Xuejun Gu, Zohaib Iqbal, and Steve Jiang. A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning. Scientific Reports, 9(1):1076, 2019

work page 2019
[7]

Moore, Thomas G

Aaron Babier, Binghao Zhang, Rafid Mahmood, Kevin L. Moore, Thomas G. Purdie, Andrea L. McNiven, and Timothy C. Y. Chan. OpenKBP: The open- access knowledge-based planning grand challenge and dataset. Medical Physics , 48(9):5549–5561, 2021

work page 2021

Show all 27 references

[8]

OpenKBP-Opt: An international and reproducible evaluation of 76 knowledge-based planning pipelines

Aaron Babier, Rafid Mahmood, Binghao Zhang, Victor G L Alves, Ana Maria Barrag´ an-Montero, Joel Beaudry, Carlos E Cardenas, Yankui Chang, Zijie Chen, Jaehee Chun, Kelly Diaz, Harold David Eraso, Erik Faustmann, Sibaji Gaj, Skylar Gay, Mary Gronberg, Bingqi Guo, Junjun He, Ger...

2022
[9]

Deep learning–based dose prediction for automated, individualized quality assurance of head and neck radiation therapy plans

Mary P Gronberg, Beth M Beadle, Adam S Garden, Heath Skinner, Skylar Gay, Tucker Netherton, Wenhua Cao, Carlos E Cardenas, Christine Chung, David T Fuentes, et al. Deep learning–based dose prediction for automated, individualized quality assurance of head and neck radiation th...

2023
[10]

A transformer-embedded multi-task model for dose distribution prediction

Lu Wen, Jianghong Xiao, Shuai Tan, Xi Wu, Jiliu Zhou, Xingchen Peng, and Yan Wang. A transformer-embedded multi-task model for dose distribution prediction. International Journal of Neural Systems , 33(08):2350043, 2023

2023
[11]

A cascade 3d u-net for dose prediction in radiotherapy

Shuolin Liu, Jingjing Zhang, Teng Li, Hui Yan, and Jianfei Liu. A cascade 3d u-net for dose prediction in radiotherapy. Medical physics, 48(9):5574–5582, 2021

2021
[12]

3D radiotherapy dose prediction on head and neck cancer patients with a hierarchically densely connected U-net deep learning architecture

Dan Nguyen, Xun Jia, David Sher, Mu-Han Lin, Zohaib Iqbal, Hui Liu, and Steve Jiang. 3D radiotherapy dose prediction on head and neck cancer patients with a hierarchically densely connected U-net deep learning architecture. Physics in Medicine & Biology , 64(6):065020, 2019

2019
[13]

Flexible- cm gan: Towards precise 3d dose prediction in radiotherapy

Riqiang Gao, Bin Lou, Zhoubing Xu, Dorin Comaniciu, and Ali Kamen. Flexible- cm gan: Towards precise 3d dose prediction in radiotherapy. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 715–725, 2023

2023
[14]

Diffdp: Radiotherapy dose prediction via a diffusion model

Zhenghao Feng, Lu Wen, Peng Wang, Binyu Yan, Xi Wu, Jiliu Zhou, and Yan Wang. Diffdp: Radiotherapy dose prediction via a diffusion model. In International Conference on Medical Image Computing and Computer-Assisted Intervention , pages 191–201. Springer, 2023

2023
[15]

Dose distribution prediction for head-and-neck cancer radiotherapy using a generative adversarial network: influence of input data

Xiaojin Gu, Victor IJ Strijbis, Ben J Slotman, Max R Dahele, and Wilko F AR Verbakel. Dose distribution prediction for head-and-neck cancer radiotherapy using a generative adversarial network: influence of input data. Frontiers in Oncology, 13, 2023

2023
[16]

Multitask learning

Rich Caruana. Multitask learning. Machine learning, 28:41–75, 1997

1997
[17]

Generalized multi-task learning from substantially unlabeled multi-source medical image data

Ayaan Haque, Abdullah-Al-Zubaer Imran, Adam Wang, and Demetri Terzopoulos. Generalized multi-task learning from substantially unlabeled multi-source medical image data. arXiv preprint arXiv:2110.13185 , 2021

2021 arXiv
[18]

Multi-task deep learning based ct imaging analysis for covid-19 pneumonia: Classification and segmentation

Amine Amyar, Romain Modzelewski, Hua Li, and Su Ruan. Multi-task deep learning based ct imaging analysis for covid-19 pneumonia: Classification and segmentation. Computers in biology and medicine , 126:104037, 2020. REFERENCES 19

2020
[19]

Multi-task attention-based semi-supervised learning for medical image segmentation

Shuai Chen, Gerda Bortsova, Antonio Garc ´ ıa-Uceda Ju´ arez, Gijs Van Tulder, and Marleen De Bruijne. Multi-task attention-based semi-supervised learning for medical image segmentation. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2019: 22nd Internatio...

2019
[20]

Deep learning for multi-task medical image segmentation in multiple modalities

Pim Moeskops, Jelmer M Wolterink, Bas HM Van Der Velden, Kenneth GA Gilhuijs, Tim Leiner, Max A Viergever, and Ivana Iˇ sgum. Deep learning for multi-task medical image segmentation in multiple modalities. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 20...

2016
[21]

Multi-task deep learning for medical image computing and analysis: A review

Yan Zhao, Xiuying Wang, Tongtong Che, Guoqing Bao, and Shuyu Li. Multi-task deep learning for medical image computing and analysis: A review. Computers in Biology and Medicine , 153:106496, 2023

2023
[22]

Conversion of single-energy ct to parametric maps of dual-energy ct using convolutional neural network.British Journal of Radiology, 97(1158):1180– 1190, 2024

Sangwook Kim, Jimin Lee, Jungye Kim, Bitbyeol Kim, Chang Heon Choi, and Seongmoon Jung. Conversion of single-energy ct to parametric maps of dual-energy ct using convolutional neural network.British Journal of Radiology, 97(1158):1180– 1190, 2024

2024
[23]

Multimodal radiotherapy dose prediction using a multi-task deep learning model

Austen Maniscalco, Ezek Mathew, David Parsons, Justin Visak, Mona Arbab, Prasanna Alluri, Xingzhe Li, Narine Wandrey, Mu-Han Lin, Asal Rahimi, et al. Multimodal radiotherapy dose prediction using a multi-task deep learning model. Medical physics, 2024

2024
[24]

Mask-free radiotherapy dose prediction via multi-task learning

Zhengyang Jiao, Xingchen Peng, Jianghong Xiao, Xi Wu, Jiliu Zhou, and Yan Wang. Mask-free radiotherapy dose prediction via multi-task learning. In 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI) , pages 1–5. IEEE, 2022

2022
[26]

Loss odyssey in medical image segmentation

Jun Ma, Jianan Chen, Matthew Ng, Rui Huang, Yu Li, Chen Li, Xiaoping Yang, and Anne L Martel. Loss odyssey in medical image segmentation. Medical Image Analysis, 71:102035, 2021

2021
[27]

End-to-end multi-task learning with attention

Shikun Liu, Edward Johns, and Andrew J Davison. End-to-end multi-task learning with attention. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 1871–1880, 2019

2019

Pith tools

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

[1] [25]

Cross-task attention network: Improving multi-task learning for medical imaging applications

Sangwook Kim, Thomas G Purdie, and Chris McIntosh. Cross-task attention network: Improving multi-task learning for medical imaging applications. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 119–128. Springer, 2023

work page 2023

[2] [1]

Chris McIntosh, Leigh Conroy, Michael C. Tjong, Tim Craig, Andrew Bayley, Charles Catton, Mary Gospodarowicz, Joelle Helou, Naghmeh Isfahanian, Vickie Kong, Tony Lam, Srinivas Raman, Padraig Warde, Peter Chung, Alejandro Berlin, and Thomas G. Purdie. Clinical integration of machine learning for curative-intent radiation treatment of patients with prostate...

work page 2021

[3] [2]

U-net: Convolutional networks for biomedical image segmentation

Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, part III 18 , pages 234–241. Springer, 2015

2015

[4] [3]

nnu-net: a self-configuring method for deep learning-based biomedical image segmentation

Fabian Isensee, Paul F Jaeger, Simon AA Kohl, Jens Petersen, and Klaus H Maier- Hein. nnu-net: a self-configuring method for deep learning-based biomedical image segmentation. Nature methods, 18(2):203–211, 2021

2021

[5] [4]

Geometric evaluations of ct and mri based deep learning segmentation for brain oars in radiotherapy

Nouf Alzahrani, Ann Henry, Anna Clark, Louise Murray, Michael Nix, and Bashar Al-Qaisieh. Geometric evaluations of ct and mri based deep learning segmentation for brain oars in radiotherapy. Physics in Medicine & Biology, 68(17):175035, 2023

work page 2023

[6] [5]

Generaliz- ability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy

Ruiyan Ni, Kathy Han, Benjamin Haibe-Kains, and Alexandra Rink. Generaliz- ability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy. Radiotherapy and Oncology, 197:110332, 2024

work page 2024

[7] [6]

A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning

Dan Nguyen, Troy Long, Xun Jia, Weiguo Lu, Xuejun Gu, Zohaib Iqbal, and Steve Jiang. A feasibility study for predicting optimal radiation therapy dose distributions of prostate cancer patients from patient anatomy using deep learning. Scientific Reports, 9(1):1076, 2019

work page 2019

[8] [7]

Moore, Thomas G

Aaron Babier, Binghao Zhang, Rafid Mahmood, Kevin L. Moore, Thomas G. Purdie, Andrea L. McNiven, and Timothy C. Y. Chan. OpenKBP: The open- access knowledge-based planning grand challenge and dataset. Medical Physics , 48(9):5549–5561, 2021

work page 2021