Pith. sign in

REVIEW 3 major objections 5 minor 46 references

Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read A multimodal deep learning model can classify pancreatic ductal adenocarcinoma resectability into the three NCCN categories from contrast-enhanced CT plus 17 clinical variables, reporting accuracy 0.85–0.87 on internal and external cohorts

desk verdict A solid applied multimodal pipeline for PDAC resectability, with known-component novelty; two upstream-information leaks and an unvalidated reference standard keep it from being more than that. read the letter →

arxiv 2607.13826 v1 pith:QIVHCTOR submitted 2026-07-15 cs.CV cs.AIcs.IRcs.LG

Vincent Ochs , Christoph Kuemmerli , Florentin Bieder , Julia Wolleb , Joel L. Lavanchy , Julia Ruppel , Jan Liechti , Stephanie Taha-Mehlitz
show 7 more authors
Christian Andreas Nebiker Beat Mueller Giuseppe Kito Fusai Joerg-Matthias Pollok Anas Taha Philippe C. Cattin Sebastian Staubli
This is my paper · ORCID
classification cs.CVcs.AIcs.IRcs.LG
keywords pancreaticcancerPDACresectabilityNCCNclassificationmultimodaldeeplearningCTimagingauxiliarysegmentationmultitask
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 aims to show that pancreatic ductal adenocarcinoma resectability—whether surgery should be attempted up front, only after neoadjuvant therapy, or not at all under NCCN rules—can be predicted automatically from a single contrast-enhanced CT scan plus routine clinical variables. The authors train a multimodal network whose imaging branch is supervised to segment the pancreas, tumor, and key peripancreatic vessels, then fuse those learned anatomical features with a compact embedding of 17 clinical variables. They report consistent performance on nested cross-validation and on an independent 52-patient cohort, with AUC around 0.86 and accuracy 0.85–0.87. They argue that because segmentation is needed only during training, the resulting system gives vessel-aware, interpretable predictions without manual masks at test time. If true, this would give clinicians a reproducible second opinion on tumor-board decisions that today vary substantially between experts.

What carries the argument

The load-bearing object is the fused representation: a 256-dimensional imaging vector from a Swin-UNETR bottleneck concatenated with a 32-dimensional clinical embedding from a three-layer MLP, forming a 288-dimensional multimodal vector for the classification head. Training is controlled by a dynamic multitask schedule in which the segmentation weight w_seg drops from 3.0 to 1.0 (and w_cls = 1/w_seg rises) as the tumor Dice score improves, so the shared encoder first learns anatomy and then is tuned for discrimination. This design transfers vessel-aware structure into the classifier without requiring segmentation masks at inference.

What would settle it

Run a prospective multi-reader study in which several blinded radiologists independently grade the same CT volumes and compare their agreement rates with the model's agreement to the MDT label; if the model agrees with the board no more often than a typical clinician does, the claim of adding reproducible objective information collapses. Additionally, checking whether the external cohort's labels were produced by an independent MDT would show whether the generalization result is merely shared-label transfer.

Watch

Extended reading notes

Core claim

The paper's central claim is that integrating anatomical supervision with clinical context yields a robust and reproducible tool for NCCN-based operability assessment in pancreatic cancer. The authors show that a Swin-UNETR encoder trained jointly on CT segmentation and classification, fused with an MLP embedding of 17 clinical variables, can predict upfront-resectable, borderline-resectable, and locally-advanced disease. They report AUC 0.86, macro-F1 0.79, and accuracy 0.85 in nested cross-validation, and AUC 0.86, macro-F1 0.81, and accuracy 0.87 on an independent 52-patient cohort. They argue that because segmentation is auxiliary to training, the system delivers vessel-aware representat

Load-bearing premise

The evaluation assumes the tumor-board-assigned NCCN labels are true resectability; if those labels are noisy or reflect local institutional treatment habits—the paper itself notes inter-observer agreement on NCCN categories is often below 70%—then the reported accuracy measures agreement with one board's decisions rather than true resectability.

Editorial extensions

If this is right

  • If the reported metrics hold in prospective testing, an automated second opinion becomes feasible for a decision whose inter-observer agreement is often below 70%.
  • The system's mask-free inference removes the need for a separate, fragile segmentation step in deployment, making the workflow closer to a click-and-report tool.
  • Ablations show CT-only (AUC 0.82) and tabular-only (AUC 0.74) each underperform the fused model (AUC 0.86), implying both modalities contribute complementary signal rather than one being redundant.
  • The external cohort results suggest transfer across institutions and scanner settings without retraining, although the paper treats this as preliminary evidence needing further validation.

Reading between the lines

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

  • Because the reference labels are MDT consensus without quantified inter-reader variability, the reported AUC/F1 values may chiefly measure agreement with one tumor board's operational decisions; a multi-reader study is needed to show the model adds signal beyond a typical clinician.
  • The geometry analysis is a testable opening: if angle-based contact surrogates were made explicit outputs, the model could generate human-readable explanations that clinicians can audit, turning a black box into a decision-support tool.
  • The adaptive segmentation-to-classification schedule could be reused in any task with scarce labels and a strong auxiliary dense task, such as tumor staging in other organs, but that transfer remains to be demonstrated.
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

3 major / 5 minor

Summary. The paper proposes a fully automated multimodal deep learning framework for classifying pancreatic ductal adenocarcinoma into three NCCN resectability categories (upfront resectable, borderline resectable, locally advanced) from 3D contrast-enhanced CT and 17 structured clinical variables. A Swin-UNETR encoder–decoder provides auxiliary multi-organ segmentation supervision during training, while a fused imaging–clinical representation is used for classification; an adaptive multitask loss shifts weight from segmentation to classification as tumor Dice improves. The authors report internal stratified nested 5-fold cross-validation on 159 patients (AUC 0.86, macro-F1 0.79, accuracy 0.85) and external validation on 52 patients from another institution (AUC 0.86, macro-F1 0.81, accuracy 0.87), with ablations showing improved performance over unimodal and adapted baseline methods. The central claim is that integrating anatomical supervision with clinical context yields a robust, reproducible tool for NCCN-based resectability assessment.

Significance. If the central claim holds, the framework is a potentially useful step toward automated, mask-free resectability assessment, and the methodological ingredients are of interest: auxiliary segmentation supervision without inference-time masks, adaptive multitask weighting, and public code release. The internal nested CV is well designed, and the ablation study is thorough. However, the validity of the reported metrics depends on two load-bearing issues: the inclusion of post-treatment clinical variables that are downstream of the label being predicted, and the reliability and consistency of the NCCN reference labels. These issues must be addressed before the headline results can be interpreted as evidence of a robust clinical tool.

major comments (3)
  1. [§3.1, Clinical Data] The clinical feature vector includes 'neoadjuvant (radio-)chemotherapy status' and 'CA 19-9 serum level after neoadjuvant therapy'. These variables are determined after diagnosis and are strongly influenced by the initial NCCN category, since borderline/locally advanced patients are typically treated neoadjuvantly while upfront resectable patients proceed to surgery. Including them as predictors therefore introduces target leakage: the model can partially infer the label from treatment history. This is especially relevant because Table 3 shows tabular-only models achieve non-trivial performance (AUC 0.74–0.75), and the imaging-only model is weaker. The authors should retrain and re-evaluate the model with all post-diagnosis treatment variables removed (or restricted to baseline CA19-9), and report whether the multimodal advantage and the reported AUC/F1 are preserved. Without this, the c
  2. [§4.1 and §5, Reference standard / label noise] The paper's own introduction cites inter-observer agreement for resectability assessment often below 70% even under standardized criteria (Giannone et al., 2021; Katz et al., 2013). Yet the training labels are MDT consensus from two Basel centers, with a single additional physician review and no reported agreement statistics, and the external KSA Aarau cohort's label generation process is not described. The reported AUC/F1 therefore measure agreement with one (or two) institutional tumor boards' decisions rather than an objective ground truth; if label noise is substantial or institutional practices differ, the model may be fitting idiosyncratic decisions. The authors acknowledge the lack of a multi-reader study in Section 5, but this is not merely a future-work item: it is load-bearing for the interpretation of every performance number. At minimum, the paper should report inter-observer
  3. [§4.5, Table 2, External validation] The external validation cohort contains only 52 patients, including 10 locally advanced cases, and results are reported as point estimates with no confidence intervals. Given the small sample, an AUC of 0.86 with no uncertainty bound cannot establish 'cross-institution generalization' in a statistically meaningful way; the 95% CI on a three-class AUC from 52 patients is likely wide (of the order of ±0.1 or more). The authors should provide bootstrap or exact CIs, per-class counts, and ideally a formal comparison with the internal performance. This is a required addition to support the paper's cross-institution claim, which is a central part of the abstract and conclusion.
minor comments (5)
  1. [§4.5/§4.4] Section 4.5 states that 'segmentation outputs were used only for evaluation', but Section 4.4 explicitly uses model-predicted segmentations to compute geometric contact angles for the NCCN geometry analysis. The wording should be clarified to 'not used for classification at inference'.
  2. [Eq. (2)] The coefficients α, β, ω in the segmentation loss are not given numeric values in the main text; the values are said to be grid-searched in the inner loop. Reporting the selected values (or range) would improve reproducibility.
  3. [Table 5] The adaptive weighting ablation reports 'Mean Tumor Dice' with standard deviations, but the main text does not explain how tumor Dice is averaged across folds and whether it is computed on the validation or test set. Please clarify.
  4. [Table 1 and Table 2] The Viviers-based baseline is adapted from DPCG to NCCN criteria, but the conversion is not described. Since this baseline underperforms the proposed method, a brief explanation of the adaptation would help readers assess fairness.
  5. [Author list] Minor typographical issues: 'Jo¨el' and 'W olleb' contain mis-encoded characters; these should be fixed in the camera-ready version.

Circularity Check

1 steps flagged · score 5.0 of 10

Clinical inputs include treatment consequences of the NCCN label, making the reported prediction partially self-referential.

  1. self definitional [Section 3.1 (Clinical Data) and Section 4.1 (Dataset)]
    "Seventeen structured variables ... encompassing ... CA 19-9 serum level after neoadjuvant therapy ... histopathological grade (G) ... and neoadjuvant (radio-)chemotherapy status. ... Ground-truth NCCN resectability labels were defined during routine clinical care by the multidisciplinary tumor board ... according to NCCN criteria."

    The NCCN category is the baseline label being predicted, but neoadjuvant status, post-neoadjuvant CA 19-9, and histopathological grade are consequences of management decisions driven by that category: borderline/locally advanced patients are preferentially treated neoadjuvantly, and grade requires a surgical specimen. Feeding these post-label variables into the classifier lets the model recover the label from its own downstream effects. The reported AUC/F1/accuracy therefore partly measure how well the model reads treatment consequences back off the target, rather than assessing resectability from independent imaging and pre-treatment context. This is load-bearing: the ablation (Table 3) shows the tabular pathway contributes substantially (tabular-only AUC 0.74; imaging knockout drops AUC

full rationale

Most of the paper is not circular. The imaging backbone, auxiliary segmentation supervision, dynamic loss weighting, nested cross-validation, and external KSA Aarau evaluation are conventional self-contained experiments; the baselines are re-implemented and re-trained rather than accepted on authority. There are no load-bearing self-citations or imported uniqueness theorems. The main circularity is the tabular modality: several of the 17 clinical variables are downstream of the NCCN category being predicted (neoadjuvant status, CA 19-9 after neoadjuvant therapy, histopathological grade, and arguably operation timing). Because the NCCN label drives the treatment decisions that generate these variables, the classifier is partly predicting the label from the label's own consequences, which is target leakage and inflates the multimodal result. The paper's own stated limitation—no dedicated multi-reader study and reliance on MDT consensus labels (Section 5)—concerns label reliability rather than derivation circularity, but it compounds the concern because the leaky variables are recorded in the same clinical workflow. Since imaging-only and non-leaky contributions still show independent signal, the paper is not wholly circular, but the central clinical prediction is partially reduced by construction. Score: 5.

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

The central empirical claims rest on a modest retrospective cohort, MDT label reliability, and several hand-selected loss weights and thresholds. No new physical entities are introduced; the main hidden burden is the inclusion of post-diagnosis treatment variables that are causally downstream of the predicted label.

free parameters (4)
  • Classification loss class weights v = [1.0, 2.0, 2.5]
    Smoothed inverse-frequency weights chosen empirically; ablation in Appendix Table 4.
  • Adaptive segmentation/classification weights (w_seg, w_cls) = (3.0, 1/3), (1.5, 2/3), (1.0, 1.0) depending on τ
    Step function in Eq. 4 with values 'empirically chosen to maintain stable gradients'; ablation in Appendix Table 5.
  • Adaptive scheduling thresholds on tumor Dice τ = τ = 0.1 and 0.5
    Thresholds 'reflect typical stages of network convergence observed during pilot training'; tuned on the same dataset family.
  • Grid-searched hyperparameters (LR, feature size, batch size, segmentation loss coefficients) = LR {1e-3,5e-4,1e-4}; feature size {24,32,48}; batch {8,16}; (α,β,ω)∈[0.2,0.6]
    Selected per outer fold on inner validation macro-F1; Section 4.2.
assumptions (5)
  • domain assumption MDT consensus NCCN labels are reliable enough to serve as ground truth.
    All training and evaluation use these labels (Section 4.1), despite cited inter-observer variability.
  • domain assumption A 160×160×160 isotropic 1 mm CT crop contains all peripancreatic structures needed for resectability.
    Preprocessing step in Section 3.1; a larger or differently centered field-of-view could alter vessel-tumor context.
  • ad hoc to paper Post-diagnosis treatment variables are valid predictors of the initial NCCN category.
    Section 3.1 includes 'neoadjuvant (radio-)chemotherapy status' and post-neoadjuvant CA19-9 as inputs; these are downstream consequences of the label, creating a feedback loop.
  • domain assumption Auxiliary segmentation masks are accurate enough to teach vessel-aware features.
    Vessel Dice is 0.67±0.06 (Section 4.5); the paper argues the signal is still useful, but poor vessel masks could misdirect the encoder.
  • domain assumption BTCV pretraining provides useful anatomical priors for this abdominal CT task.
    Stated in Section 3.2; no ablation isolates the effect of pretraining.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning." pith.science (2026). https://pith.science/paper/QIVHCTOR

@misc{pith2026260713826,
  author       = {Pith},
  title        = {Pith review of: Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QIVHCTOR}},
  note         = {Machine review of arXiv:2607.13826}
}
read the original abstract

Accurate determination of pancreatic ductal adenocarcinoma (PDAC) resectability relies on evaluating how the tumor interacts with major peripancreatic vessels on CT imaging, yet expert assessment often shows substantial variability. We introduce a fully automated multimodal deep learning framework that jointly analyzes 3D contrast enhanced CT and structured clinical information to classify patients into the three National Comprehensive Cancer Network (NCCN) resectability categories (upfront resectable, borderline resectable, locally advanced). The approach uses a Swin-UNETR backbone to obtain anatomy aware image representations through auxiliary segmentation of pancreas, tumor, and vascular structures. These features are fused with a compact clinical embedding derived from 17 routinely collected variables and processed by a lightweight classification head. Model training is guided by a dynamic multitask objective that adapts the balance between segmentation and classification based on current tumor Dice performance, promoting feature representations that remain both anatomically informed and discriminative.

Figures

Figures reproduced from arXiv: 2607.13826 by the authors.

Figure 1
Figure 1. Proposed Network Architecture. During backpropagation, gradients from both the segmentation decoder and the classification fusion head are propagated through the shared encoder. This joint optimization enables the encoder to learn anatomically meaningful, discriminative features for resectability prediction. 5 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. NCCN geometry analysis on the outer test folds. Arterial and venous tumor vessel contact angles were approximated from the model predicted segmentation masks and plotted against the ground truth NCCN resectability labels. Colors indicate the three NCCN classes (upfront resectable, borderline resectable, locally advanced), while marker shape denotes the model’s predicted class. Because the original dataset does not c… view at source ↗
Figure 3
Figure 3. Qualitative examples of model predicted multiorgan and vascular segmentations. Left: axial CT slices with overlayed predicted labels. Right: corresponding 3D renderings illustrating pancreas, tumor, and major vessels. 19 [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 4 canonical work pages

  1. [1]

    CA: A Cancer Journal for Clinicians , year=

    Cancer statistics, 2023 , author=. CA: A Cancer Journal for Clinicians , year=

    2023

  2. [2]

    Annals of Surgical Oncology , year=

    Borderline resectable pancreatic cancer: need for standardization and methods for optimal clinical trial design , author=. Annals of Surgical Oncology , year=

  3. [3]

    Nature Reviews Cancer , year=

    Artificial intelligence in radiology , author=. Nature Reviews Cancer , year=

  4. [4]

    CVPR Workshops , year=

    Multi-task learning with dynamic loss balancing for medical image analysis , author=. CVPR Workshops , year=

  5. [5]

    Giannone, Fabio and Capretti, Giovanni and Hilal, Mohammed Abu and Boggi, Ugo and Campra, Donata and Cappelli, Carla and Casadei, Riccardo and De Luca, Raffaele and Falconi, Massimo and Giannotti, Gabriele and others , year =. Resectability of Pancreatic Cancer Is in the Eye of the Observer: A Multicenter, Blinded, Prospective Assessment of Interobserver ...

    work page doi:10.1097/as9.0000000000000087

  6. [6]

    , year =

    Nasief, Haidy and Zheng, Cheng and Schott, Diane and Hall, William and Tsai, Susan and Erickson, Beth and Allen Li, X. , year =. A machine learning based delta-radiomics process for early prediction of treatment response of pancreatic cancer , volume =. npj Precision Oncology , publisher =. doi:10.1038/s41698-019-0096-z , number =

    work page doi:10.1038/s41698-019-0096-z

  7. [7]

    Deep Semantic Segmentation Feature-Based Radiomics for the Classification Tasks in Medical Image Analysis , volume =

    Huang, Bingsheng and others , year =. Deep Semantic Segmentation Feature-Based Radiomics for the Classification Tasks in Medical Image Analysis , volume =. IEEE Journal of Biomedical and Health Informatics , publisher =. doi:10.1109/jbhi.2020.3043236 , number =

    arXiv 2020

  8. [8]

    and others , year =

    Attiyeh, Marc A. and others , year =. Survival Prediction in Pancreatic Ductal Adenocarcinoma by Quantitative Computed Tomography Image Analysis , volume =. Annals of Surgical Oncology , publisher =. doi:10.1245/s10434-017-6323-3 , number =

    work page doi:10.1245/s10434-017-6323-3

Show all 46 references
  1. [9]

    2023 , eprint=

    Improved Prognostic Prediction of Pancreatic Cancer Using Multi-Phase CT by Integrating Neural Distance and Texture-Aware Transformer , author=. 2023 , eprint=

    2023

  2. [10]

    2021 , eprint=

    Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation , author=. 2021 , eprint=

    2021

  3. [11]

    and others , year =

    van Griethuysen, Joost J.M. and others , year =. Computational Radiomics System to Decode the Radiographic Phenotype , volume =. Cancer Research , publisher =. doi:10.1158/0008-5472.can-17-0339 , number =

    doi

  4. [12]

    Regression Shrinkage and Selection Via the Lasso , volume =

    Tibshirani, Robert , year =. Regression Shrinkage and Selection Via the Lasso , volume =. Journal of the Royal Statistical Society Series B: Statistical Methodology , publisher =. doi:10.1111/j.2517-6161.1996.tb02080.x , number =

    1996

  5. [13]

    U-Net: Convolutional Networks for Biomedical Image Segmentation

    Ronneberger, Olaf and Fischer, Philipp and Brox, Thomas. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention -- MICCAI 2015. 2015

    2015

  6. [14]

    DS6, Deformation-Aware Semi-Supervised Learning: Application to Small Vessel Segmentation with Noisy Training Data , volume=

    Chatterjee, Soumick and others , year=. DS6, Deformation-Aware Semi-Supervised Learning: Application to Small Vessel Segmentation with Noisy Training Data , volume=. Journal of Imaging , publisher=. doi:10.3390/jimaging8100259 , number=

    doi

  7. [15]

    2022 , eprint=

    MONAI: An open-source framework for deep learning in healthcare , author=. 2022 , eprint=

    2022

  8. [16]

    2017 , eprint=

    Mining Electronic Health Records: A Survey , author=. 2017 , eprint=

    2017

  9. [17]

    2021 , eprint=

    Fusion of medical imaging and electronic health records with attention and multi-head machanisms , author=. 2021 , eprint=

    2021

  10. [18]

    Radiomics Signature: A Potential Biomarker for the Prediction of Disease-Free Survival in Early-Stage (I or II) Non—Small Cell Lung Cancer , volume =

    Huang, Yanqi and others , year =. Radiomics Signature: A Potential Biomarker for the Prediction of Disease-Free Survival in Early-Stage (I or II) Non—Small Cell Lung Cancer , volume =. Radiology , publisher =. doi:10.1148/radiol.2016152234 , number =

    doi

  11. [19]

    2020 , eprint=

    Multi-Modality Information Fusion for Radiomics-based Neural Architecture Search , author=. 2020 , eprint=

    2020

  12. [20]

    2023 , eprint=

    Radiomics-Informed Deep Learning for Classification of Atrial Fibrillation Sub-Types from Left-Atrium CT Volumes , author=. 2023 , eprint=

    2023

  13. [21]

    2022 , eprint=

    TMSS: An End-to-End Transformer-based Multimodal Network for Segmentation and Survival Prediction , author=. 2022 , eprint=

    2022

  14. [22]

    Journal of Clinical Medicine , VOLUME =

    Khasawneh, Hala and others , TITLE =. Journal of Clinical Medicine , VOLUME =. 2023 , NUMBER =

    2023

  15. [23]

    Diagnostics , VOLUME =

    Litjens, Geke and others , TITLE =. Diagnostics , VOLUME =. 2023 , NUMBER =

    2023

  16. [24]

    2020 , eprint=

    DeepPrognosis: Preoperative Prediction of Pancreatic Cancer Survival and Surgical Margin via Contrast-Enhanced CT Imaging , author=. 2020 , eprint=

    2020

  17. [25]

    CA: A Cancer Journal for Clinicians , volume=

    Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries , author=. CA: A Cancer Journal for Clinicians , volume=. 2021 , publisher=

    2020

  18. [26]

    and Janssen, Boris V

    Bereska, Jacqueline I. and Janssen, Boris V. and Nio, C. Yung and Kop, Marnix P. M. and Kazemier, Geert and Busch, Olivier R. and Struik, Femke and Marquering, Henk A. and Stoker, Jaap and Besselink, Marc G. and Verpalen, Inez M. , year =. Artificial intelligence for assessmen...

    doi

  19. [27]

    2023 , eprint=

    Segmentation-based Assessment of Tumor-Vessel Involvement for Surgical Resectability Prediction of Pancreatic Ductal Adenocarcinoma , author=. 2023 , eprint=

    2023

  20. [28]

    2015 , eprint=

    U-Net: Convolutional Networks for Biomedical Image Segmentation , author=. 2015 , eprint=

    2015

  21. [29]

    World Journal of Oncology , year=

    Epidemiology of pancreatic cancer: Global trends, etiology and risk factors , author=. World Journal of Oncology , year=

  22. [30]

    Journal of the National Comprehensive Cancer Network , year=

    Pancreatic adenocarcinoma, version 2.2021, NCCN clinical practice guidelines in oncology , author=. Journal of the National Comprehensive Cancer Network , year=

    2021

  23. [31]

    Nature Reviews Gastroenterology & Hepatology , year=

    Therapeutic developments in pancreatic cancer: Current and future perspectives , author=. Nature Reviews Gastroenterology & Hepatology , year=

  24. [32]

    Surgery , year=

    Borderline resectable pancreatic cancer: a consensus statement by the International Study Group of Pancreatic Surgery (ISGPS) , author=. Surgery , year=

  25. [33]

    Pancreatology , year=

    International consensus on definition and criteria of borderline resectable pancreatic ductal adenocarcinoma 2017 , author=. Pancreatology , year=

    2017

  26. [34]

    Annals of Surgical Oncology , year=

    Pretreatment assessment of resectable and borderline resectable pancreatic cancer: expert consensus statement , author=. Annals of Surgical Oncology , year=

  27. [35]

    Medical Image Analysis , year =

    A survey on deep learning in medical image analysis , author =. Medical Image Analysis , year =

  28. [36]

    Front Public Health , year=

    Medical image analysis using deep learning algorithms , author=. Front Public Health , year=

  29. [37]

    Medical Physics , year=

    Automatic segmentation of pancreas in abdominal CT images using deep convolutional neural network and self-supervised learning , author=. Medical Physics , year=

  30. [38]

    2024 , issn =

    Automated peripancreatic vessel segmentation and labeling based on iterative trunk growth and weakly supervised mechanism , journal =. 2024 , issn =. doi:https://doi.org/10.1016/j.artmed.2024.102825 , url =

    2024

  31. [39]

    , author=

    Artificial intelligence in pancreatic cancer. , author=. Theranostics , year=

  32. [40]

    , author=

    Pancreatic cancer surgery in the new millennium: better prediction of outcome. , author=. Annals of Surgery , year=

  33. [41]

    , author=

    Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the United States. , author=. Cancer Research , year=

  34. [42]

    European Radiology , year =

    Automated pancreas segmentation and volumetry using deep learning: a large N study , author =. European Radiology , year =

  35. [43]

    European Radiology , year =

    End-to-end prognostication in pancreatic cancer by multimodal deep learning: a retrospective, multicenter study , author =. European Radiology , year =

  36. [44]

    2022 , eprint=

    AMOS: A Large-Scale Abdominal Multi-Organ Benchmark for Versatile Medical Image Segmentation , author=. 2022 , eprint=

    2022

  37. [45]

    Deep multi-task learning structure for segmentation and classification of supratentorial brain tumors in MR images , journal =

    Shirin Kordnoori and Maliheh Sabeti and Mohammad Hossein Shakoor and Ehsan Moradi , keywords =. Deep multi-task learning structure for segmentation and classification of supratentorial brain tumors in MR images , journal =. 2024 , issn =. doi:https://doi.org/10.1016/j.inat.202...

    2024

  38. [46]

    International Conference on Learning Representations , year=

    Decoupled Weight Decay Regularization , author=. International Conference on Learning Representations , year=

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.