Pith. sign in

REVIEW 4 major objections 5 minor 27 references

Advanced Deep Learning Techniques for Automated Segmentation of Type B Aortic Dissections

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A multi-task deep learning pipeline segments Type B aortic dissection on CT with Dice 0.91–0.92 for true lumen, 0.87–0.89 for false lumen, and 0.47 for thrombosis, beating prior published results.

desk verdict Test-set leakage in model selection and an uncontrolled baseline make the reported Dice gains over Yao et al. unreliable, though the pipeline comparison has some utility. read the letter →

arxiv 2506.22222 v1 pith:JYGJPVIG submitted 2025-06-27 eess.IV cs.CV

Hao Xu , Ruth Lim , Brian E. Chapman This is my paper · ORCID
classification eess.IVcs.CV
keywords TypeBaorticdissectionautomatedsegmentationcomputedtomographyangiographydeeplearningfalselumenthrombosismedicalimageensemblemodelSwin-UnetR
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 tries to show that automated segmentation of Type B aortic dissection from CT angiography can be made substantially more accurate than prior published methods, especially for the difficult false lumen thrombosis component. The authors build four deep learning pipelines and report that a sequential multi-task model, which dedicates a separate network to thrombosis segmentation, achieves the highest Dice scores on the three anatomical classes, while an ensemble of two architectures gives the lowest boundary errors. If these numbers hold on new data, the pipeline would let clinicians extract true lumen and false lumen diameters and thrombosis extent from a scan in minutes instead of up to 1.5 hours of manual tracing. The central comparison is against the previous ImageTBAD baseline, whose reported Dice for true lumen, false lumen, and thrombosis were 0.78, 0.68, and 0.25.

What carries the argument

The central mechanism is the sequential multi-task pipeline (Method 3), which isolates the difficult false lumen thrombosis segmentation into its own network rather than asking one network to solve all classes at once. A companion ensemble (Method 4) averages the softmax probabilities of a Swin-UnetR transformer and a 3D U-Net, trading some structure for boundary precision. The pipelines rely on preprocessing that resamples each scan to 1.5 mm isotropic voxels and crops foreground rather than downsampling, preserving small thrombosis structures, and on Dice Cross Entropy Loss, which is reported to outperform Generalized Dice Loss for the transformer network.

What would settle it

Run Method 3 on a fresh pre-registered test set that is completely untouched during model and loss selection, with the FLT-presence classifier left in the loop, and compare True FLT Dice to the reported 0.47; if the classifier's roughly 50% accuracy causes most thrombosed cases to bypass the dedicated network, the effective FLT Dice on real deployment will fall well below the headline number.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that its proposed pipelines—especially the sequential multi-task model (Method 3) and the ensemble model (Method 4)—outperform the prior state of the art for segmenting true lumen, false lumen, and false lumen thrombosis from Type B aortic dissection CTA scans. Method 3 uses a classification network to detect thrombosis, a dedicated segmentation network for the thrombosis when present, a second network for true and false lumen, and a final network that merges the outputs; the authors report test Dice of 0.91 for true lumen, 0.88 for false lumen, 0.38 for all-case false lumen thrombosis, and 0.47 for thrombosis measured only on cases that actually contain it. Method 4 averages the probability outputs of Swin-UnetR and 3D U-Net and achieves the lowest Hausdorff distances for true and false lumen, indicating the most spatially precise boundaries. The paper also reports that the thrombosis-presence classifier itself only reaches about 50% accuracy, so in practice the classification step was bypassed.

Load-bearing premise

The reported test scores assume the test set was never used to pick models or settings; the methods text says networks were selected based on their testing performance, so that assumption may not hold.

Editorial extensions

If this is right

  • If the reported Dice values generalize, clinicians could obtain true lumen, false lumen, and thrombosis segmentations from a CTA in minutes rather than the roughly 1.5 hours of manual tracing, enabling routine serial surveillance.
  • The sequential multi-task design raises the true-lumen thrombosis Dice from 0.32 for the plain two-network cascade to 0.47, suggesting that isolating the rare class in its own network is a workable route to improving segmentation of small structures.
  • Ensembling two architectures yields the lowest Hausdorff distances for true and false lumen, which matters for downstream morphological measurements like lumen diameter and thrombus volume.
  • With accurate automated masks, morphological parameters such as true lumen calibre, false lumen size, and thrombosis extent become feasible to derive routinely for treatment planning and follow-up.

Reading between the lines

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

  • A fair replication should use a three-way split in which the test set is never touched until final evaluation; the methods text says networks were selected based on testing performance, so the reported test Dice may be optimistic.
  • Because the thrombosis-presence classifier reaches only about 50% accuracy, the reported 0.47 True FLT Dice applies only when the dedicated FLT network is actually triggered; in an automated deployment that keeps the classifier in the loop, many thrombosed cases would be missed.
  • The 1.5 mm resampling and cropping may under-resolve small peripheral thrombi; testing at higher input resolution or with patch-based inference could reveal whether 0.47 is a resolution ceiling or a model ceiling.
  • The same multi-task and ensemble designs could transfer to Type A dissection or other thrombus-bearing vessels if labeled training data are available.
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

4 major / 5 minor

Summary. The paper proposes four deep learning pipelines for automated segmentation of true lumen (TL), false lumen (FL), and false lumen thrombosis (FLT) in Type B aortic dissection from CTA images. Using the ImageTBAD dataset (100 cases), the authors compare single-step, sequential, sequential multi-task, and ensemble architectures built on 3D U-Net and Swin-UnetR, with DenseNet classifiers in the multi-task pipeline. They report Dice coefficients and Hausdorff distances, claim that their Method 3 achieves the best FLT segmentation (True FLT Dice 0.47) and that their methods substantially outperform the prior state of the art reported by Yao et al.

Significance. If the reported results were unbiased, the paper would provide a useful comparison of modern segmentation architectures on a public aortic dissection dataset and a potential step forward in FLT segmentation, which is clinically relevant for surveillance and treatment planning. The paper's strengths include the use of a public dataset, the systematic comparison of multiple pipeline designs, and the evaluation of two loss functions. However, the central performance claim is not currently established because the evaluation protocol appears to use the test set for model selection, the baseline comparison to Yao et al. is not controlled, and the experimental design is internally inconsistent. The contribution is therefore more of a proof-of-concept than a reliable advance until these methodological issues are resolved.

major comments (4)
  1. [2.3] Section 2.3 states that the two segmentation networks and two classification networks were 'selecting them based on their testing performance.' This is a direct admission that the test set was used for model selection, which invalidates the test-set Dice coefficients in Table 1 as unbiased estimates of generalization. This concern is compounded by Section 3.1, where the classification step in Method 3 was 'bypassed' after observing poor classification performance, a decision that also appears to be informed by test-set results. Because the paper's central claim—that Method 3 substantially outperforms Yao et al.—rests on these test-set numbers, the reported improvements are likely optimistic. The authors should redo the evaluation with model selection based solely on validation data and report metrics from a test set that is used only once.
  2. [3.1, Table 1, Section 4] The baseline comparison to Yao et al. is not controlled. Section 3.1 says the authors 'implemented the approach of Yao et al. on our dataset also by removing some suboptimal preprocessing steps,' yet Section 4 compares Method 3 to 'their original method's performance of 0.25 ± 0.31 DC for True FLT.' Table 1 lists a row labeled 'Yao et al.'s Testing' with FLT DC 0.15 and True FLT DC 0.25, which is not the original published baseline (the original paper reported FLT DC of 0.25). Since the authors altered the preprocessing (removing downsampling), the reimplementation is a different method, and the comparison to published numbers that were obtained under different conditions is invalid. A valid comparison would reimplement the original method exactly, or clearly label the modified version and avoid quoting the original published numbers.
  3. [2.2] The experimental design is internally inconsistent. The Abstract and Section 2.2 describe a single split into training (n=80), validation (n=10), and testing (n=10), but Section 2.2 also states 'We implemented cross-validation with 5 folds to maximize the utilization of this small dataset.' It is not explained how a fixed 80/10/10 split relates to 5-fold cross-validation, nor whether the results in Tables 1-3 are from a single held-out test set or aggregated across folds. This ambiguity affects the interpretation of the reported standard deviations and the generalizability of the claims; the authors should specify the exact evaluation protocol used to produce each table.
  4. [2.3 vs 3.1] The architecture of Method 3 as evaluated is unclear. Section 2.3 describes a sequential multi-task pipeline in which a classification network first determines whether FLT is present and conditionally invokes an FLT-specific segmentation network. Section 3.1 then reports that 'the classification step was bypassed in Method 3' because the classifier underperformed. The paper does not state whether the Method 3 results in Tables 1 and 2 correspond to the full described pipeline or to a modified version without classification. If the classification step was removed, the method should be described and named accordingly, and the results should be interpreted as belonging to that modified architecture.
minor comments (5)
  1. [Abstract] The Abstract reports 'Dice Coefficients of ... 0.47 ± 0.25 for FLT,' but Table 1 shows 0.47 for True FLT and 0.38 for overall FLT. Please report the metric consistently and clearly label True FLT versus FLT in the Abstract and Key Points.
  2. [2.2] The text says 'we maintained a constant ratio of FLT cases within each fold,' but the exact ratio and the number of FLT-positive cases per fold are not provided. Please quantify the stratification.
  3. [2.2] The preprocessing description states that images were 'cropped to focus on the foreground,' but the criterion for identifying foreground is not defined. Please specify the cropping procedure.
  4. [Table 3] The Swin-UnetR DCEL row in Table 3 exactly matches the Method 1 testing row in Table 1. Please clarify whether Method 1 is Swin-UnetR trained with DCEL, so that the relationship between the two tables is explicit.
  5. [References] Reference [26] (Kingma and Ba) is missing the year; reference [8] (Isensee et al.) is missing the year and volume/page details. Please complete all bibliographic entries.

Circularity Check

2 steps flagged · score 6.0 of 10

Test-set selection in Section 2.3 makes the reported Dice and Hausdorff improvements over Yao et al. post-selection measurements rather than independent test-set predictions.

  1. fitted input called prediction [Section 2.3, Segmentation Pipeline]
    "We employed two segmentation networks, 3D U-Net (21) and Swin-UnetR (22), and two classification networks, DenseNet-121 and DenseNet-264 (23), selecting them based on their testing performance."

    The paper uses the test-set metrics in Tables 1 and 2 as evidence that its pipelines outperform Yao et al. However, the architectures and classification networks were selected 'based on their testing performance.' Using the test set as a model-selection criterion makes the reported test metrics fitted values: the headline Dice and Hausdorff numbers are, by the paper's own description, the best of the options evaluated on that same test set, so they are not unbiased predictions of generalization.

  2. fitted input called prediction [Section 3.1, Segmentation]
    "It is also worth noting that the classification network did not perform well in identifying FLT cases, with both classification networks achieving only around 50% accuracy. As a result, the classification step was bypassed in Method 3."

    Method 3's advertised test results (e.g., True FLT DC 0.47 ± 0.25) were obtained after the FLT classification step was bypassed. The decision to bypass was made after observing classification performance, so the final pipeline was modified using test outcomes. The reported numbers therefore describe a post-hoc selected variant of Method 3 rather than the performance of the originally specified pipeline operating as designed.

full rationale

This paper is an empirical segmentation benchmark, not a first-principles derivation, so the classic circularity patterns (self-definitional definitions, imported uniqueness theorems, ansatz-smuggling citations, renaming known results) do not apply. The circularity that does apply is the fitted-input-called-prediction pattern. Section 2.3 explicitly states that the two segmentation networks and two classification networks were selected 'based on their testing performance,' and Section 3.1 reports that the FLT classification step was bypassed in Method 3 after observing that the classification networks reached only around 50% accuracy. Consequently, the headline test-set Dice values in Table 1—especially Method 3's True FLT DC of 0.47 versus Yao et al.'s 0.25—are not independent held-out estimates; they are measured after the test set was used to choose architectures and to modify the pipeline. This is a statistical validity problem rather than a proof-theoretic circularity: the measurements themselves are real, but the claimed superiority over Yao et al. is partially an artifact of test-set selection. Additionally, the comparison baseline is not fully controlled: Table 1 lists Yao et al.'s FLT DC as 0.15 while the original paper reported 0.25, and the authors re-implemented Yao et al.'s method with downsampling removed; those are comparability concerns, not circularity. There is no load-bearing self-citation chain. Score 6 reflects that the central empirical claim partially reduces to test-set fitting by the paper's own stated procedure.

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

The central claim rests on the quality of the public ImageTBAD labels, the representativeness of this single-center dataset, the hand-chosen preprocessing and hyperparameters, and the questionable comparability of the Yao et al. baseline. No new entities are introduced.

free parameters (5)
  • Voxel resampling size = 1.5 mm isotropic
    Chosen by hand in preprocessing; affects resolution and memory load, and therefore segmentation performance.
  • Intensity window = -500 to 1000 HU
    Hand-selected to capture relevant anatomy; influences network input.
  • Loss function = Dice Cross Entropy Loss (DCEL)
    Selected after comparing with Generalized Dice Loss on validation; this choice affects all reported results.
  • Training epochs = 50 initial + 20 cascaded
    Hand-set schedule; not justified by convergence analysis beyond Figure 4.
  • Learning rate schedule = 1e-4, decay by 0.1 every 30 epochs
    Standard hand-set schedule; affects convergence and final performance.
assumptions (5)
  • domain assumption Ground-truth segmentation labels by two expert radiologists are accurate.
    The paper relies on these labels as training targets and evaluation reference, but later notes some cases were mislabeled ("some cases were mislabeled, indicating FLT when none was present").
  • domain assumption The ImageTBAD dataset is representative of Type B aortic dissection CTA imaging.
    The models are trained and tested on this single 100-case dataset; generalizability to other scanners or populations is assumed but not demonstrated.
  • standard math Dice and Hausdorff Distance are appropriate and sufficient metrics for this segmentation task.
    Used to rank models and support the superiority claim; no clinical validation of metric thresholds is provided.
  • domain assumption Deep learning models trained on 80 cases can generalize to held-out cases from the same distribution.
    The small training set (80, with only ~54 FLT cases after their count) is unlikely to support optimal parameter adjustment, as the authors themselves note the dataset is insufficient.
  • ad hoc to paper The reimplementation of Yao et al.'s method with modified preprocessing is a valid baseline for comparison.
    The authors removed Yao's downsampling and likely used a different split; comparing against Yao's published numbers rather than the reimplementation's numbers is not a controlled comparison.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Advanced Deep Learning Techniques for Automated Segmentation of Type B Aortic Dissections." pith.science (2026). https://pith.science/paper/JYGJPVIG

@misc{pith2026250622222,
  author       = {Pith},
  title        = {Pith review of: Advanced Deep Learning Techniques for Automated Segmentation of Type B Aortic Dissections},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JYGJPVIG}},
  note         = {Machine review of arXiv:2506.22222}
}
abstract

Purpose: Aortic dissections are life-threatening cardiovascular conditions requiring accurate segmentation of true lumen (TL), false lumen (FL), and false lumen thrombosis (FLT) from CTA images for effective management. Manual segmentation is time-consuming and variable, necessitating automated solutions. Materials and Methods: We developed four deep learning-based pipelines for Type B aortic dissection segmentation: a single-step model, a sequential model, a sequential multi-task model, and an ensemble model, utilizing 3D U-Net and Swin-UnetR architectures. A dataset of 100 retrospective CTA images was split into training (n=80), validation (n=10), and testing (n=10). Performance was assessed using the Dice Coefficient and Hausdorff Distance. Results: Our approach achieved superior segmentation accuracy, with Dice Coefficients of 0.91 $\pm$ 0.07 for TL, 0.88 $\pm$ 0.18 for FL, and 0.47 $\pm$ 0.25 for FLT, outperforming Yao et al. (1), who reported 0.78 $\pm$ 0.20, 0.68 $\pm$ 0.18, and 0.25 $\pm$ 0.31, respectively. Conclusion: The proposed pipelines provide accurate segmentation of TBAD features, enabling derivation of morphological parameters for surveillance and treatment planning

Figures

Figures reproduced from arXiv: 2506.22222 by the authors.

Figure 1
Figure 1. Visualization of a TBAD in a 3D model including FLT(Red), TL(Green), and FL(Yellow), and the corresponding CTA image with axial, [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Segmentation pipeline for Models 1, 2, and 4. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Segmentation pipeline for Model 3. output. If FLT is not detected, the process continues without the FLT segmentation network. The major difference compared to Wobben et al.’s (17) approach lies in the structure of the cascaded model. In their method, they first segmented the whole aorta and then combined this segmentation with the original input for the second network to segment the TL and FL. Finally, they combine… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Training history comparison using DCEL and GDL on Swin-UnetR [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Segmentation result of all proposed models. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 25 canonical work pages

  1. [1]

    ImageTBAD: A 3D Computed Tomography Angiography Image Dataset for Automatic Segmentation of Type-B Aortic Dissection

    Yao Z, Xie W, Zhang J, Dong Y , Qiu H, Yuan H, et al. ImageTBAD: A 3D Computed Tomography Angiography Image Dataset for Automatic Segmentation of Type-B Aortic Dissection. Frontiers in Physiology. 2021;12. Available from: https://www.frontiersin. org/journals/physiology/articles/10.3389/fphys.2021.732711

    work page arXiv 2021

  2. [2]

    The International Registry of Acute Aortic Dissection (IRAD)New Insights Into an Old Disease

    Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The International Registry of Acute Aortic Dissection (IRAD)New Insights Into an Old Disease. JAMA. 2000 02;283(7):897-903. Available from: https://doi.org/10.1001/ jama.283.7.897

    work page

  3. [3]

    Management of acute aortic dissection

    Nienaber CA, Clough RE. Management of acute aortic dissection. Lancet. 2015 Feb;385(9970):800-11

    work page 2015

  4. [4]

    Standardized Protocol to Analyze Computed Tomography Imaging of Type B Aortic Dissections

    Kamman A V , van Herwaarden JA, Orrico M, Nauta FJ, Heijmen RH, Moll FL, et al. Standardized Protocol to Analyze Computed Tomography Imaging of Type B Aortic Dissections. J Endovasc Ther. 2016 Jun;23(3):472-82

    work page 2016

  5. [5]

    Importance of false lumen thrombosis in type B aortic dissection prognosis

    Trimarchi S, Tolenaar JL, Jonker FHW, Murray B, Tsai TT, Eagle KA, et al. Importance of false lumen thrombosis in type B aortic dissection prognosis. The Journal of Thoracic and Cardiovascular Surgery. 2013;145(3):S208-12. Available from: https://www.sciencedirect. com/science/article/pii/S0022522312014894

    work page 2013

  6. [6]

    U-net: Convolutional networks for biomedical image segmentation

    Ronneberger O, Fischer P, Brox T. 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

    work page 2015

  7. [7]

    Inf-Net: Automatic COVID-19 Lung Infection Segmentation From CT Images

    Fan DP, Zhou T, Ji GP, Zhou Y , Chen G, Fu H, et al. Inf-Net: Automatic COVID-19 Lung Infection Segmentation From CT Images. IEEE Transactions on Medical Imaging. 2020;39(8):2626-37

    work page 2020

  8. [8]

    nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation

    Isensee F, Jaeger PF, Kohl SAA, Petersen J, Maier-Hein KH. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat Methods;18(2):203-11. Publisher: Nature Publishing Group. Available from: https://www.nature.com/articles/ s41592-020-01008-z

    work page

Show all 27 references
  1. [9]

    Using DUCK-Net for polyp image segmentation

    Dumitru RG, Peteleaza D, Craciun C. Using DUCK-Net for polyp image segmentation. Sci Rep. 2023 Jun;13(1):9803

    2023

  2. [10]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale; 2021

    Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, et al.. An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale; 2021

    2021

  3. [11]

    TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation; 2021

    Chen J, Lu Y , Yu Q, Luo X, Adeli E, Wang Y , et al.. TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation; 2021

    2021

  4. [12]

    Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation; 2021

    Cao H, Wang Y , Chen J, Jiang D, Zhang X, Tian Q, et al.. Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation; 2021

    2021

  5. [13]

    Efficiently Modeling Long Sequences with Structured State Spaces; 2022

    Gu A, Goel K, R ´e C. Efficiently Modeling Long Sequences with Structured State Spaces; 2022

    2022

  6. [14]

    SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation; 2024

    Xing Z, Ye T, Yang Y , Liu G, Zhu L. SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation; 2024

    2024

  7. [15]

    U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation; 2024

    Ma J, Li F, Wang B. U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation; 2024

    2024

  8. [16]

    Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning

    Cao L, Shi R, Ge Y , Xing L, Zuo P, Jia Y , et al. Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning. Eur J Radiol. 2019 Dec;121:108713

    2019

  9. [17]

    Deep Learning-Based 3D Segmentation of True Lumen, False Lumen, and False Lumen Thrombosis in Type-B Aortic Dissection

    Wobben LD, Codari M, Mistelbauer G, Pepe A, Higashigaito K, Hahn LD, et al. Deep Learning-Based 3D Segmentation of True Lumen, False Lumen, and False Lumen Thrombosis in Type-B Aortic Dissection. In: 2021 43rd Annual International Conference of the IEEE Engineering in Medicine...

    2021

  10. [18]

    Springer; 2015. p. 234-41

    2015

  11. [19]

    Multi-stage learning for segmentation of aortic dissections using a prior aortic anatomy simplification

    Chen D, Zhang X, Mei Y , Liao F, Xu H, Li Z, et al. Multi-stage learning for segmentation of aortic dissections using a prior aortic anatomy simplification. Medical Image Analysis. 2021;69:101931. Available from: https://www.sciencedirect.com/science/article/ pii/S1361841520302954

    2021

  12. [20]

    ADSeg: A flap-attention-based deep learning approach for aortic dissection segmentation

    Xiang D, Qi J, Wen Y , Zhao H, Zhang X, Qin J, et al. ADSeg: A flap-attention-based deep learning approach for aortic dissection segmentation. PATTER. 2023;4(5). Publisher: Elsevier. Available from: https://www.cell.com/patterns/abstract/S2666- 3899(23)00067-3

    2023

  13. [21]

    CT-based True- and False-Lumen Segmentation in Type B Aortic Dissection Using Machine Learning

    Hahn LD, Mistelbauer G, Higashigaito K, Koci M, Willemink MJ, Sailer AM, et al. CT-based True- and False-Lumen Segmentation in Type B Aortic Dissection Using Machine Learning. Radiol Cardiothorac Imaging. 2020;2(3):e190179

    2020

  14. [22]

    3D U-Net: Learning Dense V olumetric Segmentation from Sparse Annotation; 2016

    ¨Ozg¨un C ¸ ic ¸ek, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O. 3D U-Net: Learning Dense V olumetric Segmentation from Sparse Annotation; 2016

    2016

  15. [23]

    Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images; 2022

    Hatamizadeh A, Nath V , Tang Y , Yang D, Roth H, Xu D. Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images; 2022

    2022

  16. [24]

    Densely Connected Convolutional Networks; 2018

    Huang G, Liu Z, van der Maaten L, Weinberger KQ. Densely Connected Convolutional Networks; 2018

    2018

  17. [25]

    In: Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations

    Sudre CH, Li W, Vercauteren T, Ourselin S, Jorge Cardoso M. In: Generalised Dice Overlap as a Deep Learning Loss Function for Highly Unbalanced Segmentations. Springer International Publishing; 2017. p. 240–248. Available from: http://dx.doi.org/10.1007/978- 3-319-67558-9_28 . 8

    2017 doi

  18. [26]

    PyTorch: An Imperative Style, High-Performance Deep Learning Library; 2019

    Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, et al.. PyTorch: An Imperative Style, High-Performance Deep Learning Library; 2019

    2019

  19. [27]

    Adam: A Method for Stochastic Optimization

    Kingma DP, Ba J. Adam: A Method for Stochastic Optimization. arXiv;. Available from: http://arxiv.org/abs/1412.6980. 9

    arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.