Pith. sign in

REVIEW 3 major objections 7 minor 44 references

Intuitive Axial Augmentation Using Polar-Sine-Based Piecewise Distortion for Medical Slice-Wise Segmentation

T0 review · 3 major / 7 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a polar-sine-based piecewise affine warp of axial CT and MR slices simulates patient posture change during scans and, without extra data, raises segmentation accuracy across multiple model families.

desk verdict New distortion-based augmentation with a plausible clinical motivation, but the empirical case is undercut by validation-set parameter selection and rotation-only baselines. read the letter →

arxiv 2412.03352 v2 pith:SOZN7XKK submitted 2024-12-04 cs.CV cs.AI

Yiqin Zhang , Qingkui Chen , Chen Huang , Zhengjie Zhang , Meiling Chen , Zhibing Fu This is my paper · ORCID
classification cs.CVcs.AI
keywords medicalimageaugmentationaxialslicesegmentationpiecewiseaffinedistortionpolarcoordinatesDICOMmetadatasimilarity-guidedparametersearchCTMRI
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes PSBPD, a plug-and-play data augmentation that rewrites each axial CT or MR slice by displacing a grid of control points along sine-shaped rays in polar coordinates and then warping the slice with a piecewise affine transform. The intended effect is to generate new slices that look like reconstructions of the same patient under slight posture changes while keeping the relative positions of organs intact. The paper claims that on a seven-year gastric cancer CT dataset, on the public CT-ORG dataset, and on BraTS, this augmentation raises Dice for convolutional, transformer, and state-space segmentation models without requiring extra annotated data. It also introduces two non-neural helpers: a DICOM-metadata geometric step that masks out the scan table, and a similarity-guided check using SIFT and ORB to select safe distortion intensities. The motivation is clinical trust: unlike erasing or swapping anatomy, these samples resemble real re-scans, so clinicians can see why training on them helps.

What carries the argument

The central object is the polar-sine control-point displacement: a regular grid of control points is converted to polar coordinates around the slice center, and each point's angular coordinate is shifted by a sine function of its radius, $\vartheta_{\mathrm{new}} = \vartheta_{\mathrm{map}} + \frac{\pi}{8} a \sin\!\left(\frac{r_{\mathrm{map}}}{\delta} 2 f \pi\right)$, with per-sample random $a$ and $f$ controlled by $A$ and $\omega$. Because radial distances from the center are preserved, regions near the body surface move more than deep structures, which the authors argue matches how posture changes deform the body. The displaced grid is triangulated and each triangle is warped affinely to produce the augmented slice, with the grid density $\delta$ controlling fidelity. Two supporting mechanisms carry the robustness claims: a geometric model built from DICOM fields (reconstruction diameter, table height, pixel spacing, image position) masks out the scan table, and a similarity-guided search correlates SIFT/ORB matching counts with downstream accuracy to choose $A$ and $\omega$ without neural-network training.

What would settle it

A direct test would compare PSBPD-warped slices against true repeat scans of the same patient: if the SIFT/ORB similarity between a warped slice and the real re-scan is not higher than that between a rotation-only augmented slice and the real re-scan, the physical-realism premise is not supported. A complementary check is to train with PSBPD and with an equally strong but anatomically random warp; if accuracy is matched, the gain is a regularizer rather than realism.

Watch

Extended reading notes

Core claim

The discovery the authors are trying to establish is that posture uncertainty during a radiologic scan is a usable source of training variation, and that a polar-sine piecewise affine distortion can synthesize it cheaply. In the displacement rule, each control point keeps its radius but its polar angle becomes $\vartheta_{\mathrm{new}} = \vartheta_{\mathrm{map}} + \frac{\pi}{8} a \sin\!\left(\frac{r_{\mathrm{map}}}{\delta} 2 f \pi\right)$, with amplitude $a$ and frequency $f$ drawn randomly per sample from ranges set by $A$ and $\omega$. The displaced grid is triangulated and applied as a piecewise affine warp to the whole slice, so the paper's evidence is that this warp raises mDice on the private gastric dataset for seven model families and raises class-average Dice on CT-ORG and BraTS over rotation-only augmentation, while preserving enough feature-matching similarity that clinicians would still recognize the anatomy. The authors also claim that the similarity-guided parameter range found without training predicts the intensity range that helps neural networks most, which would spare users expensive end-to-end tuning.

Load-bearing premise

The load-bearing premise is that warping an axial slice along sine-shaped rays produces images that resemble real re-scans of the same patient under posture change; the paper does not test this against actual same-subject multi-posture scans.

Editorial extensions

If this is right

  • As a plug-in preprocessing step before normalization, PSBPD improves mDice for ResNet50, ConvNeXt, SegNeXt, MAE, Poolformer, Segformer, and Swin Transformer V2 on the gastric cancer dataset, and for MedNext and SwinUMamba on public CT and MR datasets.
  • The method works with as little as one annotated axial slice per CT scan, so it addresses the annotation-cost bottleneck in medical imaging.
  • The similarity-guided search identifies a usable intensity range (around $A=6$, $\omega=2$) without training, and the accuracy surface is smooth, so the method is forgiving to parameter choice.
  • The scan-table removal step is optional: it helps most models but hurts MAE, because masked reconstruction benefits from the extra pixels.
  • Since the augmentation is model-agnostic and the helper modules are non-neural, it can be added to existing training pipelines with minimal computational overhead, measured at sub-millisecond to a few milliseconds per slice.

Reading between the lines

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

  • A direct validity check the paper leaves open is whether a PSBPD-warped slice actually resembles a real re-scan of the same patient under posture change; comparing warped slices with true repeat scans would settle the physical claim and could calibrate the intensity parameters.
  • Part of the accuracy gain may come from the warp acting as a strong geometric regularizer rather than from anatomical realism; comparing PSBPD against an equally strong but anatomically random warp, matched in displacement statistics, would separate these explanations.
  • The per-slice independent warping ignores slice-to-slice coherence, so sampling one distortion field and applying it consistently across neighboring axial slices is a natural 3D extension that could improve volumetric segmentation.
  • The similarity-guided tuning recipe is portable: any geometric augmentation whose pre/post images can be feature-matched could reuse the SIFT/ORB screening to find a safe intensity range, which would make the tuning step cheaper for other augmentations.
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 / 7 minor

Summary. The paper proposes Polar-Sine-Based Piecewise Distortion (PSBPD), an augmentation method for axial medical slices that distorts a slice in polar coordinates using a sinusoidal ray mapping, motivated by posture uncertainty during CT/MR scanning. It also introduces a DICOM-metadata-driven scan table removal module and a SIFT/ORB similarity-guided hyperparameter search for the distortion amplitude and frequency. Experiments are reported on the CT-ORG and BraTS public datasets and on a private gastric cancer CT dataset, across convolution-, transformer-, and mamba-based segmentation models, with Dice/Recall/Precision improvements claimed relative to rotation-only or no augmentation.

Significance. If the empirical claims are reliable, PSBPD is a plug-and-play, model-agnostic augmentation that improves slice-wise segmentation in low-annotation settings and has an interpretability rationale that clinicians may find more acceptable than black-box augmentations. The paper has concrete strengths: it provides a public code repository, evaluates on two public benchmarks plus a clinically collected private dataset, and introduces non-adaptive modules with a computational overhead analysis showing that preprocessing is not a training bottleneck. The main limitation is that the evaluation protocol does not clearly separate hyperparameter selection from test reporting and does not provide seed-averaged uncertainty, so the size and even the direction of the reported gains are not yet firmly established.

major comments (3)
  1. [Section 3.4, Algorithm 1; Tables 4-6] The central claim that the method 'improves accuracy' is not yet substantiated because the evaluation does not state a clean train/validation/test split. Algorithm 1 selects the hyperparameters (A, omega) as k* = argmin_k L_k on D_val, and the manuscript never says explicitly that the Dice values in Tables 4-6 are measured on a test partition that was not used to choose k*. If the reported numbers come from the same D_val used for selection, the improvements are inflated by selection bias. The problem is compounded by the absence of any multiple-seed results or standard deviations. For example, in Table 4 the bladder Dice jumps from 3.21 to 48.55 for MedNext, a change far larger than one would expect from an augmentation effect without seed-averaged error bars. The authors should report an explicit data split, a fixed rule for choosing A and omega, and mean +/- std over at least three random seeds.
  2. [Section 4.3 and Section 4.5, Fig. 8 and Fig. 11] The similarity-guided parameter search is not validated as a predictor of segmentation accuracy. Section 4.3 reports Pearson correlations of -0.625 and -0.618 between SIFT/ORB matching counts and accuracy and states that similarity begins to plummet at a ~= 3, f ~= 1.5. However, Section 4.5 states that accuracy only starts to decline at A > 7, omega > 3. These two thresholds are not consistent, so it is unclear that the similarity metric identifies the same operating region as the actual accuracy landscape. Moreover, the correlation is computed on the same parameter grid that Algorithm 1 uses to select hyperparameters, so it does not demonstrate predictive value for a new dataset. A concrete fix is to evaluate the similarity-accuracy correlation on a separate grid or a holdout dataset and report the rank correlation between predicted and actual optimal parameters.
  3. [Section 3.2.2, Eqs. (2)-(4); Section 1] The premise that sinusoidal polar-coordinate distortion simulates genuine posture variation is load-bearing for the paper's interpretability and clinical-acceptance claims, but it is never directly validated. The paper shows that the distorted images are continuous and visually plausible, and that downstream accuracy often improves, but it does not compare the augmented samples to actual repeated scans of the same subject under different postures, nor does it provide a clinical/radiological evaluation of the generated slices. Without such evidence, the claim that the method 'simulates uncertain postures' is an assumption, not an established property. The authors should either provide a validation experiment with paired same-subject scans or soften the interpretability claim accordingly.
minor comments (7)
  1. [Section 3.2.2, Eq. (3)] The text says 'with theta as the pole' where it should presumably refer to the pole coordinate Theta; Eq. (3) also uses inconsistent notation for array height and width (S_h/S_w vs. H/W).
  2. [Section 3.4, Algorithm 1] The similarity threshold t ~= 90% is introduced without justification or a sensitivity analysis; it is unclear how the size of K10% changes with t and how sensitive the selected hyperparameters are to this choice.
  3. [Table 6, Poolformer row] The text says that most frameworks improve 'excluding the Poolformer model,' but Table 6 shows that Poolformer's 'Ours' column is worse than both 'rotate +/-180' and 'w/o Table Removal' on mIoU and mDice; this exception should be acknowledged explicitly and discussed.
  4. [Section 4.3, Fig. 8] The caption and text refer to 'SIRF' in one place, which appears to be a typo for SIFT; please make the feature-descriptor names consistent.
  5. [Section 4.2] The statement that 'there are currently two main architectures, convolutional structure and Transformer structure' is outdated given that the paper itself evaluates SwinUMamba, a state-space model; the sentence should be revised to reflect the broader set of modern architectures.
  6. [Section 3.3, Eq. (12)] The formatting of the conditional 'if ||p - theta2|| > lambda2: p = epsilon' is unclear; please use proper subscripts and explain what happens to pixels outside the valid mask.
  7. [References and Section 4.4] The text mentions SA-Med2D in Section 4.4 but the dataset description in Section 4.1 lists only CT-ORG and BraTS; please clarify the relationship between SA-Med2D and the CT-ORG results reported in Table 4.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the accuracy gains are empirical, the distortion is geometrically defined, and no load-bearing self-citations or in-construction predictions were found.

full rationale

The paper's central claim is that PSBPD improves segmentation accuracy; this is supported by experiments on external benchmarks (CT-ORG, BraTS) and a private dataset across seven architectures. The distortion is defined geometrically via polar-coordinate sine warp (Eqs. 1-9), and the reported accuracy numbers are measured after training, not algebraically derived from the method's definition. Hyperparameter selection in Algorithm 1 uses validation loss to choose A and omega, which is standard model selection; even if the same validation split were reused in the tables, that would be selection bias rather than circularity because the reported Dice is not forced by the argmin. The similarity-guide claim is validated empirically by correlating SIFT/ORB match counts with measured segmentation accuracy (Fig. 8), not by assuming the correlation, so the proxy is independently evidence-backed. The paper contains no load-bearing self-citations: it cites its own GitHub for code and a preprint, but no prior theoretical result is imported to justify the method. The premise that sinusoidal distortion mimics patient posture is a clinical assumption, not a result derived from the target outcome; whether it holds is testable but not circular. The derivation chain is therefore self-contained, and the empirical claims are externally falsifiable.

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

The augmentation has two tunable hyperparameters (A, omega) selected by validation-based search; delta and the similarity threshold are hand-set. No new physical or conceptual entities are introduced; the distortion is a geometric transformation of existing pixels. The main empirical claims are tested on external datasets but do not include multiple seeds or error bars.

free parameters (4)
  • A (distortion amplitude) = A about 6 (best on validation, Fig. 11)
    Controls the angular distortion magnitude; selected by grid search over [0.25, 12] using validation accuracy.
  • omega (angular frequency) = omega about 2 (best on validation, Fig. 11)
    Controls sine frequency of the distortion; selected jointly with A via validation accuracy.
  • delta (control point grid density) = delta >= 16 recommended
    Number of control points per dimension; affects fidelity and computational cost; chosen by hand, not tuned.
  • Similarity threshold t = t about 90%
    Defines the K10% parameter spectrum in Algorithm 1; chosen by hand, not derived.
assumptions (4)
  • domain assumption Sinusoidal polar-coordinate distortion of an axial slice generates anatomically plausible images resembling real patient posture variation.
    Core premise of the augmentation mechanism (Section 3.2). Not validated directly against re-scans of the same patient; only supported indirectly by downstream accuracy gains.
  • domain assumption SIFT/ORB feature-match count is a valid proxy for the learnability of augmented samples.
    Used to define the K10% parameter set in Algorithm 1; the paper provides only in-sample correlation coefficients (rho about -0.62), not a predictive validation.
  • standard math Delaunay triangulation and piecewise affine mapping preserve the continuity of anatomical structures.
    Standard property of piecewise-affine warping used in Eqs. (8) and (9); reasonable but not proven for tissue boundaries.
  • domain assumption DICOM metadata values (Table Height, Image Position, Pixel Spacing, Reconstruction Diameter) are accurate enough to geometrically locate the CT table.
    Scan Table Removal (Section 3.3) depends on these tags; most public NIFTI datasets do not provide them, limiting scope.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Intuitive Axial Augmentation Using Polar-Sine-Based Piecewise Distortion for Medical Slice-Wise Segmentation." pith.science (2026). https://pith.science/paper/SOZN7XKK

@misc{pith2026241203352,
  author       = {Pith},
  title        = {Pith review of: Intuitive Axial Augmentation Using Polar-Sine-Based Piecewise Distortion for Medical Slice-Wise Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SOZN7XKK}},
  note         = {Machine review of arXiv:2412.03352}
}
read the original abstract

Most data-driven models for medical image analysis rely on universal augmentations to improve accuracy. Experimental evidence has confirmed their effectiveness, but the unclear mechanism underlying them poses a barrier to the widespread acceptance and trust in such methods within the medical community. We revisit and acknowledge the unique characteristics of medical images apart from traditional digital images, and consequently, proposed a medical-specific augmentation algorithm that is more elastic and aligns well with radiology scan procedure. The method performs piecewise affine with sinusoidal distorted ray according to radius on polar coordinates, thus simulating uncertain postures of human lying flat on the scanning table. Our method could generate human visceral distribution without affecting the fundamental relative position on axial plane. Two non-adaptive algorithms, namely Meta-based Scan Table Removal and Similarity-Guided Parameter Search, are introduced to bolster robustness of our augmentation method. In contrast to other methodologies, our method is highlighted for its intuitive design and ease of understanding for medical professionals, thereby enhancing its applicability in clinical scenarios. Experiments show our method improves accuracy with two modality across multiple famous segmentation frameworks without requiring more data samples. Our preview code is available in: https://github.com/MGAMZ/PSBPD.

Figures

Figures reproduced from arXiv: 2412.03352 by the authors.

Figure 1
Figure 1. Overview of the proposed method. (a) shows common frameworks tend to utilize universal augment method on CT dataset, without considering the effect of posture uncertainty. (b) is a pair of slices from the same subject with different posture, leading to variance on slices. Our augmentation method is able to simulate this malleable variance, thus generating more samples for neural networks. standard data analysis proc… view at source ↗
Figure 2
Figure 2. Overview of data preprocess. Symbol definitions are available in [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Generated control point map with 𝐴 = 1, 𝜔 = 1. The points of the same color correspond in two sub-figs. The positions of the converted control points have changed, with the points closer to the far end undergoing more drastic changes. However, the relative positional relationships between the control points remain unchanged, and points of similar colors still cluster together. map only performs a fixed transformatio… view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Distorted reconstructed slices with different parameters. Ψ = 1 to better compare the effect. Larger 𝐴 and 𝜔 will lead to extreme distortion effect, but still remains continuity of adjacent areas. The scanning table is also distorted, this may introduce unnecessary noi…
Figure 5
Figure 5. Figure 5: Geometric Positional Modeling used in scan table removal. The definition of DICOM symbols are available in [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Scan table removal. The subject’s body in the left subfigure presents a continuous region resembling an ellipse, and the distorted rectangle below it is the scan table. The metal shell frame of the table absorbs X-rays strongly, so its shape is clearly shown in the rec…
Figure 7
Figure 7. Figure 7: Removal effectiveness on distorted slice. The left subfigure shows that after distortion, the curve of the Scan Table presents a very messy shape, which can introduce significant noise in the feature extraction process. The right one is a slice after the Table Removal …
Figure 8
Figure 8. Figure 8: Similarity guide results. Relations between the Count of Feature Points Matched Successfully and Actual Segmentation Accuracy ( [PITH_FULL_IMAGE:figures/full_fig_p016_8.png]
Figure 9
Figure 9. Figure 9: Similarity distribution around labelled slices. The vertical axis in the first and second subfigure correspond to SIFT and ORB value. The number of similarity detection points shows an upward trend when △ → 𝑢, 𝑢 ∉ 𝑠𝑡𝑜𝑚𝑎𝑐ℎ ≈ [−20, 20]. The similarity results indicate a …
Figure 10
Figure 10. Figure 10: Segmentation metrics. In the left subfigure, the orange dots represent actual data points, and the gray dashed line indicates the estimated PR curve. A curve closer to upper right corner represents better. The right subfigure illustrates the counterbalance between acc…
Figure 11
Figure 11. Figure 11: Parameter search on 𝐴 and 𝜔. Too small or too large 𝐴 and 𝜔 will lead to reduced accuracy, and their distribution is relatively moderate, reflecting the high stability of the algorithm under different parameters. The most effective setting seems to be around 𝐴 = 6, 𝜔 …
Figure 12
Figure 12. Figure 12: Computational Overhead Analysis. Distinct shapes are utilized to denote varying sizes 𝑛 of the source pixel matrices. Different colors are assigned to indicate the selection of post-sampling implementation options. Additionally, the size of the shapes corresponds to t…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 23 canonical work pages

  1. [1]

    Artifacts among Cone Beam Computed Tomography Images of Patients of Department of Oral Medicine and Radiology in a Tertiary Care Centre: A Descriptive Cross-sectional Study

    Bali, H., Luitel, A., Upadhyaya, C., 2023. Artifacts among Cone Beam Computed Tomography Images of Patients of Department of Oral Medicine and Radiology in a Tertiary Care Centre: A Descriptive Cross-sectional Study. JNMA J Nepal Med Assoc 61, 18–22. doi:10.31729/jnma.7949

    work page doi:10.31729/jnma.7949 2023

  2. [2]

    2d object recognition: a comparative analysis of sift, surf and orb feature descriptors

    Bansal, M., Kumar, M., Kumar, M., 2021. 2d object recognition: a comparative analysis of sift, surf and orb feature descriptors. Multimedia Tools and Applications 80, 18839–18857. URL: https://doi.org/10.1007/s11042-021-10646-0, doi:10.1007/ s11042-021-10646-0

    work page doi:10.1007/s11042-021-10646-0 2021

  3. [3]

    Deepconvolutionalneuralnetworksforbrainimage analysis on magnetic resonance imaging: a review

    Bernal,J.,Kushibar,K.,Asfaw,D.S.,Valverde,S.,Oliver,A.,Martí,R.,Lladó,X.,2019. Deepconvolutionalneuralnetworksforbrainimage analysis on magnetic resonance imaging: a review. Artificial Intelligence in Medicine 95, 64–81. URL:https://www.sciencedirect. com/science/article/pii/S0933365716305206, doi:https://doi.org/10.1016/j.artmed.2018.08.008

    work page doi:10.1016/j.artmed.2018.08.008 2019

  4. [4]

    Measuring image similarity: an overview of some useful applications

    Chalom, E., Asa, E., Biton, E., 2013. Measuring image similarity: an overview of some useful applications. IEEE Instrumentation & Measurement Magazine 16, 24–28. doi:10.1109/MIM.2013.6417053

    work page arXiv 2013

  5. [5]

    Chen,Y.W.,Shih,C.T.,Lin,H.H.,Chuang,K.S.,2016.Physicalmodel-basedcontrastenhancementofcomputedtomographyimages:Contrast enhancementofcomputedtomography,in:2016IEEE16thInternationalConferenceonBioinformaticsandBioengineering(BIBE),pp.238–

    work page 2016

  6. [6]

    A review of medical image data augmentation techniques for deep learning applications

    Chlap, P., Min, H., Vandenberg, N., Dowling, J., Holloway, L., Haworth, A., 2021. A review of medical image data augmentation techniques for deep learning applications. Journal of medical imaging and radiation oncology 65, 545–563. doi:10.1111/1754-9485.13261. pMID: 34145766

    arXiv 2021

  7. [7]

    Longitudinal image registration with temporally-dependent image similarity measure

    Csapo, I., Davis, B., Shi, Y., Sanchez, M., Styner, M., Niethammer, M., 2013. Longitudinal image registration with temporally-dependent image similarity measure. IEEE Transactions on Medical Imaging 32, 1939–1951. doi:10.1109/TMI.2013.2269814

    arXiv 2013

  8. [8]

    Areviewondelaunaytriangulationwithapplicationoncomputervision

    Dinas,S.,Bañón,J.,2014. Areviewondelaunaytriangulationwithapplicationoncomputervision. IJCSE-InternationalJournalofComputer Science and Engineering 3, 9–18

    work page 2014

Show all 44 references
  1. [9]

    Overcomemedicalimagedatascarcitybydataaugmentationtechniques:Areview,in:2022 International Conference on Microelectronics (ICM), pp

    ElJiani,L.,ElFilali,S.,Benlahmer,E.H.,2022. Overcomemedicalimagedatascarcitybydataaugmentationtechniques:Areview,in:2022 International Conference on Microelectronics (ICM), pp. 21–24. doi:10.1109/ICM56065.2022.10005544

    2022

  2. [10]

    Data augmentation for medical imaging: A systematic literature review

    Garcea, F., Serra, A., Lamberti, F., Morra, L., 2023. Data augmentation for medical imaging: A systematic literature review. Computers in Biology and Medicine 152, 106391. URL: https://www.sciencedirect.com/science/article/pii/S001048252201099X, doi:https://doi.org/10.1016/j.c...

    2023

  3. [11]

    Using dicommetadataforradiologicalimageseriescategorization:afeasibilitystudyonlargeclinicalbrainmridatasets

    Gauriau, R., Bridge, C., Chen, L., Kitamura, F., Tenenholtz, N.A., Kirsch, J.E., Andriole, K.P., Michalski, M.H., Bizzo, B.C., 2020. Using dicommetadataforradiologicalimageseriescategorization:afeasibilitystudyonlargeclinicalbrainmridatasets. JournalofDigitalImaging 33, 747–76...

    2020 doi

  4. [12]

    Two-stage unsupervised learning method for affine and deformable medical image registration, in: 2019 IEEE International Conference on Image Processing (ICIP), pp

    Gu, D., Liu, G., Tian, J., Zhan, Q., 2019. Two-stage unsupervised learning method for affine and deformable medical image registration, in: 2019 IEEE International Conference on Image Processing (ICIP), pp. 1332–1336. doi:10.1109/ICIP.2019.8803794

    2019

  5. [13]

    The virtues of interpretable medical artificial intelligence

    Hatherley, J., Sparrow, R., Howard, M., 2022. The virtues of interpretable medical artificial intelligence. Cambridge Quarterly of Healthcare Ethics , 1–10doi:10.1017/S0963180122000305

    2022 doi

  6. [14]

    Review on self supervised learning in medical image analysis, in: 2023 IEEE 7th Conference on Information and Communication Technology (CICT), pp

    Kumari, N., Agrawal, S., 2023. Review on self supervised learning in medical image analysis, in: 2023 IEEE 7th Conference on Information and Communication Technology (CICT), pp. 1–6. doi:10.1109/CICT59886.2023.10455714

    2023

  7. [15]

    Casper:Conventional ct database augmentation using deep learning based spectral ct images generation, in: 2020 15th IEEE International Conference on Signal Processing (ICSP), pp

    Lartaud,P.J.,Rouchaud,A.,Dessouky,R.,Vlachomitrou,A.S.,Rouet,J.M.,Nempont,O.,Boussel,L.,Douek,P.,2020. Casper:Conventional ct database augmentation using deep learning based spectral ct images generation, in: 2020 15th IEEE International Conference on Signal Processing (ICSP),...

    2020

  8. [16]

    Swin-umamba: Mamba-based unet with imagenet-based pretraining

    Liu, J., Yang, H., Zhou, H.Y., Xi, Y., Yu, L., Yu, Y., Liang, Y., Shi, G., Zhang, S., Zheng, H., Wang, S., 2024. Swin-umamba: Mamba-based unet with imagenet-based pretraining. URL:https://arxiv.org/abs/2402.03302, arXiv:2402.03302

    2024 arXiv

  9. [17]

    Mixing data augmentation with preserving foreground regions in medical image segmentation, in: 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), pp

    Liu, X., Ono, K., Bise, R., 2023. Mixing data augmentation with preserving foreground regions in medical image segmentation, in: 2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI), pp. 1–5. doi:10.1109/ISBI53787.2023.10230495

    2023

  10. [18]

    Object recognition from local scale-invariant features, in: Proceedings of the Seventh IEEE International Conference on Computer Vision, pp

    Lowe, D., 1999. Object recognition from local scale-invariant features, in: Proceedings of the Seventh IEEE International Conference on Computer Vision, pp. 1150–1157 vol.2. doi:10.1109/ICCV.1999.790410

    1999

  11. [19]

    Distinctive image features from scale-invariant keypoints

    Lowe, D.G., 2004. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 60, 91–110. URL: https://doi.org/10.1023/B:VISI.0000029664.99615.94, doi:10.1023/B:VISI.0000029664.99615.94

    2004

  12. [20]

    Computed tomography imaging artifacts in the head and neck region: Pitfalls and solutions

    Marshall, E.L., Ginat, D.T., Sammet, S., 2022. Computed tomography imaging artifacts in the head and neck region: Pitfalls and solutions. Neuroimaging Clinics of North America 32, 271–277. URL:https://www.sciencedirect.com/science/article/pii/ S1052514922000016, doi:https://do...

    2022 doi

  13. [21]

    pydicom/pydicom: pydicom v2.4.4.https://doi.org/10.5281/zenodo.10385252

    Mason, D., Scaramallion, Mrbean-bremen, Rhaxton, Suever, J., Vanessasaurus, Orfanos, D.P., Lemaitre, G., Panchal, A., Rothberg, A., Herrmann, M.D., Massich, J., Kerns, J., van Golen, K., Bridge, C., Robitaille, T., Biggs, S., Moloney, Shun-Shin, M., Clauss, C., 2023. pydicom/p...

    2023 doi

  14. [22]

    The multimodal brain tumor image segmentation benchmark (brats)

    Menze, B.H., Jakab, A., Bauer, S., Kalpathy-Cramer, J., Farahani, K., Kirby, J., Burren, Y., Porz, N., Slotboom, J., Wiest, R., Lanczi, L., Gerstner, E., Weber, M.A., Arbel, T., Avants, B.B., Ayache, N., Buendia, P., Collins, D.L., Cordier, N., Corso, J.J., Criminisi, A., Das,...

    2015

  15. [23]

    A ct database for research, development and education: Concept and potential

    Messmer, P., Matthews, F., Jacob, A.L., Kikinis, R., Regazzoni, P., Noser, H., 2007. A ct database for research, development and education: Concept and potential. Journal of Digital Imaging 20, 17–22. URL: https://doi.org/10.1007/s10278-006-0771-9, doi:10.1007/s10278-006-0771-9

    2007 doi

  16. [24]

    Prospects of structural similarity index for medical image analysis

    Mudeng, V., Kim, M., Choe, S.w., 2022. Prospects of structural similarity index for medical image analysis. Applied Sciences 12. URL: https://www.mdpi.com/2076-3417/12/8/3754, doi:10.3390/app12083754

    2022 doi

  17. [25]

    sciencedirect.com/science/article/pii/S2590005622000911, doi:https://doi.org/10.1016/j.array.2022.100258

    Mumuni,A.,Mumuni,F.,2022.Dataaugmentation:Acomprehensivesurveyofmodernapproaches.Array16,100258.URL: https://www. sciencedirect.com/science/article/pii/S2590005622000911, doi:https://doi.org/10.1016/j.array.2022.100258

    2022

  18. [26]

    Applications of self-supervised learning to biomedical signals: A survey

    Pup, F.D., Atzori, M., 2023. Applications of self-supervised learning to biomedical signals: A survey. IEEE Access 11, 144180–144203. doi:10.1109/ACCESS.2023.3344531

    2023

  19. [27]

    Efficient unstructured mesh generation by means of delaunay triangulation and bowyer-watson algorithm

    Rebay, S., 1993. Efficient unstructured mesh generation by means of delaunay triangulation and bowyer-watson algorithm. Journal of Computational Physics 106, 125–138. URL: https://www.sciencedirect.com/science/article/pii/S0021999183710971, doi:https://doi.org/10.1006/jcph.1993.1097

    1993

  20. [28]

    Explainable ai (xai): Explained, in: 2023 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), pp

    Reddy, G.P., Kumar, Y.V.P., 2023. Explainable ai (xai): Explained, in: 2023 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), pp. 1–6. doi:10.1109/eStream59056.2023.10134984

    2023

  21. [29]

    Ct-org, a new dataset for multiple organ segmentation in computed tomography

    Rister, B., Yi, D., Shivakumar, K., Nobashi, T., Rubin, D.L., 2020. Ct-org, a new dataset for multiple organ segmentation in computed tomography. Scientific Data 7, 381. URL:https://doi.org/10.1038/s41597-020-00715-8, doi:10.1038/s41597-020-00715-8

    2020 doi

  22. [30]

    Roy, S., Koehler, G., Ulrich, C., Baumgartner, M., Petersen, J., Isensee, F., Jäger, P.F., Maier-Hein, K.H., 2023. Mednext: Transformer- driven scaling of convnets for medical image segmentation, in: Greenspan, H., Madabhushi, A., Mousavi, P., Salcudean, S., Duncan, J., Syeda-...

    2023

  23. [31]

    Orb: An efficient alternative to sift or surf, in: 2011 International Conference on Computer Vision, pp

    Rublee, E., Rabaud, V., Konolige, K., Bradski, G., 2011. Orb: An efficient alternative to sift or surf, in: 2011 International Conference on Computer Vision, pp. 2564–2571. doi:10.1109/ICCV.2011.6126544

    2011

  24. [32]

    Transparencyofdeepneuralnetworksformedicalimageanalysis:Areview of interpretability methods

    Salahuddin,Z.,Woodruff,H.C.,Chatterjee,A.,Lambin,P.,2021. Transparencyofdeepneuralnetworksformedicalimageanalysis:Areview of interpretability methods. URL:https://arxiv.org/abs/2111.02398, arXiv:2111.02398

    2021 arXiv

  25. [33]

    Differentiableimagedataaugmentationanditsapplications:Asurvey

    Shi,J.,Ghazzai,H.,Massoud,Y.,2024. Differentiableimagedataaugmentationanditsapplications:Asurvey. IEEETransactionsonPattern Analysis and Machine Intelligence 46, 1148–1164. doi:10.1109/TPAMI.2023.3330862

    2024

  26. [34]

    Hsmix:Hardandsoftmixingdataaugmentationformedicalimagesegmentation

    Sun,D.,Dornaika,F.,Barrena,N.,2025. Hsmix:Hardandsoftmixingdataaugmentationformedicalimagesegmentation. InformationFusion 115, 102741. URL: https://www.sciencedirect.com/science/article/pii/S1566253524005190, doi:https://doi.org/10. 1016/j.inffus.2024.102741

    2025

  27. [35]

    (Eds.), 2012

    Tack, D., Kalra, M.K., Gevenois, P.A. (Eds.), 2012. Image Quality in CT: Challenges and Perspectives. Springer Berlin Heidelberg, Berlin, Heidelberg. pp. 81–100. URL:https://doi.org/10.1007/174_2011_482, doi:10.1007/174_2011_482

    2012 doi

  28. [36]

    Explainable artificial intelligence (xai) in deep learning-based medical image analysis

    van der Velden, B.H., Kuijf, H.J., Gilhuijs, K.G., Viergever, M.A., 2022. Explainable artificial intelligence (xai) in deep learning-based medical image analysis. Medical Image Analysis 79, 102470. URL: https://www.sciencedirect.com/science/article/pii/ S1361841522001177, doi:...

    2022

  29. [37]

    SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python

    Virtanen, P., Gommers, R., Oliphant, T.E., Haberland, M., Reddy, T., Cournapeau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J., van der Walt, S.J., Brett, M., Wilson, J., Millman, K.J., Mayorov, N., Nelson, A.R.J., Jones, E., Kern, R., Larson, E., Carey, C.J., Pola...

    2020

  30. [38]

    Ethicsofcollectingandusinghealthcaredata

    Wade,D.,2007. Ethicsofcollectingandusinghealthcaredata. BMJ(Clinicalresearched.)334,1330–1331. URL: https://doi.org/10. 1136/bmj.39247.679329.80, doi:10.1136/bmj.39247.679329.80. publisher: BMJ Publishing Group

    2007

  31. [39]

    Xedgeconv: Leveraging graph convolutions for efficient, permutation- and rotation-invariant dense 3d medical image segmentation, in: Bekkers, E., Wolterink, J.M., Aviles-Rivero, A

    Weihsbach, C., Hansen, L., Heinrich, M., 2022. Xedgeconv: Leveraging graph convolutions for efficient, permutation- and rotation-invariant dense 3d medical image segmentation, in: Bekkers, E., Wolterink, J.M., Aviles-Rivero, A. (Eds.), Proceedings of the First International Wo...

    2022

  32. [40]

    Sa-med2d-20m dataset: Segment anything in 2d medical imaging with 20 million masks

    Ye,J.,Cheng,J.,Chen,J.,Deng,Z.,Li,T.,Wang,H.,Su,Y.,Huang,Z.,Chen,J.,Jiang,L.,Sun,H.,Zhu,M.,Zhang,S.,He,J.,Qiao,Y.,2023. Sa-med2d-20m dataset: Segment anything in 2d medical imaging with 20 million masks. URL:https://arxiv.org/abs/2311.11969, arXiv:2311.11969

    2023 arXiv

  33. [41]

    Deep long-tailed learning: A survey

    Zhang, Y., Kang, B., Hooi, B., Yan, S., Feng, J., 2023a. Deep long-tailed learning: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence 45, 10795–10816. doi:10.1109/TPAMI.2023.3268118

    2023

  34. [42]

    Affine image registration of arterial spin labeling mri using deep learning networks

    Zhang, Z., Yang, H., Guo, Y., Bolo, N.R., Keshavan, M., DeRosa, E., Anderson, A.K., Alsop, D.C., Yin, L., Dai, W., 2023b. Affine image registration of arterial spin labeling mri using deep learning networks. NeuroImage 279, 120303. URL:https://www.sciencedirect. com/science/ar...

  35. [43]

    Spineregnet:Spineregistrationnetworkforvolumetric mr and ct image by the joint estimation of an affine-elastic deformation field

    Zhao,L.,Pang,S.,Chen,Y.,Zhu,X.,Jiang,Z.,Su,Z.,Lu,H.,Zhou,Y.,Feng,Q.,2023. Spineregnet:Spineregistrationnetworkforvolumetric mr and ct image by the joint estimation of an affine-elastic deformation field. Medical Image Analysis 86, 102786. URL:https://www. sciencedirect.com/sci...

    2023

  36. [241]

    doi:10.1109/BIBE.2016.39

    2016 doi

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.