Pith. sign in

REVIEW 4 major objections 5 minor 32 references

Incorporating Cyclic Group Equivariance into Deep Learning for Reliable Reconstruction of Rotationally Symmetric Tomography Systems

T0 review · 4 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Pith's one-line read The paper derives a cyclic equivariance identity coupling rotation of projections to rotation of images, and shows that enforcing it in a deep network improves tomographic reconstruction under distribution shift.

desk verdict A plausible equivariance-based improvement for domain-transform reconstruction, with a sound derivation but experimental gaps and an interpolation caveat that keep me from fully trusting the symmetry claims. read the letter →

arxiv 2502.02418 v1 pith:RPHMVPHJ submitted 2025-02-04 physics.med-ph eess.SP

classification physics.med-pheess.SP
keywords cyclicgroupequivarianceimagereconstructioncomputedtomographydeeplearningrotationalsymmetryequivariantconvolutiondomaintransformgeneralization
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 establish that the rotational symmetry of tomography hardware should be treated as a hard constraint on deep reconstruction networks, not just a property of the data. It derives the equivariance condition that cyclically shifting the angular index of the projection data must rotate the reconstructed image by the corresponding discrete angle, and it identifies the cyclic group underlying both domains. To enforce the condition, the paper contributes two mechanisms: cyclic rotation equivariant convolutions in the projection domain and a cyclic group equivariance regularizer over the full network. On digital brain phantoms, these additions stabilize training and improve reconstruction quality specifically when the test data look different from the training data. If the claim holds, cyclic group equivariance becomes a unifying design principle for any rotationally symmetric imaging system.

What carries the argument

The load-bearing object is the cyclic group $C_N$ together with the equivariance identity that couples its two representations: cyclic shifts $\mathcal{T}_m$ on the sinogram's angular axis and discrete rotations $R_m$ on the image plane. The CREC layer realizes group convolution on the angular axis, wrapping kernels around the boundary so that the projection feature maps remain circular. The CGER regularizer adds a term summing $\lVert \mathcal{F}(\mathcal{T}_i \mathbf{g}, \theta) - R_i \mathcal{F}(\mathbf{g}, \theta) \rVert^2$ for $i$ in a subgroup $C_M$, balancing it dynamically against the reconstruction loss. Together they cover the two ways a symmetry can enter a network: by architectural design in the projection domain and by optimization across the whole reconstruction map.

What would settle it

Retrain the CGER-constrained model with an exact discrete rotation (e.g., Fourier-based or three-shear rotation) in place of bicubic interpolation and replot the polar symmetry-consistency error; if the error spikes at 90, 180, and 270 degrees persist, the residual inconsistency is not interpolation noise and the discrete-grid equivariance identity itself is the weak point.

Watch

Extended reading notes

Core claim

The central claim is equation (11): for every $m \in \mathbb{Z}_N$, $\mathcal{F}[\mathcal{T}_m \mathbf{g}] = R_m[\mathbf{y}]$, where $\mathcal{F}$ is the reconstruction map, $\mathcal{T}_m$ cyclically rotates the angular axis of the projection data, and $R_m$ rotates the reconstructed discrete image by $2\pi m/N$. The paper argues this identity follows from the forward model when the $N$ detector arrays are identical and uniformly distributed around the central axis, and that it equips the projection space with a uniformly distributed circular structure. It then shows how to enforce the identity in practice: CREC layers perform group convolution along the angular axis with circular boundary handling so that projection features inherit the symmetry, and CGER penalizes deviations from the equivariance relation, computed over a subgroup $C_{32}$ of $C_{128}$ with image rotations performed by bicubic interpolation. Across discrete and fuzzy phantom experiments, the constrained networks exhibit lower rotational inconsistency and better out-of-distribution reconstruction than the unconstrained baseline.

Load-bearing premise

The load-bearing premise is that rotating a reconstructed image with bicubic interpolation is a faithful discrete stand-in for the continuous rotation $R_m$; for angles far from multiples of 90 degrees, interpolation error could be misread as an equivariance violation and distort the reconstruction.

Editorial extensions

If this is right

  • A domain-transform network with CREC layers outperforms its unconstrained counterpart on in-distribution discrete phantoms, improving SSIM from 0.8506 to 0.9144 while keeping rRMSE below filtered back-projection.
  • Adding CGER sharply improves out-of-distribution reconstruction, with SSIM rising by up to 0.3114 for the unmodified network and by up to 0.1862 for the CREC-modified network on fuzzy phantoms.
  • CGER narrows the training-validation gap and makes convergence smoother, which is a sign of reduced overfitting in the limited-data setting.
  • Under a full $C_{128}$ symmetry-consistency check, both mechanisms lower the rotation-error distribution, and the CREC-plus-CGER model gives the most uniform errors across all 127 non-identity rotations.

Reading between the lines

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

  • Beyond the paper: because the equivariance identity is derived purely from detector geometry, the same CREC and CGER modules should transfer to PET, SPECT, and optoacoustic tomography, and the natural test is whether they reduce the number of training angles needed for a target image quality.
  • Beyond the paper: the regularizer is evaluated only on a subgroup $C_{32}$ of $C_{128}$; sweeping over all divisors of $N$ would reveal how much of the group's symmetry is responsible for the stability gain and whether full-group regularization is worth its extra cost.
  • Beyond the paper: the observed interpolation anomalies at 90, 180, and 270 degrees suggest replacing the bicubic rotation with an exact discrete rotation operator; this is a directly testable modification that could make the regularizer an exact symmetry constraint rather than an approximate one.
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

4 major / 5 minor

Summary. The paper argues that tomography systems with N-fold rotational symmetry induce a cyclic group C_N that acts by cyclic rotation on projection data and by discrete rotation on the reconstructed image. The central theoretical claim, Eq. (11), states that any reconstruction map F must be equivariant under this group, i.e., F[T_m g] = R_m[y] for all m in Z_N. Based on this, the authors propose two mechanisms to enforce the symmetry in a deep-learning reconstruction network: cyclic rotation equivariant convolution (CREC) layers applied to the projection domain, and a cyclic group equivariance regularization (CGER) term that penalizes deviations from equivariance during training. They integrate these modules into an AUTOMAP-style domain-transform network and evaluate on digital brain phantoms, training on discrete phantoms and testing on fuzzy phantoms from the same brain models. The results show that the modified network with CREC and CGER yields higher SSIM, better artifact suppression, and more uniform angular consistency in polar plots than the unmodified AUTOMAP baseline.

Significance. If the central claim is correct, the paper offers a principled, hardware-derived symmetry constraint that could improve generalization and data efficiency of deep-learning tomographic reconstruction, which is a timely and practically relevant contribution. The derivation from sensitivity functions to Eq. (11) is self-contained and does not rely on fitted parameters, which is a clear strength. The experimental design includes a meaningful distribution shift (discrete to fuzzy phantoms) and reports notable quantitative gains. However, the paper does not release code or data, does not control for parameter count in ablations, and the implementation of the proposed CREC layer appears degenerate when the kernel width is 1. The symmetry-consistency evaluation and the CGER regularizer both depend on a bicubic-interpolation surrogate for rotations that is not an exact group action, so the quantitative gains may partly reflect fitting this surrogate rather than true C_N equivariance. These issues must be addressed before the claims can be fully accepted.

major comments (4)
  1. [Implementation Details and Eq. (14) (CGER)] The CGER regularizer and the symmetry-consistency evaluation in Fig. 9 realize the image-domain rotations R_i with bicubic interpolation. Bicubic rotation of a discrete pixel grid is not a group representation of C_N: it is not exactly invertible, and composition R_i after R_j only approximates R_{i+j} due to resampling. Thus the regularizer enforces consistency with an angle-dependent approximate surrogate rather than with true C_N equivariance. The paper acknowledges interpolation anomalies at 90/180/270 degrees but does not calibrate the surrogate error. Please provide a quantitative measure of the surrogate's deviation from a true group action (e.g., composition error ||R_i(R_j(x)) - R_{i+j}(x)|| for representative images), compare with exact discrete rotation implementations such as Fourier-domain rotation or shear-based rotation, and report the CGER error for a known equivariant method like FBP. Without such calibration, the reported SSIM gains and the 'more uniform' polar plots in Fig. 9 may largely reflect fitting the surrogate.
  2. [CREC layer, Fig. 2 and Implementation Details] The paper states in the CREC design section that the kernel width is set to 1, eliminating the need for wrapping around. A kernel of width 1 along the angular axis performs no mixing of neighboring angular positions, so the cyclic equivariance is trivially realized but the layer does not exploit the circular structure of the projection domain as claimed. This makes the contribution of the CREC layer ambiguous: the observed improvements of Mod-AUTOMAP over AUTOMAP could stem from the added fully-connected or convolutional parameters rather than from cyclic equivariance. Please clarify the exact kernel size used in the experiments, and if the width is indeed 1, explain how this demonstrates a cyclic equivariant convolution design; otherwise, provide results with a non-trivial kernel width and report the kernel size.
  3. [Experimental comparisons in Table I] The comparison between AUTOMAP and Mod-AUTOMAP (with three CREC layers) is not an ablation that controls for parameter count or model capacity. Adding three layers increases the number of trainable parameters and can improve performance independently of equivariance. To support the claim that the gains are due to cyclic equivariance, the authors should include a control model with the same architectural changes but without the cyclic constraint (e.g., replacing CREC layers with standard convolutional layers of identical shape and parameter count, or adding a comparable-capacity autoencoder branch). This is load-bearing because the headline results—e.g., SSIM improvement from 0.8506 to 0.9144 on discrete phantoms—could simply reflect a larger network.
  4. [Datasets and generalization claims] The fuzzy phantom test sets are derived from the same two brain models as the discrete test set (as stated in the Datasets section). While the pixel-intensity distributions differ (discrete vs. fuzzy), the underlying anatomy is identical, so the 'external test' evaluates a restricted form of distribution shift, not anatomical variability. This limits the strength of the generalization conclusions. Please either add tests on fuzzy phantoms from held-out brain models not used in training or clearly moderate the claims of generalization to new patient anatomies.
minor comments (5)
  1. [Equation numbering] Equation (14) is used twice: once for the cyclic rotation convolution definition and once for the CGER regularized objective. Please renumber the equations to avoid ambiguity.
  2. [Typos] There are several typographical errors, e.g., 'Equitation (10)' should be 'Equation (10)', and '𝒚𝒚( 𝒓𝒓′)' should be '𝒚𝒚(𝐫𝐫′)' in the derivation. Please proofread the manuscript.
  3. [References] Reference [23] (Hashemi et al.) is listed with the placeholder arXiv number 2301.00000, which appears to be a dummy identifier. Please update to the actual arXiv reference or remove the placeholder.
  4. [Figure 9] The polar plots in Fig. 9 are said to show rRMSE across 127 angular transformations, but the axis labels are not visible in the caption and the plots appear to be on different scales. Please add a shared legend and consistent radial-axis scaling so that 'more uniform' and 'lower' errors can be compared across panels.
  5. [Code and data availability] The manuscript does not state whether the code, trained models, or phantom datasets will be made available. Given the reproducibility standards of the field, please include a data/code availability statement.

Circularity Check

1 steps flagged · score 2.0 of 10

The Fig. 9 symmetry-consistency metric is the CGER training objective itself, so that particular evaluation is partly self-referential; the reconstruction-quality results on fuzzy phantoms remain independent evidence.

  1. self definitional [Eq. (14); Section 'Evaluation of the Symmetry Consistency' (Fig. 9)]
    "arg min_θ (1/(N−1)) Σ_{i=1}^{N−1} ||F(T_i g, θ) − R_i F(g, θ)||^2, (14) ... To assess rotational consistency under the full cyclic group C128, we extended the CGER constraint as an evaluation metric. Specifically, we computed the rRMSE between F(T_j g, θ) and R_j F(g, θ) for all j = 1 … 127"

    The evaluation metric used in Fig. 9 is exactly the functional minimized by the CGER regularizer in Eq. (14), merely summed over the full group C128 instead of the training subgroup C32. Models trained with CGER are therefore scored on the same objective they were optimized against, so their lower and more uniform polar errors are partly forced by construction. This is a minor self-referential evaluation, not the paper's central claim; the SSIM/PSNR gains on the fuzzy phantoms are independent external measures and do not reduce to the equivariance loss.

full rationale

The central equivariance claim, Eq. (11), is derived from the detector-symmetry relation h_n(r) = h_0(R_n^{-1}r) and the forward projection model; it involves no fitted parameters and is not equivalent to its inputs. The CREC layer is a standard cyclic group convolution whose equivariance follows from the group definition, and the CGER regularizer is explicitly presented as an optimization penalty. The only place where a reported result coincides with the training objective is the symmetry-consistency evaluation of Fig. 9, where the rRMSE between F(T_j g) and R_j F(g) is precisely the quantity minimized in Eq. (14). The paper is transparent about this, saying it 'extended the CGER constraint as an evaluation metric,' so this is a minor self-referential check rather than a hidden prediction. The fuzzy-phantom reconstruction improvements (SSIM gains such as +0.31 for CSF) are independent external benchmarks, and the bicubic-rotation caveat raised in the implementation is a correctness/robustness concern about whether the discrete rotation R_i faithfully represents the continuous group action, not a circularity. The single self-citation [31] appears only in a list of conventional iterative methods and is not load-bearing. Overall, the derivation is self-contained and the main empirical claims are independently supported; the score is 2 due only to the self-referential symmetry-consistency evaluation.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. Its derivation uses standard group theory and an ideal linear forward model. The free parameters are experimental design choices (subgroup size M and kernel width), not fitted constants in a theory. The main assumptions are the ideal symmetry of the scanner and the fidelity of discrete rotations.

free parameters (2)
  • Subgroup order M for CGER = 32
    Chosen to balance computational cost; only 32 of 128 rotations are used in training, approximating full C_128 equivariance. This is a hand-selected experimental design choice, not a fitted constant.
  • CREC kernel width = 1
    Set to 1 in the simple implementation, making the convolution pointwise along the angular axis and avoiding wrapping. This limits angular mixing and nearly trivializes the architectural contribution.
assumptions (4)
  • domain assumption Ideal N-fold rotational symmetry of the imaging system with N identical detector arrays and time-invariance.
    Invoked in Section 'From Rotational Symmetry of Systems to Cyclic Group Equivariance' and Eq. (7). Real hardware may have sensitivity variations and scatter that break strict symmetry.
  • domain assumption The reconstruction mapping F is treated as a deterministic inverse of the forward model, so equivariance of the forward map transfers to F.
    Used to state Eq. (11). For regularized or learned reconstructions this is an idealization; the paper relies on training (via CREC and CGER) to approximate it.
  • domain assumption Discrete image rotation by R_m via bicubic interpolation faithfully represents continuous rotation.
    Invoked in the CGER implementation and the symmetry-consistency evaluation. Interpolation errors are acknowledged for some angles, which weakens the fidelity of this assumption.
  • domain assumption The Radon transform simulates the forward projection of CT.
    Used in experiments to generate 128-angle projections from 190x190 brain phantom slices, matching the discrete forward model of Eq. (7).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Incorporating Cyclic Group Equivariance into Deep Learning for Reliable Reconstruction of Rotationally Symmetric Tomography Systems." pith.science (2026). https://pith.science/paper/RPHMVPHJ

@misc{pith2026250202418,
  author       = {Pith},
  title        = {Pith review of: Incorporating Cyclic Group Equivariance into Deep Learning for Reliable Reconstruction of Rotationally Symmetric Tomography Systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RPHMVPHJ}},
  note         = {Machine review of arXiv:2502.02418}
}
read the original abstract

Rotational symmetry is a defining feature of many tomography systems, including computed tomography (CT) and emission computed tomography (ECT), where detectors are arranged in a circular or periodically rotating configuration. This study revisits the image reconstruction process from the perspective of hardware-induced rotational symmetry and introduces a cyclic group equivariance framework for deep learning-based reconstruction. Specifically, we derive a mathematical correspondence that couples cyclic rotations in the projection domain to discrete rotations in the image domain, both arising from the same cyclic group inherent in the hardware design. This insight also reveals the uniformly distributed circular structure of the projection space. Building on this principle, we provide a cyclic rotation equivariant convolution design method to preserve projection domain symmetry and a cyclic group equivariance regularization approach that enforces consistent rotational transformations across the entire network. We further integrate these modules into a domain transform reconstruction framework and validate them using digital brain phantoms, training on discrete models and testing on more complex and realistic fuzzy variants. Results indicate markedly improved generalization and stability, with fewer artifacts and better detail preservation, especially under data distribution deviation. These findings highlight the potential of cyclic group equivariance as a unifying principle for tomographic reconstruction in rotationally symmetric systems, offering a flexible and interpretable solution for scenarios with limited data.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 31 canonical work pages

  1. [1]

    Deep learning for tomographic image reconstruction,

    G. Wang, J. C. Ye, and B. De Man, "Deep learning for tomographic image reconstruction," Nat. Mach. Intell., vol. 2, no. 12, pp. 737-748, Dec. 2020

  2. [2]

    Deep learning for biomedical image reconstruction: A survey,

    H. Ben Yedder, B. Cardoen, and G. Hamarneh , "Deep learning for biomedical image reconstruction: A survey," Artif. Intell. Rev., vol. 54, no. 1, pp. 215 - 251, Jan. 2021

  3. [3]

    Image reconstruction by domain- transform manifold learning,

    B. Zhu, J. Z. Liu, S. F. Cauley, B. R. Rosen, and M. S. Rosen, "Image reconstruction by domain- transform manifold learning," Nature, vol. 555, no. 7697, pp. 487-492, Mar. 2018

  4. [4]

    Learning to reconstruct computed tomography images directly from sinogram data under a variety of data acquisition conditions,

    Y. Li, K. Li, C. Zhang, J. Montoya, and G.- H. Chen, "Learning to reconstruct computed tomography images directly from sinogram data under a variety of data acquisition conditions," IEEE Trans. Med. Imag., vol. 38, no. 10, pp. 2469-2481, Oct. 2019

  5. [5]

    DeepPET: A deep encoder –decoder network for directly solving the PET image reconstruction inverse problem,

    I. Häggström, C. R. Schmidtlein, G. Campanella, and T. J. Fuchs, "DeepPET: A deep encoder –decoder network for directly solving the PET image reconstruction inverse problem," Med. Image Anal., vol. 54, pp. 253-262, 2019

  6. [6]

    DirectPET: full -size neural network PET reconstruction from sinogram data,

    W. Whiteley, W. K. Luk, and J. Gregor, "DirectPET: full -size neural network PET reconstruction from sinogram data," J. Med. Imag., vol. 7, no. 3, pp. 032503-032503, 2020

  7. [7]

    A learned reconstruction network for SPECT imaging,

    W. Shao, M. G. Pomper, and Y. Du, "A learned reconstruction network for SPECT imaging," IEEE Trans. Radiat. Plasma Med. Sci., vol. 5, no. 1, pp. 26-34, Jan. 2020

  8. [8]

    Deep embedding- attention-refinement for sparse-view CT reconstruction,

    W. Wu, X. Guo, Y. Chen, S. Wang, and J. Chen, “Deep embedding- attention-refinement for sparse-view CT reconstruction,” IEEE Trans. Instrum. Meas., vol. 72, pp. 1–11, 2022

Show all 32 references
  1. [9]

    WNet: A data-driven dual-domain denoising model for sparse- view computed tomography with a trainable reconstruction layer,

    T. Cheslerean-Boghiu, F. C. Hofmann, M. Schultheiß, F. Pfeiffer, D. Pfeiffer, and T. Lasser, “WNet: A data-driven dual-domain denoising model for sparse- view computed tomography with a trainable reconstruction layer,” IEEE Trans. Comput. Imag., vol. 9, pp. 120–132, 2023

  2. [10]

    Direct reconstruction for simultaneous dual- tracer PET imaging based on multi- task learning,

    F. Zeng, J. Fang, A. Muhashi, and H. Liu, “Direct reconstruction for simultaneous dual- tracer PET imaging based on multi- task learning,” EJNMMI Res., vol. 13, no. 1, p. 7, 2023

  3. [11]

    ReconU-Net: A direct PET image reconstruction using U- Net architecture with back projection-induced skip connection,

    F. Hashimoto and K. Ote, “ReconU-Net: A direct PET image reconstruction using U- Net architecture with back projection-induced skip connection,” Phys. Med. Biol., vol. 69, no. 10, p. 105022, 2024

  4. [12]

    On instabilities of deep learning in image reconstruction and the potential costs of AI,

    V. Antun, F. Renna, C. Poon, B. Adcock, and A. C. Hansen, "On instabilities of deep learning in image reconstruction and the potential costs of AI," Proc. Natl. Acad. Sci. USA, vol. 117, no. 48, pp. 30088-30095, 2020

  5. [13]

    A review of deep learning CT reconstruction: Concepts, limitations, and promise in clinical practice,

    T. P. Szczykutowicz, G. V. Toia, A. Dhanantwari, and B. Nett, "A review of deep learning CT reconstruction: Concepts, limitations, and promise in clinical practice," Curr. Radiol. Rep., vol. 10, no. 9, pp. 101-115, 2022

  6. [14]

    Geometric deep learning and equivariant neural networks,

    J. E. Gerken, J. Aronsson, O. Carlsson, H. Linander, F. Ohlsson, C. Petersson, and D. Persson, “Geometric deep learning and equivariant neural networks,” Artif. Intell. Rev., vol. 56, no. 12, pp. 14605–14662, 2023

  7. [15]

    Harmonic networks: Deep translation and rotation equivariance,

    D. E. Worrall, S. J. Garbin, D. Turmukhambetov, and G. J. Brostow, "Harmonic networks: Deep translation and rotation equivariance," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit., pp. 5028-5037, 2017

  8. [16]

    Robust equivariant imaging: A fully unsupervised framework for learning to image fro m noisy and partial measurements,

    D. Chen, J. Tachella, and M. E. Davies, "Robust equivariant imaging: A fully unsupervised framework for learning to image fro m noisy and partial measurements," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., pp. 5647-5656, 2022

  9. [17]

    Provably strict generalisation benefit for equivariant models,

    B. Elesedy and S. Zaidi, “Provably strict generalisation benefit for equivariant models,” in *Proc. Int. Conf. Mach. Learn. (ICML)*, PMLR, 2021, pp. 2959– 2969

  10. [18]

    Rotation-equivariant deep learning for diffusion MRI,

    P. Müller, V. Golkov, V. Tomassini, and D. Cremers, “Rotation-equivariant deep learning for diffusion MRI,” arXiv preprint arXiv:2102.06942, 2021

  11. [19]

    Image classification based on CNN: A sur vey,

    A. A. Elngar, M. Arafa, A. Fathy, B. Moustafa, O. Mahmoud, M. Shaban, and N. Fawzy, “Image classification based on CNN: A sur vey,” J. Cybersecurity Inf. Manag., vol. 6, no. 1, pp. 18–50, 2021

  12. [20]

    Dense steerable filter CNNs for exploiting rotational symmetry in histology images,

    S. Graham, D. Epstein, and N. Rajpoot, "Dense steerable filter CNNs for exploiting rotational symmetry in histology images," IEEE Trans. Med. Imag., vol. 39, no. 12, pp. 4124-4136, Dec. 2020

  13. [21]

    Pulmonary nodule detection in CT scans with equivariant CNNs,

    M. Winkels and T. S. Cohen, “Pulmonary nodule detection in CT scans with equivariant CNNs,” Med. Image Anal., vol. 55, pp. 15 –26, 2019

  14. [22]

    Roto -translation equivariant convolutional networks: Application to histopathology image analysis,

    M. W. Lafarge, E. J. Bekkers, J. P. W. Pluim, R. Duits, and M. Veta, “Roto -translation equivariant convolutional networks: Application to histopathology image analysis,” Med. Image Anal., vol. 68, p. 101849, 2021

  15. [23]

    Spherical CNN for medical imaging applications: Importance of equivariance in image recon struction and denoising,

    A. Hashemi, Y. Feng, and H. Sabet, "Spherical CNN for medical imaging applications: Importance of equivariance in image recon struction and denoising," arXiv preprint arXiv:2301.00000, 2023

  16. [24]

    SE(3)- equivariant and noise- invariant 3D rigid motion tracking in brain MRI,

    B. Billot, N. Dey, D. Moyer, M. Hoffmann, E. A. Turk, B. Gagoski, P. E. Grant, and P. Golland, "SE(3)- equivariant and noise- invariant 3D rigid motion tracking in brain MRI," IEEE Trans. Med. Imag., 2024 (to be published)

  17. [25]

    Equivariant neural networks for inverse problems,

    E. Celledoni, M. J. Ehrhardt, C. Etmann, B. Owren, C.-B. Schönlieb, and F. Sherry, “Equivariant neural networks for inverse problems,” Inverse Probl., vol. 37, no. 8, p. 085006, 2021

  18. [26]

    Equivariant plug -and-play image reconstruction,

    M. Terris, T. Moreau, N. Pustelnik, and J. Tachella , "Equivariant plug -and-play image reconstruction," in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit., pp. 25255-25264, 2024

  19. [27]

    Twenty new digital brain phantoms for creation of validation image data bases,

    B. Aubert-Broche, M. Griffin, G. B. Pike, A. C. Evans, and D. L. Collins, "Twenty new digital brain phantoms for creation of validation image data bases," IEEE Trans. Med. Imag., vol. 25, no. 11, pp. 1410-1416, Nov. 2006

  20. [28]

    Implementing and accelerating the EM algorithm for positron emission tomography,

    L. Kaufman, "Implementing and accelerating the EM algorithm for positron emission tomography," IEEE Trans. Med. Imag., vol. 6 , no. 1, pp. 37-51, Mar. 1987

  21. [29]

    SVD for imaging systems with discrete rotational symmetry,

    E. Clarkson, R. Palit, and M. A. Kupinski, "SVD for imaging systems with discrete rotational symmetry," Opt. Express, vol. 18 , no. 24, pp. 25306- 25320, 2010

  22. [30]

    Fully -3D PET image reconstruction using scanner -independent, adaptive projection data and highly rotation - symmetric voxel assemblies,

    J. J. Scheins, H. Herzog, and N. J. Shah, "Fully -3D PET image reconstruction using scanner -independent, adaptive projection data and highly rotation - symmetric voxel assemblies," IEEE Trans. Med. Imag., vol. 30, no. 3, pp. 879-892, Mar. 2011. Fuzzy Model: Cerebrospinal Flui...

  23. [31]

    Monte- Carlo system modeling for PET reconstruction: A rotator approach,

    L. Zhang, S. Vandenberghe, S. Staelens, J. Verhaeghe, I. Kawrakow, and I. Lemahieu, "Monte- Carlo system modeling for PET reconstruction: A rotator approach," in 2008 IEEE Nucl. Sci. Symp. Conf. Rec., pp. 5101-5106, 2008

  24. [32]

    Discrete imaging models for three- dimensional optoacoustic tomography using radially symmetric expansion functions,

    K. Wang, R. W. Schoonover, R. Su, A. Oraevsky, and M. A. Anastasio, "Discrete imaging models for three- dimensional optoacoustic tomography using radially symmetric expansion functions," IEEE Trans. Med. Imag., vol. 33, no. 5, pp. 1180-1193, May 2014

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.