Pith. sign in

REVIEW 2 major objections 1 minor 43 references

What neurosurgeons need to see: synthetic intra-operative MRI from ultrasound for brain-shift compensation in brain tumour surgery

T0 review · 2 major / 1 minor · reviewed 2026-06-28 · grok-4.3

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

Pith's one-line read A pipeline merges preoperative MRI with ultrasound-derived synthetic images to produce updated whole-brain volumes reflecting post-resection state.

desk verdict Pipeline generates updated whole-brain volumes from ioUS but registration accuracy matches the classical baseline with no targeted checks on resection cavity or residual tumor. read the letter →

arxiv 2606.07658 v1 pith:G7CZR5ZJ submitted 2026-06-03 cs.CV cs.LG

Santiago Cepeda , Olga Esteban-Sinovas , Ignacio Arrese , Rosario Sarabia This is my paper · ORCID
classification cs.CVcs.LG
keywords intraoperativeultrasoundsyntheticMRIbrainshiftneuronavigationgliomasurgeryimagesynthesisdeformableregistrationpost-resectionimaging
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 an end-to-end pipeline that generates a new whole-brain MRI volume in preoperative imaging space by combining the original preoperative MRI, a synthetic MRI generated from intraoperative ultrasound, and a deformable registration anchored on the synthetic image. It uses a 2.5D residual-transformer synthesis model and a two-stage registration process operating on raw scanner inputs. This addresses brain shift after dural opening without needing dedicated intraoperative MRI infrastructure, relying instead on inexpensive and repeatable ultrasound. A sympathetic reader would care because the resulting volume inside the ultrasound field of view reflects the intraoperative post-resection state, including resection cavity and residual tumor, with potential for direct integration into surgical navigation workflows. On a post-resection cohort the method yields synthetic images matching intraoperative T2 metrics and produces diffeomorphic deformation fields in every subject while matching a classical registration baseline in target error.

What carries the argument

The ResViT-2.5D residual-transformer synthesis backbone that generates a synthetic MRI from ioUS to anchor subsequent deformable registration and produce the merged whole-brain volume.

What would settle it

Direct visual or quantitative comparison of the synthetic volume against actual intraoperative MRI in additional post-resection cases focused on fidelity inside the resection cavity.

Watch

Extended reading notes

Core claim

The central claim is that the integrated volume reflects the intraoperative post-resection state inside the ultrasound field of view, achieved by synthesis from ioUS followed by synthesis-anchored registration, thereby providing the surgeon with an MRI-like update of the operative field.

Load-bearing premise

The synthetic MRI generated from intraoperative ultrasound accurately represents structures absent from the preoperative scan such as the resection cavity and residual tumor.

Editorial extensions

If this is right

  • The synthesis-anchored registration reduces mean target registration error from 6.27 mm to 5.86 mm while producing a diffeomorphic deformation field in every subject.
  • The integrated volume reflects the post-resection state inside the ultrasound field of view and matches intraoperative T2 across structural, intensity, and perceptual metrics.
  • The pipeline operates directly on raw scanner inputs and yields results comparable to a strong classical NiftyReg baseline.

Reading between the lines

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

  • The whole-brain update could allow surgeons to visualize changes outside the narrow ultrasound field of view if the synthesis extrapolates reliably.
  • Real-time integration testing in the operating room would determine whether the updated volumes alter surgical decisions compared with standard neuronavigation.
  • The same synthesis-plus-anchored-registration pattern might apply to other procedures where ultrasound is available but full intraoperative MRI is not.
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, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes an end-to-end pipeline that synthesizes a whole-brain intraoperative MRI volume from preoperative MRI and intraoperative ultrasound (ioUS) using a ResViT-2.5D residual-transformer network, then merges it with a two-stage deformable registration (NiftyReg followed by synthesis-anchored SynthMorph) to compensate for brain shift during glioma resection. On a post-resection ReMIND cohort, the synthetic images match intraoperative T2 across structural, intensity, and perceptual metrics; in 14 subjects with 215 expert landmarks the synthesis-anchored registration yields a mean target registration error (TRE) of 5.86 mm (down from 6.27 mm), statistically equivalent to a strong classical NiftyReg baseline (5.85 mm) while producing a diffeomorphic deformation field in every case. The stated contribution is the integrated volume itself, which is claimed to reflect the post-resection state inside the ultrasound field of view.

Significance. If the synthesis accurately renders resection cavities and residual tumor (structures absent from the preoperative scan) and the deformation field extrapolates reliably outside the narrow ultrasound FOV, the method could supply MRI-like intraoperative updates using only standard ioUS hardware, removing the need for dedicated intraoperative MRI infrastructure. The equivalence of registration accuracy to the classical baseline underscores that any clinical value resides entirely in the fidelity and usability of the synthesized volume rather than in improved alignment metrics.

major comments (2)
  1. [Abstract] Abstract: The reported TRE reduction (6.27 mm o 5.86 mm) exactly matches the NiftyReg baseline (5.85 mm), so the central claim that the integrated volume supplies a usable whole-brain update rests solely on the untested ability of ResViT-2.5D to correctly depict resection cavity and residual tumor inside the US FOV and to support reliable extrapolation outside it; only aggregate structural/intensity/perceptual metrics are presented, with no isolated error quantification or independent ground-truth validation on these new structures.
  2. [Abstract] Abstract / Methods (implied): No information is supplied on training/validation splits, hyperparameter selection protocol, or the precise landmark placement and annotation procedure used for the 215 landmarks in 14 subjects; given the small cohort size, these omissions leave open the possibility that the reported matching metrics and TRE values are sensitive to post-hoc choices.
minor comments (1)
  1. [Abstract] Abstract: The sentence fragment 'which inside the ultrasound field of view it reflects the intraoperative post-resection state' is grammatically incomplete and should be rephrased for clarity.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the thorough review and constructive feedback. We address each major comment below and outline the revisions we will make to the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The reported TRE reduction (6.27 mm → 5.86 mm) exactly matches the NiftyReg baseline (5.85 mm), so the central claim that the integrated volume supplies a usable whole-brain update rests solely on the untested ability of ResViT-2.5D to correctly depict resection cavity and residual tumor inside the US FOV and to support reliable extrapolation outside it; only aggregate structural/intensity/perceptual metrics are presented, with no isolated error quantification or independent ground-truth validation on these new structures.

    Authors: We agree that the TRE improvement is negligible and statistically equivalent to the NiftyReg baseline, consistent with our manuscript's emphasis that the primary contribution is the integrated synthetic volume rather than registration accuracy. The structural, intensity, and perceptual metrics are computed across the full image, including the resection cavity and residual tumor within the ioUS field of view. However, we acknowledge the value of more targeted validation. In the revised manuscript, we will include additional qualitative examples highlighting the synthesis of resection cavities and residual tumor, and discuss the limitations of aggregate metrics. Independent ground-truth for these structures beyond the provided intraoperative T2 is not available in the ReMIND dataset, limiting quantitative isolation. revision: partial

  2. Referee: [Abstract] Abstract / Methods (implied): No information is supplied on training/validation splits, hyperparameter selection protocol, or the precise landmark placement and annotation procedure used for the 215 landmarks in 14 subjects; given the small cohort size, these omissions leave open the possibility that the reported matching metrics and TRE values are sensitive to post-hoc choices.

    Authors: We apologize for these omissions in the manuscript. The revised version will explicitly detail the training/validation splits used for ResViT-2.5D, the hyperparameter selection protocol (including any cross-validation approach), and the landmark annotation procedure, including how the 215 expert landmarks were placed and verified in the 14 subjects. These details will clarify the robustness of the results despite the small cohort size. revision: yes

standing simulated objections not resolved
  • Independent ground-truth validation specifically isolating error on resection cavities and residual tumor is not feasible with the current ReMIND dataset, as it lacks paired annotations for these post-resection structures beyond the aggregate image metrics.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; empirical evaluation on held-out data is self-contained

full rationale

The paper trains ResViT-2.5D on data to synthesize MRI from ioUS, then evaluates the output against real intraoperative T2 images using structural/intensity/perceptual metrics on a post-resection ReMIND cohort. Registration performance is measured via expert landmarks (215 in 14 subjects) and compared to an external classical NiftyReg baseline. No equations or steps reduce a claimed prediction to a fitted input by construction, no self-citations are load-bearing for uniqueness or ansatzes, and the central claim (integrated volume reflecting post-resection state) rests on external empirical matches rather than internal redefinitions. This is the standard non-circular case for a learned synthesis pipeline.

Assumptions & free parameters 1 free parameters · 1 assumptions · 0 invented entities

The claim rests on a trained synthesis network whose weights are fitted to the ReMIND cohort and on the domain assumption that synthetic images supply adequate contrast for anchoring registration outside the ultrasound field of view.

free parameters (1)
  • ResViT-2.5D network weights
    Learned parameters that map ultrasound to synthetic MRI; central to producing the anchor image
assumptions (1)
  • domain assumption Synthetic images from ioUS can serve as reliable anchors for diffeomorphic deformable registration
    Invoked in the two-stage registration step that produces the final volume

how reviews work

0 comments
Cite this review

Pith. "Pith review of What neurosurgeons need to see: synthetic intra-operative MRI from ultrasound for brain-shift compensation in brain tumour surgery." pith.science (2026). https://pith.science/paper/G7CZR5ZJ

@misc{pith2026260607658,
  author       = {Pith},
  title        = {Pith review of: What neurosurgeons need to see: synthetic intra-operative MRI from ultrasound for brain-shift compensation in brain tumour surgery},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/G7CZR5ZJ}},
  note         = {Machine review of arXiv:2606.07658}
}
read the original abstract

Maximal safe resection is the primary objective in glioma surgery. Neuronavigation guidance is progressively degraded by brain shift after dural opening. Intraoperative MRI can compensate but needs dedicated infrastructure and is rarely available, whereas intraoperative ultrasound (ioUS) is inexpensive, repeatable, and compatible with routine workflows. Navigation systems combining ioUS with preoperative MRI usually rely on rigid registration; even deformable multimodal registration is limited by ultrasound speckle contrast, a narrow field of view, and the inability to represent structures absent from the preoperative scan, most critically the resection cavity and residual tumor. We propose an end-to-end pipeline that generates a new whole-brain MRI volume in the preoperative imaging space by merging the preoperative MRI, a synthetic MRI generated from the ioUS, and a deformable registration anchored on that synthetic image. It integrates a 2.5D residual-transformer synthesis backbone (ResViT-2.5D) and a two-stage registration coupling NiftyReg with a synthesis-anchored SynthMorph stage, operating directly on raw scanner inputs. On a post-resection ReMIND cohort, ResViT-2.5D produced synthetic images closely matching the intraoperative T2 across structural, intensity, and perceptual metrics. In 14 subjects with 215 expert landmarks, the synthesis-anchored registration reduced the mean target registration error from 6.27 to 5.86 mm, matching a strong classical NiftyReg baseline (5.85 mm) while yielding a diffeomorphic deformation field in every subject. The contribution is not a gain in registration accuracy but the integrated volume itself, which inside the ultrasound field of view it reflects the intraoperative post-resection state. This provides the surgeon with an MRI-like update of the operative field with potential for integration into surgical-navigation workflows.

Figures

Figures reproduced from arXiv: 2606.07658 by the authors.

Figure 1
Figure 1. Overview of the proposed end-to-end pipeline. From a single post-resection intraoperative ultrasound (ioUS) [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Architectures of the three candidate synthesis backbones. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Qualitative registration across four representative subjects (rows). Each panel overlays the intraoperative [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Geometry of the composed transformation for three representative subjects (rows). [PITH_FULL_IMAGE:figures/full_fig_p020_4.png]
Figure 5
Figure 5. Figure 5: Synthetic intraoperative MRI (siMRI) for three representative subjects, each shown in three orthogonal planes [PITH_FULL_IMAGE:figures/full_fig_p022_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 40 canonical work pages

  1. [1]

    Q-id: Lightweight quantum network server identification through fingerprinting.Netwrk

    Azampour, M.F., Mach, K., Fatemizadeh, E., Demiray, B., Westenfelder, K., Steiger, K., Eiber, M., Wendler, T., Kainz, B., Navab, N., 2024. Multitask weakly supervised generative network for MR-US registration. IEEE Transactions on Medical Imaging 43, 3780–3793.https://doi.org/10.1109/TMI.2024.3400899

    work page doi:10.1109/tmi.2024.3400899 2024

  2. [2]

    DiffUS: differentiable ultrasound rendering from volumetric imaging.https://arxiv.org/abs/2508.06768

    Bertramo, N., Duguey, G., Gopalakrishnan, V ., 2025. DiffUS: differentiable ultrasound rendering from volumetric imaging.https://arxiv.org/abs/2508.06768

    work page arXiv 2025

  3. [3]

    IEEE Transactions on Medical Imaging42(5), 1289–1300 (2023).https://doi.org/10.1109/TMI.2022

    Dalmaz, O., Yurt, M., Çukur, T., 2022. ResViT: residual vision transformers for multimodal medical image synthesis. IEEE Transactions on Medical Imaging 41, 2598–2614. https://doi.org/10.1109/TMI.2022. 3167808

    work page doi:10.1109/tmi.2022 2022

  4. [4]

    Unified brain MR-ultrasound synthesis using multi-modal hierarchical representations

    Dorent, R., Haouchine, N., Kögl, F., Joutard, S., Juvekar, P., Torio, E., Golby, A., Ourselin, S., Frisken, S., Vercauteren, T., Kapur, T., Wells, W.M., 2023. Unified brain MR-ultrasound synthesis using multi-modal hierarchical representations. In: Medical Image Computing and Computer-Assisted Intervention – MICCAI 2023, LNCS 14229. Springer, pp. 448–458....

    work page doi:10.1007/978-3-031-43999-5_43 2023

  5. [5]

    The Brain Resection Multimodal Image Registration (ReMIND2Reg) 2025 challenge

    Dorent, R., Rigolo, L., Galvin, C.P., Chen, J., Heinrich, M.P., Carass, A., Colliot, O., Wassermann, D., Golby, A., Kapur, T., Wells, W., 2025. The Brain Resection Multimodal Image Registration (ReMIND2Reg) 2025 challenge. https://arxiv.org/abs/2508.09649

    work page arXiv 2025

  6. [6]

    Unified cross-modal medical image synthesis with hierarchical mixture of product-of-experts

    Dorent, R., Haouchine, N., Golby, A., Frisken, S., Kapur, T., Wells, W., 2026. Unified cross-modal medical image synthesis with hierarchical mixture of product-of-experts. IEEE Transactions on Pattern Analysis and Machine Intelligence 48, 1641–1656.https://doi.org/10.1109/TPAMI.2025.3616632

    work page doi:10.1109/tpami.2025.3616632 2026

  7. [7]

    A Systematic Benchmark of Intraoperative Ultrasound-to-MR Synthesis for Brain Tumour Surgery

    Esteban-Sinovas, O., Cepeda, S., Arrese, I., Sarabia, R., 2026. A systematic benchmark of intraoperative ultrasound-to-MR synthesis for brain tumour surgery.https://arxiv.org/abs/2606.00630

    work page Pith review arXiv 2026

  8. [8]

    3d slicer as an image computing platform for the quantitative imaging network,

    Fedorov, A., Beichel, R., Kalpathy-Cramer, J., Finet, J., Fillion-Robin, J.C., Pujol, S., Bauer, C., Jennings, D., Fennessy, F., Sonka, M., Buatti, J., Aylward, S., Miller, J.V ., Pieper, S., Kikinis, R., 2012. 3D Slicer as an image computing platform for the Quantitative Imaging Network. Magnetic Resonance Imaging 30, 1323–1341. https://doi.org/10.1016/j...

    work page doi:10.1016/j.mri.2012.05.001 2012

Show all 43 references
  1. [9]

    Brain shift in neuronavi- gation of brain tumors: a review

    Gerard, I.J., Kersten-Oertel, M., Petrecca, K., Sirhan, D., Hall, J.A., Collins, D.L., 2017. Brain shift in neuronavi- gation of brain tumors: a review. Medical Image Analysis 35, 403–420. https://doi.org/10.1016/j.media. 2016.08.007. 27 Synthetic intra-operative MRI from ultrasound

    2017 doi

  2. [10]

    Learn2Reg 2024: new benchmark datasets driving progress on new challenges

    Hansen, L., Heyer, W., Großbröhmer, C., et al., 2025. Learn2Reg 2024: new benchmark datasets driving progress on new challenges. Journal of Machine Learning for Biomedical Imaging (MELBA) 3, 775–791. https://doi.org/10.59275/j.melba.2025-gc8c

    2025 doi

  3. [11]

    Towards realtime multimodal fusion for image-guided interventions using self-similarities

    Heinrich, M.P., Jenkinson, M., Papie˙z, B.W., Brady, S.M., Schnabel, J.A., 2013. Towards realtime multimodal fusion for image-guided interventions using self-similarities. In: MICCAI 2013, LNCS 8149. Springer, pp. 187–194.https://doi.org/10.1007/978-3-642-40811-3_24

    2013 doi

  4. [12]

    Maximizing safe resection of low- and high-grade glioma

    Hervey-Jumper, S.L., Berger, M.S., 2016. Maximizing safe resection of low- and high-grade glioma. Journal of Neuro-Oncology 130, 269–282.https://doi.org/10.1007/s11060-016-2110-4

    2016 doi

  5. [13]

    SynthMorph: learning contrast-invariant registration without acquired images

    Hoffmann, M., Billot, B., Greve, D.N., Iglesias, J.E., Fischl, B., Dalca, A.V ., 2022. SynthMorph: learning contrast-invariant registration without acquired images. IEEE Transactions on Medical Imaging 41, 543–558. https://doi.org/10.1109/TMI.2021.3116879

    2022 doi

  6. [14]

    Anatomy-aware and acquisition-agnostic joint registration with SynthMorph

    Hoffmann, M., Hoopes, A., Greve, D.N., Fischl, B., Dalca, A.V ., 2024. Anatomy-aware and acquisition-agnostic joint registration with SynthMorph. Imaging Neuroscience 2, 1–33. https://doi.org/10.1162/imag_a_ 00197

    2024 doi

  7. [15]

    Uncertainty estimates and multi-hypotheses networks for optical flow

    Ilg, E., Çiçek, Ö., Galesso, S., Klein, A., Makansi, O., Hutter, F., Brox, T., 2018. Uncertainty estimates and multi-hypotheses networks for optical flow. In: Computer Vision – ECCV 2018, LNCS 11211. Springer, pp. 677–693.https://doi.org/10.1007/978-3-030-01234-2_40

    2018 doi

  8. [16]

    Cross-modal conditional latent diffusion model for brain MRI to ultrasound image translation

    Jiang, S., Wang, L., Li, Y ., Yang, Z., Zhou, Z., Li, B., 2025. Cross-modal conditional latent diffusion model for brain MRI to ultrasound image translation. Physics in Medicine & Biology 70, 155005. https://doi.org/10. 1088/1361-6560/adf0bc

    2025

  9. [17]

    Self-supervised ultrasound to MRI fetal brain image synthesis

    Jiao, J., Namburete, A.I.L., Papageorghiou, A.T., Noble, J.A., 2020. Self-supervised ultrasound to MRI fetal brain image synthesis. IEEE Transactions on Medical Imaging 39, 4413–4424. https://doi.org/10.1109/TMI. 2020.3018560

    2020 doi

  10. [18]

    ReMIND: the brain resection multimodal imaging database

    Juvekar, P., Dorent, R., Kögl, F., Torio, E., Barr, C., Rigolo, L., Galvin, C., Jowkar, N., Kazi, A., Haouchine, N., Cheema, H., Navab, N., Pieper, S., Wells, W.M., Bi, W.L., Golby, A., Frisken, S., Kapur, T., 2024. ReMIND: the brain resection multimodal imaging database. Scie...

    2024

  11. [19]

    Kuklisova-Murgasova, A

    M. Kuklisova-Murgasova, A. Cifor, R. Napolitano, A. Papageorghiou, G. Quaghebeur, M. A. Rutherford, J. V . Hajnal, J. A. Noble, and J. A. Schnabel. Registration of 3D fetal neurosonography and MRI. Medical Image Analysis, 17(8):1137–1150, 2013.https://10.1016/j.media.2013.07.004

    2013 doi

  12. [20]

    Non-rigid registration of 3D ultrasound for neurosurgery using automatic feature detection and matching

    Machado, I., Toews, M., Luo, J., Unadkat, P., Essayed, W., George, E., Teodoro, P., Carvalho, H., Martins, J., Golland, P., Pieper, S., Frisken, S., Golby, A., Wells, W.M., 2018. Non-rigid registration of 3D ultrasound for neurosurgery using automatic feature detection and mat...

    2018 doi

  13. [21]

    Deformable MRI-ultrasound registration using correlation-based attribute matching for brain shift correction: accuracy and generality in multi-site data

    Machado, I., Toews, M., George, E., Unadkat, P., Essayed, W., Luo, J., Teodoro, P., Carvalho, H., Martins, J., Golland, P., Pieper, S., Frisken, S., Golby, A., Wells, W., Ou, Y ., 2019. Deformable MRI-ultrasound registration using correlation-based attribute matching for brain...

    2019 doi

  14. [22]

    Least squares generative adversarial networks

    Mao, X., Li, Q., Xie, H., Lau, R.Y .K., Wang, Z., Smolley, S.P., 2017. Least squares generative adversarial networks. In: IEEE International Conference on Computer Vision (ICCV). pp. 2794–2802. https://doi.org/ 10.1109/ICCV.2017.304

    2017 doi

  15. [23]

    Fast free-form deformation using graphics processing units

    Modat, M., Ridgway, G.R., Taylor, Z.A., Lehmann, M., Barnes, J., Hawkes, D.J., Fox, N.C., Ourselin, S., 2010. Fast free-form deformation using graphics processing units. Computer Methods and Programs in Biomedicine 98, 278–284.https://doi.org/10.1016/j.cmpb.2009.09.002

    2010 doi

  16. [24]

    Global image registration using a symmetric block-matching approach

    Modat, M., Cash, D.M., Daga, P., Winston, G.P., Duncan, J.S., Ourselin, S., 2014. Global image registration using a symmetric block-matching approach. Journal of Medical Imaging 1, 024003. https://doi.org/10.1117/1. JMI.1.2.024003

    2014 doi

  17. [25]

    A 3D cross-modal keypoint descriptor for MR-US matching and registration.https://arxiv.org/abs/2507.18551

    Morozov, D., Dorent, R., Haouchine, N., 2025. A 3D cross-modal keypoint descriptor for MR-US matching and registration.https://arxiv.org/abs/2507.18551

    2025

  18. [26]

    Serial intraoperative magnetic resonance imaging of brain shift

    Nabavi, A., Black, P.M., Gering, D.T., Westin, C.F., Mehta, V ., Pergolizzi, R.S., Ferrant, M., Warfield, S.K., Hata, N., Schwartz, R.B., Wells, W.M., Kikinis, R., Jolesz, F.A., 2001. Serial intraoperative magnetic resonance imaging of brain shift. Neurosurgery 48, 787–798.htt...

    2001 doi

  19. [27]

    Reconstructing a 3D structure from serial histological sections

    Ourselin, S., Roche, A., Subsol, G., Pennec, X., Ayache, N., 2001. Reconstructing a 3D structure from serial histological sections. Image and Vision Computing 19, 25–31. https://doi.org/10.1016/S0262-8856(00) 00052-4

    2001 doi

  20. [28]

    Unsupervised medical image translation with adversarial diffusion models

    Özbey, M., Dalmaz, O., Dar, S.U.H., Bedel, H.A., Öztürk, ¸ S., Güngör, A., Çukur, T., 2023. Unsupervised medical image translation with adversarial diffusion models. IEEE Transactions on Medical Imaging 42, 3524–3539. https://doi.org/10.1109/TMI.2023.3290149

    2023 doi

  21. [29]

    Influence of high-performance image-to-image translation networks on clinical visual assessment and outcome prediction: utilizing ultrasound to MRI translation in prostate cancer

    Salmanpour, M.R., Mousavi, A., Xu, Y ., Weeks, W.B., Hacihaliloglu, I., 2026. Influence of high-performance image-to-image translation networks on clinical visual assessment and outcome prediction: utilizing ultrasound to MRI translation in prostate cancer. International Journ...

    2026 doi

  22. [30]

    Glioma extent of resection and its impact on patient outcome

    Sanai, N., Berger, M.S., 2008. Glioma extent of resection and its impact on patient outcome. Neurosurgery 62, 753–764.https://doi.org/10.1227/01.neu.0000318159.21731.cf

    2008 doi

  23. [31]

    Navigated ultrasound-based image guidance during resection of gliomas: practical utility in intraoperative decision-making and outcomes

    Shetty, P., Yeole, U., Singh, V ., Moiyadi, A., 2021. Navigated ultrasound-based image guidance during resection of gliomas: practical utility in intraoperative decision-making and outcomes. Neurosurgical Focus 50, E14. https://doi.org/10.3171/2020.10.FOCUS20550

    2021 doi

  24. [32]

    ConvexAdam: self-configuring dual-optimization- based 3D multitask medical image registration

    Siebert, H., Großbröhmer, C., Hansen, L., Heinrich, M.P., 2025. ConvexAdam: self-configuring dual-optimization- based 3D multitask medical image registration. IEEE Transactions on Medical Imaging 44, 738–748. https: //doi.org/10.1109/TMI.2024.3462248

    2025 doi

  25. [33]

    Translation of fetal brain ultrasound images into pseudo-MRI images using artificial intelligence.https://arxiv.org/abs/2504.02408

    Silverstein, N., Beloosesky, R., Leibowitz, E., Azhari, H., 2025. Translation of fetal brain ultrasound images into pseudo-MRI images using artificial intelligence.https://arxiv.org/abs/2504.02408

    2025

  26. [34]

    BrainV oxGen: deep learning framework for synthesis of ultrasound to MRI.https://arxiv.org/abs/2310.08608

    Singh, S., Bewoor, M., Ranapurwala, A., Rai, S., Patil, S., 2023. BrainV oxGen: deep learning framework for synthesis of ultrasound to MRI.https://arxiv.org/abs/2310.08608

    2023

  27. [35]

    Ability of navigated 3D ultrasound to delineate gliomas and metastases: comparison of image interpretations with histopathology

    Unsgård, G., Selbekk, T., Brostrup Müller, T., Ommedal, S., Torp, S.H., Myhr, G., Bang, J., Nagelhus Hernes, T.A., 2005. Ability of navigated 3D ultrasound to delineate gliomas and metastases: comparison of image interpretations with histopathology. Acta Neurochirurgica 147, 1...

    2005

  28. [36]

    Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks

    Wang, G., Li, W., Aertsen, M., Deprest, J., Ourselin, S., Vercauteren, T., 2019. Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks. Neurocomputing 338, 34–45.https://doi.org/10.1016/j.neucom.2019.01.103

    2019 doi

  29. [37]

    Image quality assessment: from error visibility to structural similarity

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P., 2004. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing 13, 600–612. https://doi.org/10.1109/TIP. 2003.819861

    2004 doi

  30. [38]

    High-resolution image synthesis and semantic manipulation with conditional GANs

    Wang, T.C., Liu, M.Y ., Zhu, J.Y ., Tao, A., Kautz, J., Catanzaro, B., 2018. High-resolution image synthesis and semantic manipulation with conditional GANs. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 8798–8807.https://doi.org/10.1109/CVPR.2018.00917

    2018 doi

  31. [39]

    Unsupervised multimodal 3D medical image registration with multilevel correlation balanced optimization

    Wang, J., Chen, X., Zhang, Y ., Liu, M., Zhang, H., 2024. Unsupervised multimodal 3D medical image registration with multilevel correlation balanced optimization. https://arxiv.org/abs/2409.05040. Learn2Reg 2024 challenge submission

    2024

  32. [40]

    Unsupervised MR-US multimodal image registration with multilevel correlation pyramidal optimization.https://arxiv.org/abs/2602.06288

    Wang, J., Liu, Z., Liu, M., Chen, X., Yu, X., Wang, Y ., Zhang, H., 2026. Unsupervised MR-US multimodal image registration with multilevel correlation pyramidal optimization.https://arxiv.org/abs/2602.06288

    2026

  33. [41]

    Global registration of ultrasound to MRI using the LC2 metric for enabling neurosurgical guidance

    Wein, W., Ladikos, A., Fuerst, B., Shah, A., Sharma, K., Navab, N., 2013. Global registration of ultrasound to MRI using the LC2 metric for enabling neurosurgical guidance. In: MICCAI 2013, LNCS 7908. Springer, pp. 34–41.https://doi.org/10.1007/978-3-642-40811-3_5

    2013 doi

  34. [42]

    Evaluation of MRI to ultrasound registration methods for brain shift correction: the CuRIOUS2018 challenge

    Xiao, Y ., Rivaz, H., Chabanas, M., Fortin, M., Machado, I., Ou, Y ., Heinrich, M.P., Schnabel, J.A., Zhong, X., Maier, A., Wein, W., Shams, R., Kadoury, S., Drobny, D., Modat, M., Reinertsen, I., 2020. Evaluation of MRI to ultrasound registration methods for brain shift corre...

    2020 doi

  35. [43]

    The unreasonable effectiveness of deep features as a perceptual metric

    Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O., 2018. The unreasonable effectiveness of deep features as a perceptual metric. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR). pp. 586–595. https://doi.org/10.1109/CVPR.2018.00068. 29

    2018 doi

Pith tools

Reviewed June 28, 2026 · model on record in the stance chip above.