Pith. sign in

REVIEW 3 major objections 5 minor 30 references

Data-Driven Registration and Modeling of Brain Deformation for Image-Guided Neurosurgery: A Systematic Review

T0 review · 3 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read This systematic review of 41 studies argues that deep learning now dominates brain-deformation research, yet classic iterative registration still wins on the hardest multimodal tasks, so the field's real progress lies in hybrid learning-plu

desk verdict Useful field-mapping review of 2020-2025 brain-shift registration, but the central 'classic methods still beat DL on multimodal' claim rests on challenge data the review itself excludes. read the letter →

arxiv 2602.10155 v3 pith:QJECWRKZ submitted 2026-02-09 eess.IV cs.CV

Tiago Assis , Colin P. Galvin , Joshua P. Castillo , Nazim Haouchine , Marta Kersten-Oertel , Zeyu Gao , Mireia Crispin-Ortuzar , Stephen J. Price
show 7 more authors
Thomas Santarius Yangming Ou Sarah Frisken Nuno C. Garcia Alexandra J. Golby Reuben Dorent Ines P. Machado
This is my paper · ORCID
classification eess.IVcs.CV
keywords brainshiftimageregistrationdeeplearningsystematicreviewneurosurgeryintraoperativeultrasounddeformationmodelingmultimodal
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 asks whether the recent explosion in deep learning for neuroimaging actually solves the clinical problem of brain shift—the progressive deformation of brain tissue during tumor surgery. It reviews 41 studies from 2020 to 2025 and concludes that learning-based methods now represent nearly three-quarters of the literature, but this popularity does not translate into superior accuracy. On the most demanding tasks, especially aligning preoperative MRI with intraoperative ultrasound and handling resection cavities, classic optimization-based methods still often perform better. The review's contribution is to organize the field into clear methodological families and to show that hybrid systems—deep learning for speed plus iterative refinement for accuracy—are the most promising direction. This matters because brain shift is a leading source of navigational error in neurosurgery, and the field needs an honest map of what works before clinical deployment.

What carries the argument

The argument is carried by a methodological taxonomy that sorts the 41 included studies into six families—direct displacement-field regression, landmark/feature-based alignment, transformer architectures, synthesis-driven and adversarial multimodal alignment, resection-aware handling of absent correspondences, and biomechanics-informed physics-guided models. On top of this taxonomy, the review rests its strongest claim on a cross-study comparison of results from two recent community benchmark challenges: one for longitudinal pre- to post-resection MRI, where top teams used hybrid learning-plus-optimization pipelines, and one for multimodal MRI-to-ultrasound registration, where the classic bl

What would settle it

A concrete test would be a comprehensive re-run of the search that adds non-English articles and unpublished challenge reports, then recomputes the proportion of deep-learning versus classic studies and re-evaluates the multimodal leaderboard. If the added evidence showed a learning-based method beating the classic baseline on a standardized MRI-to-ultrasound task with statistical significance, or if the publication proportions flipped, the central claim would be refuted. Short of that, a single prospective multi-center study where a deep-learning registration system outperforms classic optimi

Watch

Extended reading notes

Core claim

The central finding of the review is an inversion of the field's apparent progress: although deep learning now accounts for roughly three-quarters of recent publications on data-driven brain deformation registration, direct comparisons on shared benchmark tasks show that classic iterative optimization methods still match or beat learning-based approaches when the going gets hard—particularly in multimodal registration of MRI to intraoperative ultrasound and in the presence of resection cavities where correspondences are missing. The most successful learning systems are not pure deep networks but hybrids that use a network for a fast initial estimate and then refine with case-specific optimiz

Load-bearing premise

The entire assessment rests on the literature search actually capturing the relevant 2020-2025 work; if the English-only, peer-reviewed, four-database query and the decision to exclude unpublished challenge entries missed a meaningful share of the field, the conclusions about dominance and performance inherit that bias.

Editorial extensions

If this is right

  • Clinical adoption should treat deep learning as a speed layer, not an accuracy fix; expect it to provide real-time initial estimates that still need refinement.
  • Researchers should compare new learning-based methods against strong classic baselines on shared multimodal benchmarks; a new method that only competes against other deep networks may show illusory gains.
  • Handling missing correspondences—resection cavities, new lesions—is a first-order problem; masking and consistency-based approaches are currently the most effective response.
  • Uncertainty estimation and error maps should become standard outputs of registration systems before they can support surgical decision-making.
  • The 30-second-to-1-minute update window with sub-3mm accuracy is a realistic target that hybrid models are best positioned to meet.

Reading between the lines

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

  • The dominance of deep learning in publication counts may partly reflect publication bias and the low cost of incremental network tweaks, rather than genuine clinical superiority; a bibliometric analysis of rejected or unpublished challenge entries would test this.
  • If the review's performance assessment is right, a promising testable extension is to feed physics-informed biomechanical losses into the learning process for multimodal MRI-ultrasound, since the gap appears largest when handcrafted intensity metrics fail.
  • The review's findings suggest that prospective clinical studies should measure extent of resection or patient outcomes, not just landmark error, because sub-millimeter registration gains may not translate into surgical benefit.
  • The exclusion of non-English literature and unpublished challenge work is a known limitation; a multilingual or grey-literature search might shift the trend lines, especially for classic methods reported in non-English venues.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This review applies a PRISMA-style protocol to literature published between January 2020 and April 2025 on data-driven compensation of brain deformation for image-guided neurosurgery. The authors report screening 712 records and including 41 studies, which they classify into classic iterative optimization, biomechanical modeling, and deep learning-based registration, with subcategories for direct displacement regression, landmark/feature-based methods, Transformers, synthesis/adversarial methods, missing-correspondence handling, and hybrid frameworks. They also survey datasets, similarity losses, regularization terms, evaluation metrics, uncertainty estimation, validation practices, and two MICCAI challenges (BraTS-Reg and ReMIND2Reg). The central conclusion is that DL methods now dominate the literature, comprising nearly three-quarters of the included studies, but that their advantage is mainly speed and scalability; in demanding multimodal settings, especially preoperative MRI-to-intraoperative ultrasound registration, classic optimization methods such as NiftyReg still often outperform learning-based submissions. The paper closes with recommendations for clinical workflow integration, validation, and future directions.

Significance. If the reporting inconsistencies were corrected, this would be a useful reference for the field. Its strengths are the formal PRISMA protocol, explicit inclusion/exclusion criteria, a coherent methodological taxonomy, detailed tables of datasets and DL-based methods, and a nuanced, critical message about evaluation practices and clinical readiness. The paper does not claim machine-checked proofs or novel technical derivations, and none are needed for a review; the contribution is the synthesis itself. However, the main performance-caveat claim—that classic methods still outperform DL in multimodal registration—is supported by challenge results that the manuscript itself excludes from the reviewed corpus, so the current version does not yet meet the auditability standard expected of a systematic review.

major comments (3)
  1. [Abstract vs. §3.4 / Fig. 3] The number of eligible studies is inconsistent: the arXiv abstract states 46 eligible studies, while the full-text abstract, §3.4, and the PRISMA flow diagram report 41. This is not a cosmetic issue because the headline trend 'DL represents nearly three-quarters of the reviewed studies' depends on the quotient 30/41. The final corpus needs a single, defensible count and a complete enumeration of the included studies.
  2. [§8.2, Table 3, Fig. 4, §10] There is an internal inconsistency in the evidence base for the central performance claim. Fig. 4 explicitly states that ReMIND2Reg challenge papers (2024/2025) remain unpublished and 'were therefore not included in the present review.' Yet §8.2 and Table 3 use exactly those results—NiftyReg baseline TRE 2.87 mm vs. 3.63–4.42 mm for the top submitted teams—as the principal evidence that classic iterative optimization outperforms learning-based methods on preMRI-to-iUS registration, and §10 repeats this conclusion. Either the challenge results are part of the evidence base, in which case the count and eligibility decisions must be revised, or they are outside it, in which case this load-bearing evidence must be removed or clearly separated and replaced with evidence from the 41 included peer-reviewed studies.
  3. [§3.4 / PRISMA reporting] The manuscript does not provide a complete list of the 41 included studies, in the text or in a supplementary table. Without such a list, the category counts (5 classic, 6 biomechanical, 30 DL), the modality-pair statistics, and the per-year trend analysis cannot be audited by the reader. A systematic review should include a full citation list of included studies, ideally with the specific inclusion criteria each study satisfies. This should be added and cross-referenced with the PRISMA flow diagram.
minor comments (5)
  1. [§3.1] The final query is stated, but database-specific query strings, search dates, and the exact number of hits per database are not reported. Since the review follows PRISMA, the search strategy section should include this information or a supplementary appendix.
  2. [§2.4.1, Eq. (1)] The mathematical notation is imprecise: 'space Ω ∈ R^3' should be 'domain Ω ⊂ R^3', and the dimensions D′×H′×W′ in the image definitions are introduced without connecting notation. Please revise for consistency.
  3. [§2.4.2] There are typographical errors, e.g., '(Sotiras et al., 2013)).' contains a double closing parenthesis, and §2.4.3 has an unformatted 'identityI'. Please copyedit throughout.
  4. [Data availability] The statement 'No data was used for the research described in the article' is misleading for a systematic review; the data are the published studies. Rephrase to clarify that no new clinical or experimental data were generated.
  5. [Table 2] The 'State-of-the-Art Accuracy' row cites sources that are challenge reports and preprints, some of which are excluded from the review corpus. Consider labeling this row as 'Reported accuracy in the literature' and ensure the citations are consistent with the inclusion criteria.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: the review's central claims are descriptive statistics and external benchmark results; self-citations and evidence-base inconsistencies do not make the argument circular.

full rationale

This paper is a systematic review, not a derivation: it contains no equations that map inputs to predictions, no fitted parameters, and no claimed first-principles result. The central claims—DL accounts for 30/41 studies (~73%) and classic iterative methods still outperform learning-based methods on some multimodal tasks—are descriptive statistics and externally benchmarked challenge outcomes, respectively. The DL-dominance count follows directly from the PRISMA selection (Section 3.4) rather than from any self-referential definition. The performance caveat is supported by Table 3's ReMIND2Reg and BraTS-Reg results, which were produced by independent challenge evaluations, not by the review's own fitting procedure. Several cited works share authors with the present paper (Machado et al., Haouchine et al., Dorent et al., Assis et al., ReMIND), but none of these self-citations functions as a load-bearing derivation: the review does not invoke a self-authored uniqueness theorem or adopt an ansatz solely by self-citation. The manuscript does contain internal inconsistencies—the abstract reports 46 eligible studies while the full text and PRISMA flow state 41, and Fig. 4 says unpublished ReMIND2Reg challenge papers were excluded while Section 8.2 and Table 3 rely on those results. These are auditability/evidence-base concerns, not circularity: the claims do not reduce to their inputs by construction. Accordingly, no circular step is identified.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

As a review, the central claims rest on assumptions about search completeness, representativeness, and trustworthiness of primary-study reports rather than on mathematical axioms or fitted parameters.

assumptions (3)
  • domain assumption The 41 included studies are representative of data-driven brain deformation methods published 2020-2025
    Search relied on four databases, English-only, peer-reviewed venues (Section 3.1-3.3); unpublished challenge entries were excluded (Section 8.2). If the corpus is incomplete, the trend and gap claims fail.
  • domain assumption Reported evaluation metrics in primary studies are accurate and comparable
    The review synthesizes reported TRE/Dice/Jacobian values without re-running experiments; between-study comparability is assumed (Sections 7 and Table 3).
  • domain assumption PRISMA-style narrative synthesis is an adequate method for this review's claims
    The review applies PRISMA reporting and qualitative categorization, but does not perform quantitative meta-analysis or risk-of-bias assessment, so its conclusions are interpretive summaries rather than pooled evidence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Data-Driven Registration and Modeling of Brain Deformation for Image-Guided Neurosurgery: A Systematic Review." pith.science (2026). https://pith.science/paper/QJECWRKZ

@misc{pith2026260210155,
  author       = {Pith},
  title        = {Pith review of: Data-Driven Registration and Modeling of Brain Deformation for Image-Guided Neurosurgery: A Systematic Review},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QJECWRKZ}},
  note         = {Machine review of arXiv:2602.10155}
}
read the original abstract

Accurate compensation of brain deformation is critical for reliable image-guided neurosurgery. Surgical manipulation and tumor resection induce tissue motion, causing preoperative planning images to become misaligned with the intraoperative anatomy. In this systematic review, we examine data-driven methods developed between 2020 and 2025 for brain deformation registration and modeling, with a particular focus on learning-based approaches. A comprehensive literature search was conducted in PubMed, IEEE Xplore, Scopus, and Web of Science using predefined inclusion and exclusion criteria for computational methods addressing brain deformation in neurosurgical imaging, resulting in 46 eligible studies. We provide a unified analysis of methodological strategies, including deep learning-based image registration, direct deformation field regression, synthesis-driven multimodal alignment, resection-aware architectures for handling missing correspondences, and hybrid models integrating biomechanical priors. We also examine dataset utilization, evaluation metrics, validation protocols, and the assessment of uncertainty and generalization across studies. While learning-based methods demonstrate promising accuracy and computational efficiency, current approaches remain limited by out-of-distribution robustness, standardized benchmarking, interpretability, and readiness for clinical deployment. Our review highlights these gaps and outlines future directions toward more robust, generalizable, and clinically translatable solutions for neurosurgical guidance. By organizing recent advances and critically assessing evaluation practices, this work provides a comprehensive reference for researchers and clinicians working on data-driven registration and modeling of brain deformation.

Figures

Figures reproduced from arXiv: 2602.10155 by the authors.

Figure 1
Figure 1. Neurosurgical workflow from preoperative assessment through postoperative care, including intraoperative phases. Green stars indicate the same location [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Comparison between classic instance optimization and deep learning-based medical image registration frameworks. (A) Classic methods optimize the [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Diagram illustrating the literature search process. (Top) A total of 712 records were first identified through database searches in PubMed, IEEE Xplore, [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Summary of key trends across studies meeting inclusion criteria. (A) Number of publications per year from January 2020 to *April 2025, comparing all [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Geographic distribution of the reviewed publications. Darker shad [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Representative deep learning frameworks for medical image registration. Blue arrows show the inputs to the objective function that guides backpropa [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 7
Figure 7. Figure 7: Overview of the biomechanical modeling workflow for image-guided [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 4 canonical work pages

  1. [9]

    Talbot, H., Haouchine, N., Peterlik, I., Dequidt, J., Duriez, C., Delingette, H., Cotin, S., 2015

    doi:10.1109/cvpr.2015.7298594. Talbot, H., Haouchine, N., Peterlik, I., Dequidt, J., Duriez, C., Delingette, H., Cotin, S., 2015. Surgery training, planning and guidance using the SOFA framework, in: Eurographics. doi:10.1016/b978-0-12-415873-3.00034-1. Taleb, A., Guigou, C., Leclerc, S., Lalande, A., Bozorg Grayeli, A., 2023. Image-to-patient registratio...

    arXiv 2015

  2. [10]

    Employing ConvexAdam for BraTS-Reg, in: Inter- national MICCAI Brainlesion Workshop, Springer. pp. 252–

  3. [16]

    Frisken, S., Luo, M., Juvekar, P., Bunevicius, A., Machado, I., Unadkat, P., Bertotti, M.M., Toews, M., Wells, W.M., Miga, M.I., et al., 2020

    doi:10.1109/tmrb.2021.3122357. Frisken, S., Luo, M., Juvekar, P., Bunevicius, A., Machado, I., Unadkat, P., Bertotti, M.M., Toews, M., Wells, W.M., Miga, M.I., et al., 2020. A comparison of thin-plate spline deformation and finite element modeling to compensate for brain shift during tumor resection. International Journal of Computer Assisted Radiology an...

    arXiv 2021

  4. [17]

    Applied Sciences 12, 2828

    2D/3D multimode medical image registration based on normalized cross-correlation. Applied Sciences 12, 2828. doi:10.3390/app12062828. Liu, Y ., Li, X., Li, R., Huang, S., Yang, X., 2023. A multi-view assisted registration network for MRI registration pre-and post-therapy. Medical & Biological Engineering & Comput- ing 61, 3181–3191. doi:10.1007/s11517-023...

    arXiv 2023

  5. [20]

    International Journal of Computer As- sisted Radiology and Surgery 18, 501–508

    Robust landmark-based brain shift correction with 27 a siamese neural network in ultrasound-guided brain tu- mor resection. International Journal of Computer As- sisted Radiology and Surgery 18, 501–508. doi:10.1007/ s11548-022-02770-5. Puchalski, R.B., Shah, N., Miller, J., Dalley, R., Nomura, S.R., Yoon, J.G., Smith, K.A., Lankerovich, M., Bertagnolli, ...

    arXiv 2018

  6. [23]

    Frontiers in Neuroscience 16, 883584

    Navigated intraoperative 3D ultrasound in glioblas- toma surgery: analysis of imaging features and impact on extent of resection. Frontiers in Neuroscience 16, 883584. doi:10.3389/fnins.2022.883584. Scarpace, L., Mikkelsen, T., Cha, S., Rao, S., Tekchandani, S., Gutman, D., Saltz, J.H., Erickson, B.J., Pedano, N., Flanders, A.E., et al., 2016. The Cancer ...

    arXiv 2022

  7. [24]

    PULPo: Probabilistic unsupervised laplacian pyramid registration, in: International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer. pp. 717–727. doi:10.1007/978-3-031-72069-7_67. Sokooti, H., De V os, B., Berendsen, F., Lelieveldt, B.P., Išgum, I., Staring, M., 2017. Nonrigid image registra- tion using multi-scale 3D co...

    arXiv 2017

  8. [26]

    18084–18094

    GradICON: Approximate diffeomorphisms via gradi- ent inverse consistency, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 18084–18094. doi:10.1109/cvpr52729.2023.01734. Tonutti, M., Gras, G., Yang, G.Z., 2017. A machine learn- ing approach for real-time modelling of tissue deforma- tion in image-guided neurosurge...

    arXiv 2023

Show all 30 references
  1. [28]

    Computers in Biology and Medicine 143, 105271

    Automatic framework for patient-specific modelling of tumour resection-induced brain shift. Computers in Biology and Medicine 143, 105271. doi:10.1016/j.compbiomed. 2022.105271. Zaffino, P., Moccia, S., De Momi, E., Spadea, M.F., 2020. A re- view on advances in intra-operative...

    2022

  2. [29]

    IEEE Access 9, 147579–147590

    iRegNet: Non-rigid registration of MRI to inter- ventional US for brain-shift compensation using convolu- tional neural networks. IEEE Access 9, 147579–147590. doi:10.1109/access.2021.3120306. Zeineldin, R.A., Karar, M.E., Mathis-Ullrich, F., Burgert, O.,

    2021

  3. [30]

    Self-supervised iRegNet for the registration of lon- gitudinal brain MRI of diffuse glioma patients, in: Interna- tional MICCAI Brainlesion Workshop, Springer. pp. 25–34. doi:10.1007/978-3-031-44153-0_3. Zhang, J., Xie, X., Cheng, X., Li, T., Zhong, J., Hu, X., Sun, L., Yan, H...

    2024

  4. [41]

    Avants, B.B., Tustison, N.J., Song, G., Cook, P.A., Klein, A., Gee, J.C., 2011

    doi:10.1016/j.media.2007.06.004. Avants, B.B., Tustison, N.J., Song, G., Cook, P.A., Klein, A., Gee, J.C., 2011. A reproducible evaluation of ANTs similar- ity metric performance in brain image registration. NeuroIm- age 54, 2033–2044. doi:10.1016/j.neuroimage.2010. 09.025. Az...

    2007

  5. [154]

    Safdar, S., Zwick, B.F., Yu, Y ., Bourantas, G.C., Joldes, G.R., Warfield, S.K., Hyde, D.E., Frisken, S., Kapur, T., Kiki- nis, R., et al., 2023

    doi:10.1007/978-1-4419-8204-9_9. Safdar, S., Zwick, B.F., Yu, Y ., Bourantas, G.C., Joldes, G.R., Warfield, S.K., Hyde, D.E., Frisken, S., Kapur, T., Kiki- nis, R., et al., 2023. SlicerCBM: automatic framework for biomechanical analysis of the brain. International Journal of C...

    2023

  6. [261]

    Ha, I.Y ., Heinrich, M.P., 2021

    doi:10.1007/978-3-031-33842-7_22. Ha, I.Y ., Heinrich, M.P., 2021. Modality-agnostic self- supervised deep feature learning and fast instance optimi- sation for multimodal fusion in ultrasound-guided interven- tions. Computer Methods and Programs in Biomedicine 211, 106374. do...

    2021

  7. [272]

    Canalini, L., Klein, J., Miller, D., Kikinis, R., 2020

    doi:10.1007/978-3-031-33842-7_23. Canalini, L., Klein, J., Miller, D., Kikinis, R., 2020. Enhanced registration of ultrasound volumes by segmentation of resec- tion cavity in neurosurgical procedures. International Jour- nal of Computer Assisted Radiology and Surgery 15, 1963–

    2020 doi

  8. [502]

    Peng, Y ., Liu, X., Wang, A., Han, R., 2014

    doi:10.14444/8780. Peng, Y ., Liu, X., Wang, A., Han, R., 2014. The effect of mannitol on intraoperative brain relaxation in patients un- dergoing supratentorial tumor surgery: study protocol for a randomized controlled trial. Trials 15, 165. doi:10.1186/ 1745-6215-15-165. Pfe...

    2014 doi

  9. [522]

    26 Miller, K., Chinzei, K., Orssengo, G., Bednarz, P., 2000

    doi:10.1007/s11548-010-0535-3. 26 Miller, K., Chinzei, K., Orssengo, G., Bednarz, P., 2000. Me- chanical properties of brain tissue in-vivo: experiment and computer simulation. Journal of Biomechanics 33, 1369–

    2000 doi

  10. [1317]

    Haouchine, N., Juvekar, P., Wells III, W.M., Cotin, S., Golby, A., Frisken, S., 2020

    doi:10.1109/tbme.2021.3113841. Haouchine, N., Juvekar, P., Wells III, W.M., Cotin, S., Golby, A., Frisken, S., 2020. Deformation aware augmented reality for craniotomy using 3D/2D non-rigid registration of corti- cal vessels, in: International Conference on Medical Image Compu...

    2021

  11. [1376]

    Miller, K., Joldes, G.R., Bourantas, G., Warfield, S.K., Hyde, D.E., Kikinis, R., Wittek, A., 2019

    doi:10.1016/s0021-9290(00)00120-2. Miller, K., Joldes, G.R., Bourantas, G., Warfield, S.K., Hyde, D.E., Kikinis, R., Wittek, A., 2019. Biomechanical modeling and computer simulation of the brain during neurosurgery. International journal for numerical methods in biomedical eng...

    2019

  12. [1435]

    Heinrich, M.P., Jenkinson, M., Papie˙ z, B.W., Brady, S.M., Schnabel, J.A., 2013

    doi:10.1016/j.media.2012.05.008. Heinrich, M.P., Jenkinson, M., Papie˙ z, B.W., Brady, S.M., Schnabel, J.A., 2013. Towards realtime multimodal fu- sion for image-guided interventions using self-similarities, in: International Conference on Medical Image Computing and Computer-...

    2012

  13. [1974]

    Chel, H., Bora, P., Ramchiary, K., 2023

    doi:10.1007/s11548-020-02273-1. Chel, H., Bora, P., Ramchiary, K., 2023. Segmentation- based registration of pre and intra-operative brain ultra- sound images using particle swarm optimization. Research on Biomedical Engineering 39, 805–818. doi:10.1007/ s42600-023-00304-9. Ch...

    2023

  14. [2003]

    IEEE Transactions on Medical Imaging 22, 730–741

    V olume-preserving nonrigid registration of MR breast images using free-form deformation with an incompressibil- ity constraint. IEEE Transactions on Medical Imaging 22, 730–741. doi:10.1109/tmi.2003.814791. Ronchetti, M., Wein, W., Navab, N., Zettinig, O., Pre- vost, R., 2023...

    2003

  15. [2017]

    Medical Physics 44, 3875–3882

    REtroSpective Evaluation of Cerebral Tumors (RE- SECT): A clinical database of pre-operative mri and intra- operative ultrasound in low-grade glioma surgeries. Medical Physics 44, 3875–3882. doi:https://doi.org/10.1002/ mp.12268. Xiao, Y ., Rivaz, H., Chabanas, M., Fortin, M.,...

    2019

  16. [2019]

    (Ed.), Biomechanics of the Brain

    Intra-operative measurement of brain deformation, in: Miller, K. (Ed.), Biomechanics of the Brain. 2 ed.. Springer Cham, pp. 303–319. doi:https://doi.org/10. 1007/978-3-030-04996-6. Fuerst, B., Wein, W., Müller, M., Navab, N., 2014. Automatic ultrasound–MRI registration for ne...

    2014

  17. [2021]

    Frontiers in Oncology 11, 656519

    Challenges and opportunities of intraoperative 3D ul- trasound with neuronavigation in relation to intraoperative MRI. Frontiers in Oncology 11, 656519. doi:10.3389/ fonc.2021.656519. Bay, H., Tuytelaars, T., Gool, L.V ., 2006. SURF: Speeded Up Robust Features, in: Leonardis, ...

    2021

  18. [2022]

    Neurosurgical Review 45, 2503–2515

    Intraoperative ultrasound in brain tumor surgery: A review and implementation guide. Neurosurgical Review 45, 2503–2515. doi:10.1007/s10143-022-01778-4. Dorent, R., Haouchine, N., Golby, A., Frisken, S., Kapur, T., Wells, W., 2025a. Unified cross-modal medical image syn- thesi...

    2025

  19. [2023]

    IEEE Transactions on Ultra- sonics, Ferroelectrics, and Frequency Control 70, 999–1015

    Toward estimating MRI-ultrasound registration error in image-guided neurosurgery. IEEE Transactions on Ultra- sonics, Ferroelectrics, and Frequency Control 70, 999–1015. doi:10.1109/tuffc.2023.3239320. Bilger, A., Dequidt, J., Duriez, C., Cotin, S., 2011. Biome- chanical simul...

    2023

  20. [2024]

    Scientific Data 11, 494

    ReMIND: The Brain Resection Multimodal Imag- ing Database. Scientific Data 11, 494. doi:10.1038/ s41597-024-03295-z. Juvekar, P., Torio, E., Bi, W.L., Bastos, D.C.D.A., Golby, A.J., Frisken, S.F., 2023. Mapping resection progress by tool- tip tracking during brain tumor surger...

    2023 arXiv

  21. [2555]

    Liu, J., Cao, S., Imbach, K.J., Gritsenko, M.A., Lih, T.S.M., Kyle, J.E., Yaron-Barir, T.M., Binder, Z.A., Li, Y ., Strunilin, I., et al., 2024

    doi:10.1109/iccv.2011.6126542. Liu, J., Cao, S., Imbach, K.J., Gritsenko, M.A., Lih, T.S.M., Kyle, J.E., Yaron-Barir, T.M., Binder, Z.A., Li, Y ., Strunilin, I., et al., 2024. Multi-scale signaling and tumor evolution in high-grade gliomas. Cancer Cell 42, 1217–1238. doi:10. 1...

    2011

  22. [6532]

    Bergström, P., Edlund, O., 2014

    doi:10.1002/mp.17317. Bergström, P., Edlund, O., 2014. Robust registration of point sets using iteratively reweighted least squares. Computa- tional Optimization and Applications 58, 543–561. doi:10. 1007/s10589-014-9643-2. Bierbrier, J., Eskandari, M., Di Giovanni, D.A., Coll...

    2014 doi

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.