Pith. sign in

REVIEW 3 major objections 4 minor 35 references

Toward Reliable AR-Guided Surgical Navigation: Interactive Deformation Modeling with Data-Driven Biomechanics and Prompts

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A hybrid neural–biomechanical solver matches finite-element deformation accuracy while running about 2.5 times faster, and surgeon-drawn prompts lower registration error further.

desk verdict Solid hybrid method with a clean one-pass regularization trick, but the phantom SOTA claim needs a fair interactive baseline and a significance test. read the letter →

arxiv 2506.08048 v2 pith:LMBWFCDM submitted 2025-06-08 cs.CV cs.AIcs.RO

Zheng Han , Jun Zhou , Jialun Pei , Jing Qin , Yingfang Fan , Qi Dou This is my paper
classification cs.CVcs.AIcs.RO
keywords augmentedrealitysurgicalnavigationdeformationmodelingnon-rigidregistrationbiomechanicalmodelphysics-informedneuralnetworkhuman-in-the-looptargeterror
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

AR-guided surgery needs a preoperative organ model deformed to match the live anatomy. This paper claims that a hybrid pipeline—a per-instance physics-integrated neural network followed by a one-pass biomechanical solve—can match finite-element deformation accuracy while running about 2.5 times faster, and that letting a surgeon draw corrective prompts on the misaligned surface improves accuracy further. On a public liver-phantom benchmark the automated pipeline reaches 3.42 mm mean target registration error, and prompted interactions reduce this to 2.78 mm, below all prior methods listed in the comparison. If correct, this makes physically plausible deformation modeling fast enough for intraoperative use and gives surgeons a principled way to fix correspondence errors that automated methods cannot resolve.

What carries the argument

The load-bearing object is the finite-element stiffness matrix $K$ assembled on a tetrahedral mesh of the organ, used twice in different roles. Inside BiomPINN, $K$ appears as a quadratic strain-energy regularizer (via the interpolation matrix $\Phi$ that maps surface displacements to volumetric ones), penalizing locally inconsistent boundary motion; in the PBM stage, the same $K$ is the elastic term of a Tikhonov-regularized linear system, $(\Phi^T P \Phi + \beta \sigma^2 K) u_\Omega = \Phi^T P b$, whose data weight $\sigma^2$ is computed directly from BiomPINN's residual rather than by iterative expectation-maximization. This single-pass formulation is what replaces iterative regularization tuning, and the correspondence-refinement loop—prompt lines expanded to local patches, aligned by iterative closest point, and re-matched by mutual nearest neighbors—is what turns surgeon feedback into a biomechanically propagated correction.

What would settle it

Measure target registration error on the phantom dataset while artificially adding a retractor-shaped patch of points to the intraoperative cloud; if the embedded-bead TRE degrades by more than the roughly 0.6 mm gain that prompts provide, then the clean-surface assumption is the binding limit on real-surgery transfer.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central claim is that the expensive iterative tuning of a patient-specific biomechanical model can be replaced by a direct prediction from a physics-regularized neural network, with no loss of accuracy. BiomPINN is a four-level MLP deformation pyramid optimized per instance; its displacement field is projected onto a tetrahedral finite-element mesh, and the FE stiffness matrix supplies a strain-energy regularizer at each level. The resulting surface displacement is then fed into the PBM as a residual-based estimate of the data weight, so the volumetric deformation is obtained by solving a single sparse linear system instead of repeatedly re-solving it. On synthetic liver, kidney, and prostate data, the method matches the GMM-FEM baseline's mean target registration error (2.47, 3.40, and 0.95 mm) at runtimes of 1.65, 1.25, and 1.50 seconds versus 4.28, 3.18, and 3.38 seconds. The interactive layer treats surgeon line prompts as corrections to the correspondence matrix, re-optimizes the deformation incrementally, and on the phantom challenge improves TRE from 3.42 mm to 2.78 mm, surpassing all compared prior methods.

Load-bearing premise

The pipeline assumes the intraoperative point cloud is a clean, accurate capture of the organ surface, without significant contamination from surgical instruments or surrounding tissue; if that fails, the correspondences driving every correction are computed against the wrong geometry.

Editorial extensions

If this is right

  • Any deformation case processed by BiomPINN-PBMs avoids offline training, so accuracy does not depend on how well synthetic training data match the operating room.
  • The same pipeline runs in under two seconds per case for liver, kidney, and prostate test cases, making biomechanical correction compatible with intraoperative decision points.
  • Prompt-based corrections are local updates to the correspondence matrix, so repeated rounds of surgeon guidance accumulate as a chain of incremental deformation fields rather than requiring a full restart.
  • Retaining the mesh topology after deformation lets preoperative surgical plans be carried onto the intraoperative anatomy in the same coordinate frame.

Reading between the lines

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

  • The annotation phase, not the solver, is the end-to-end bottleneck of a correction cycle, so replacing line-drawing with touchscreen or semi-automated prompts could make the interactive loop viable for continuous, not just on-demand, navigation.
  • Because the prompt mechanism only edits correspondences, it is a general repair operator: the same interface could be bolted onto any correspondence-driven non-rigid registration algorithm, not just the proposed hybrid.
  • If the residual-derived data weight estimate generalizes, any finite-element registration that currently re-estimates its regularization weight by outer-loop iteration could adopt a one-pass solve.
  • The reported mid-air annotation error suggests prompt accuracy, not model capacity, may become the next limiting factor; explicitly visualizing inferred correspondences during annotation could test that directly.
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 / 4 minor

Summary. The manuscript proposes a hierarchical per-instance MLP deformation network (BiomPINN) regularized by a finite-element strain-energy term, integrates it with a patient-specific biomechanical model (BiomPINN-PBMs) through the linear system in Eqs. (21)-(24), and adds an interactive AR framework in which surgeons draw line prompts to correct correspondences and re-optimize the deformation. Experiments on a synthetic liver/kidney/prostate dataset show that BiomPINN-PBMs achieves mean TRE statistically equivalent to GMM-FEM while reducing runtime, and experiments on the public Image-to-Physical Liver Registration Sparse Data Challenge phantom dataset report a mean TRE of 3.42 mm without prompts and 2.78 mm with prompts, which the authors claim surpasses all listed prior methods. The paper also reports qualitative in-vivo laparoscopic cases and an end-to-end timing analysis of the interactive loop.

Significance. The synthetic validation is a genuine strength: the paired statistical comparison with GMM-FEM, the runtime measurements, and the Jacobian-determinant field-consistency analysis are concrete and reproducible, and the use of a public challenge dataset with blinded targets is appropriate for benchmarking. The interactive prompt mechanism is novel and clinically motivated, and the timing breakdown of the annotation loop is useful for assessing intraoperative feasibility. However, the headline claim of surpassing state-of-the-art volumetric accuracy currently rests on a non-comparable interactive condition, and the fully automatic variant is not competitive with several prior methods. If the authors add a proper interactive baseline or reframe the claim, the paper would be a solid contribution to non-rigid registration for surgical navigation; as written, the significance is contingent on that additional evidence.

major comments (3)
  1. [Section V-B, Table V] The claim of surpassing state-of-the-art is not supported by the reported comparison. In the fully automatic setting, BiomPINN-PBMs (3.42 ± 0.72 mm) is worse than Heiselman et al. (3.08 ± 0.85 mm), Mestdagh et al. (3.31 ± 1.86 mm), and Yang et al. (2.93 ± 0.68 mm). The only condition in which the method beats prior work is the 'w/ prompt' condition, where no prior method is given the same interactive correction mechanism. Moreover, the improvement over the best prior method is 0.15 mm (2.78 vs. 2.93 mm), which is well within the reported standard deviations, and no paired significance test or per-case analysis is provided for the phantom data. The abstract's statement that the method surpasses state-of-the-art in volumetric accuracy is therefore not an apples-to-apples comparison.
  2. [Section V-B and Section IV-C] The interactive prompt protocol is underspecified, making the 'w/ prompt' result difficult to interpret. The paper does not report the number of prompts per case, the number of interaction iterations, the time spent per case, or the inter-user variability, and the annotators were computer science and biomedical engineering students rather than surgeons. Without a controlled interactive baseline—for example, applying the same line-prompt correction procedure to Yang et al. or to a simple ICP-based correspondence update—the measured improvement cannot be attributed specifically to the proposed framework rather than to the general benefit of additional user-provided correspondences.
  3. [Section III-C, Eq. (22)] The residual-based estimate of σ² in Eq. (22) uses the same surface target that appears in the data term of Eq. (21) and is computed from the BiomPINN prediction, but the paper provides no sensitivity analysis for σ² or for the Tikhonov weight β, and β is never reported anywhere in the experiments. Since the linear system in Eq. (24) depends on the product βσ², the numerical results are not fully reproducible, and the claim that the approach avoids iterative regularization tuning should be accompanied by evidence that the results are stable over reasonable ranges of these parameters.
minor comments (4)
  1. [Section III-C] There is a typo in the sentence following Eq. (22): 'Taking the derivative of the derivative of E(u_Ω)' should read 'Taking the derivative of E(u_Ω)'. Additionally, in the definition of C̃, the phrase 'is is the Kronecker product' contains a duplicated 'is'.
  2. [Section V-A, Table III] The runtime comparison is asymmetric: GMM-FEM runtime is reported as only the linear-system solve time, whereas BiomPINN-PBMs runtime includes the full per-instance optimization and biomechanical solve. The text claims the setup makes runtimes directly comparable, but the reporting should clarify whether both methods exclude or include correspondence estimation and iterative regularization tuning, since this affects the interpretation of the speedup.
  3. [Section IV-C] The interactive experiments were performed by two students rather than clinicians. The paper should either use the term 'users' instead of 'surgeons' in the experimental protocol or justify why the student annotations are representative of surgical expertise.
  4. [Section V-C] The human in-vivo evaluation relies solely on a one-sided Chamfer distance to the reconstructed surface and on qualitative inspection; no comparison with existing non-rigid registration methods is provided. The authors should state more explicitly that these results are feasibility demonstrations rather than quantitative performance claims.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: Eq. (22) is a residual-based heuristic rather than a fitted parameter; phantom TRE is evaluated on independent internal fiducials; self-citations are not load-bearing.

full rationale

Walking the derivation chain: BiomPINN optimizes Eq. (17) with the alignment loss Eq. (18) against the intraoperative surface Y. BiomPINN-PBMs then computes sigma^2 in Eq. (22) from the BiomPINN-projected residual and solves Eq. (24), a standard Tikhonov-regularized data-fit problem. This is a one-step empirical residual weighting, not a circular reduction: sigma^2 is not fitted to TRE or to the final deformation, and the final u_Omega is obtained by solving a linear system rather than by plugging in the fitted constant. The PBM's elastic energy and the forward-propagation step reuse the same biomechanical model, which is self-consistency rather than circularity. On the phantom benchmark, the reported TRE is measured at 159 internal stainless-steel beads that are not used in the optimization, so the evaluation target is independent of the surface data term. The interactive 'w/ prompt' comparison in Table V is not apples-to-apples with non-interactive baselines, and the 0.15 mm improvement over Yang et al. lacks a significance test, but that is a fairness and statistical-support issue, not a circularity issue. The self-citations ([19], [27]) motivate the interactive framework and provide the in-vivo dataset; neither carries the deformation derivation. No equation in the paper defines its output from a fitted constant by construction, and no load-bearing claim reduces to a self-citation chain.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central contribution is a method, so the ledger is mostly modeling choices rather than invented entities. The most load-bearing items are the linear-elastic FE assumption, the cleanliness of the surface point cloud, and the residual-based data variance estimate. No new physical entities are introduced.

free parameters (3)
  • lambda_1, lambda_2 (per-organ regularization weights) = not reported; tuned in [1e-6, 1e-3] per organ
    Equation (17) uses lambda_1 L_rigid plus lambda_2 L_fem; tuning per organ means part of the reported performance may come from organ-specific fitting rather than the method itself. Section IV-C.
  • beta (Tikhonov weight in PBM energy) = not reported
    Equations (21) and (24) depend on beta for the balance between biomechanical energy and the data term, but no value or tuning procedure is given. Without it, the one-pass solution is not fully specified.
  • sigma^2 (residual-based data variance) = computed per case from Equation (22)
    Replaces iterative EM tuning of the regularization weight; it is derived from the BiomPINN prediction residual, so it is data-dependent rather than an independent calibration. This is not necessarily a flaw, but it is an extra modeling choice that affects accuracy.
assumptions (4)
  • domain assumption Linear elastic constitutive model adequately represents organ deformation for the PBM.
    Section II-B uses Equations (4)-(6) with linear elasticity, while the synthetic ground truth is generated with a neo-Hookean hyperelastic model. If intraoperative deformations are large, the linear assumption may limit accuracy.
  • domain assumption Sparse surface correspondences are reliable enough to drive whole-volume deformation.
    Section II-D describes correspondences via UTOPIC, GraphSCNet, or mutual nearest neighbors. Large morphological changes such as falciform ligament dissection can break these correspondences, which the paper itself discusses in the introduction.
  • domain assumption The intraoperative point cloud is a clean and accurate capture of the true organ surface.
    Section VI-A explicitly states that point clouds should not be significantly contaminated by surrounding tissues or instruments, which is often not true in real surgical settings.
  • ad hoc to paper The data fidelity can be modeled as Gaussian noise with variance sigma^2 estimated from the BiomPINN residual.
    Equations (21)-(22) assume this noise model and use the residual heuristic to set sigma^2. No validation is given that this heuristic yields a calibrated variance, and it is central to the one-pass claim.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Toward Reliable AR-Guided Surgical Navigation: Interactive Deformation Modeling with Data-Driven Biomechanics and Prompts." pith.science (2026). https://pith.science/paper/LMBWFCDM

@misc{pith2026250608048,
  author       = {Pith},
  title        = {Pith review of: Toward Reliable AR-Guided Surgical Navigation: Interactive Deformation Modeling with Data-Driven Biomechanics and Prompts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LMBWFCDM}},
  note         = {Machine review of arXiv:2506.08048}
}
read the original abstract

In augmented reality (AR)-guided surgical navigation, preoperative organ models are superimposed onto the patient's intraoperative anatomy to visualize critical structures such as vessels and tumors. Accurate deformation modeling is essential to maintain the reliability of AR overlays by ensuring alignment between preoperative models and the dynamically changing anatomy. Although the finite element method (FEM) offers physically plausible modeling, its high computational cost limits intraoperative applicability. Moreover, existing algorithms often fail to handle large anatomical changes, such as those induced by pneumoperitoneum or ligament dissection, leading to inaccurate anatomical correspondences and compromised AR guidance. To address these challenges, we propose a data-driven biomechanics algorithm that preserves FEM-level accuracy while improving computational efficiency. In addition, we introduce a novel human-in-the-loop mechanism into the deformation modeling process. This enables surgeons to interactively provide prompts to correct anatomical misalignments, thereby incorporating clinical expertise and allowing the model to adapt dynamically to complex surgical scenarios. Experiments on a publicly available dataset demonstrate that our algorithm achieves a mean target registration error of 3.42 mm. Incorporating surgeon prompts through the interactive framework further reduces the error to 2.78 mm, surpassing state-of-the-art methods in volumetric accuracy. These results highlight the ability of our framework to deliver efficient and accurate deformation modeling while enhancing surgeon-algorithm collaboration, paving the way for safer and more reliable computer-assisted surgeries.

Figures

Figures reproduced from arXiv: 2506.08048 by the authors.

Figure 1
Figure 1. Interactive deformation modeling for AR-guided surgical navi￾gation. The proposed data-driven biomechanics algorithm (BiomPINN￾PBMs) deforms the preoperative 3D model based on intraoperative imaging and presents the result for surgeon evaluation. This human￾in-the-loop process allows the surgeon to either accept the deforma￾tion for AR overlay or provide corrective prompts to further refine the model. A video demons… view at source ↗
Figure 2
Figure 2. Illustration of the surface-based deformation modeling process. After rigid registration, sparse correspondences with the intraoperative surface are established (left). These local displacements are biome￾chanically propagated to estimate the full volumetric deformation field, including unobserved regions (right). To overcome these issues, more recent works have adopted flexible geometric representations, such as po… view at source ↗
Figure 3
Figure 3. Illustration of the interpolation matrix Φ that maps volumetric displacements to boundary displacements. The tetrahedral mesh Ω is partitioned into boundary nodes X ⊂ ∂Ω and interior nodes, yielding a block matrix structure. A cross-sectional view reveals the internal mesh structure, while the surface mesh ∂Ω retains only the boundary points. denoted as UΩ = {uΩ,i ∈ R 3 | i = 1, . . . , nv}, can be obtained by solvi… view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Overview of the proposed interactive deformation modeling framework for AR-guided surgical navigation. The process consists of four sequential steps: (1) input of the preoperative model and intraoperative point cloud with initial coarse matching; (2) deformation estima…
Figure 5
Figure 5. Figure 5: Interactive deformation modeling workflow. User-annotated misaligned regions are expanded into local surface patches, aligned via rigid ICP, and used to update the correspondence matrix via mutual nearest neighbor matching. The deformation field is then re-optimized an…
Figure 6
Figure 6. Figure 6: (a). The beads were localized through CT scans for each deformation state and paired with the corresponding deformed surface. The evaluated algorithms were tasked with estimating the volumetric deformation field of the preoperative organ based on sparse intraoperative …
Figure 7
Figure 7. Figure 7: Performance comparison on the synthetic dataset across liver, kidney, and prostate. The plots show how average target registration error varies with geodesic distance from observed regions, reflecting each method’s ability to propagate deformation to unobserved areas …
Figure 8
Figure 8. Figure 8: Field consistency assessment: Jacobian determinant heatmap. TABLE IV TRE (MEAN ± STD) [MM] UNDER DIFFERENT POINT CLOUD SAMPLING STRATEGIES ON THE PHANTOM DATASET. Methods Swabbed Surface Digitized Surface Rigid Alignment 5.70 ± 1.96 5.06 ± 1.10 Ours (w/o prompt) 3.45 ±…
Figure 10
Figure 10. Figure 10: Qualitative and quantitative results on laparoscopic liver surgery cases. Each row corresponds to one patient, showing two different views (View 1 and View 2) of the surface alignment outcomes. For each patient, we compare rigid alignment, BiomPINN-PBMs without prompt…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    Utility of optical see-through head mounted displays in augmented reality-assisted surgery: A systematic review,

    M. Birlo, P. E. Edwards, M. Clarkson, and D. Stoyanov, “Utility of optical see-through head mounted displays in augmented reality-assisted surgery: A systematic review,”Medical Image Anal., vol. 77, p. 102361, 2022

    work page 2022

  2. [2]

    Augmented reality navigation for liver resection with a stereoscopic laparoscope,

    H. Luoet al., “Augmented reality navigation for liver resection with a stereoscopic laparoscope,”Comput. Methods Programs Biomed., vol. 187, p. 105099, 2020

    work page 2020

  3. [3]

    Patient-specific biomechanical modeling for guidance during minimally-invasive hepatic surgery,

    R. Plantefeve, I. Peterlik, N. Haouchine, and S. Cotin, “Patient-specific biomechanical modeling for guidance during minimally-invasive hepatic surgery,”Ann Biomed Eng., vol. 44, pp. 139–153, 2016

    work page 2016

  4. [4]

    Febio: Finite elements for biomechanics,

    S. A. Maas, B. J. Ellis, G. A. Ateshian, and J. A. Weiss, “Febio: Finite elements for biomechanics,”J Biomech Eng., vol. 134, p. 011005, 2012

    work page 2012

  5. [5]

    Non-rigid volume to surface registration using a data- driven biomechanical model,

    M. Pfeifferet al., “Non-rigid volume to surface registration using a data- driven biomechanical model,” inMICCAI 2020, vol. 12264. Springer, 2020, pp. 724–734

    work page 2020

  6. [6]

    Compensating for intraoperative soft-tissue deformations using incomplete surface data and finite elements,

    D. M. Cash, M. I. Miga, T. K. Sinha, R. L. Galloway, and W. C. Chapman, “Compensating for intraoperative soft-tissue deformations using incomplete surface data and finite elements,”IEEE Trans. Medical Imaging, vol. 24, no. 11, pp. 1479–1491, 2005

    work page 2005

  7. [7]

    Data-driven intra-operative estimation of anatomical attachments for autonomous tissue dissection,

    E. Tagliabueet al., “Data-driven intra-operative estimation of anatomical attachments for autonomous tissue dissection,”IEEE Robotics Autom. Lett., vol. 6, no. 2, pp. 1856–1863, 2021

    work page 2021

  8. [8]

    Non-rigid medical image registration using physics- informed neural networks,

    Z. Minet al., “Non-rigid medical image registration using physics- informed neural networks,” inIPMI 2023, vol. 13939. Springer, 2023, pp. 601–613

    work page 2023

Show all 35 references
  1. [9]

    Biomechanically constrained surface registration: Application to mr-trus fusion for prostate interventions,

    S. Khallaghiet al., “Biomechanically constrained surface registration: Application to mr-trus fusion for prostate interventions,”IEEE Trans. Medical Imaging, vol. 34, no. 11, pp. 2404–2414, 2015

    2015

  2. [10]

    An optimal control problem for elastic registration and force estimation in augmented surgery,

    G. Mestdagh and S. Cotin, “An optimal control problem for elastic registration and force estimation in augmented surgery,” inMICCAI 2022, vol. 13437. Springer, 2022, pp. 74–83

    2022

  3. [11]

    Intraoperative estima- tion of liver boundary conditions from multiple partial surfaces,

    A. Mendizabal, E. Tagliabue, and D. Dall’Alba, “Intraoperative estima- tion of liver boundary conditions from multiple partial surfaces,”Int. J. Comput. Assist. Radiol. Surg., vol. 18, no. 7, pp. 1295–1302, 2023

    2023

  4. [12]

    Parallel incomplete-lu and cholesky factorization in the preconditioned iterative methods on the gpu,

    M. Naumov, “Parallel incomplete-lu and cholesky factorization in the preconditioned iterative methods on the gpu,”Nvidia Technical Report NVR-2012-003, 2012

    2012

  5. [13]

    Real-time biomechanical modelling of the liver using lightgbm model,

    J. Zhuet al., “Real-time biomechanical modelling of the liver using lightgbm model,”Int J Med Robot., vol. 18, no. 6, p. e2433, 2022

    2022

  6. [14]

    Learning soft tissue behavior of organs for surgical navigation with convolutional neural networks,

    M. Pfeiffer, C. Riediger, J. Weitz, and S. Speidel, “Learning soft tissue behavior of organs for surgical navigation with convolutional neural networks,”Int. J. Comput. Assist. Radiol. Surg., vol. 14, pp. 1147–1155, 2019

    2019

  7. [15]

    Anatomy-aware computed tomography-to- ultrasound spine registration,

    M. F. Azampouret al., “Anatomy-aware computed tomography-to- ultrasound spine registration,”Med Phys., vol. 51, no. 3, pp. 2044–2056, 2024

    2024

  8. [16]

    Improving intraoperative liver registration in image-guided surgery with learning-based reconstruction,

    M. Jia and M. Kyan, “Improving intraoperative liver registration in image-guided surgery with learning-based reconstruction,” inICASSP

  9. [17]

    Tracking tumors under deformation from partial point clouds using occupancy networks,

    P. Henrichet al., “Tracking tumors under deformation from partial point clouds using occupancy networks,” inIROS 2024. IEEE, 2024, pp. 7227–7234

    2024

  10. [18]

    The image-to-physical liver registration sparse data challenge: Comparison of state-of-the-art using a common dataset,

    J. S. Heiselman, J. A. Collins, M. J. Ringel, T. Peter Kingham, W. R. Jarnagin, and M. I. Miga, “The image-to-physical liver registration sparse data challenge: Comparison of state-of-the-art using a common dataset,”J Med Imaging, vol. 11, no. 1, pp. 015 001–015 001, 2024

    2024

  11. [19]

    A review on organ deformation modeling approaches for reliable surgical navigation using augmented reality,

    Z. Han and Q. Dou, “A review on organ deformation modeling approaches for reliable surgical navigation using augmented reality,” Comput Assist Surg, vol. 29, no. 1, p. 2357164, 2024

    2024

  12. [20]

    Bonet and R

    J. Bonet and R. D. Wood,Nonlinear continuum mechanics for finite element analysis. Cambridge University Press, 1997

    1997

  13. [21]

    Utopic: Uncertainty-aware overlap prediction network for partial point cloud registration,

    Z. Chenet al., “Utopic: Uncertainty-aware overlap prediction network for partial point cloud registration,” inComput. Graph. Forum, vol. 41, no. 7. Wiley Online Library, 2022, pp. 87–98

    2022

  14. [22]

    Deep graph-based spatial consistency for robust non-rigid point cloud registration,

    Z. Qin, H. Yu, C. Wang, Y . Peng, and K. Xu, “Deep graph-based spatial consistency for robust non-rigid point cloud registration,” inCVPR 2023. IEEE, 2023, pp. 5394–5403

    2023

  15. [23]

    Non-rigid point cloud registration with neural deformation pyramid,

    Y . Li and T. Harada, “Non-rigid point cloud registration with neural deformation pyramid,” inNeurIPS 2022, vol. 35. Curran Associates Inc., 2024, pp. 27 757–27 768

    2022

  16. [24]

    Elmer finite element solver for multiphysics and multiscale problems,

    M. Malinen and P. R ˚aback, “Elmer finite element solver for multiphysics and multiscale problems,”Multiscale Model. Methods Appl. Mater. Sci., vol. 19, pp. 101–113, 2013

    2013

  17. [25]

    The image-to- physical liver registration sparse data challenge,

    E. L. Brewer, L. W. Clements, J. A. Collins, D. J. Doss, J. S. Heiselman, M. I. Miga, C. D. Pavas, and E. H. Wisdom III, “The image-to- physical liver registration sparse data challenge,” inMedical imaging 2019: image-guided procedures, robotic interventions, and modeling, vol...

    2019

  18. [26]

    Depth anything: Unleashing the power of large-scale unlabeled data,

    L. Yang, B. Kang, Z. Huang, X. Xu, J. Feng, and H. Zhao, “Depth anything: Unleashing the power of large-scale unlabeled data,” inCVPR

  19. [27]

    Landmark- free preoperative-to-intraoperative registration in laparoscopic liver re- section,

    J. Zhou, B. Gao, K. Wang, J. Pei, P.-A. Heng, and J. Qin, “Landmark- free preoperative-to-intraoperative registration in laparoscopic liver re- section,”IEEE Trans. Medical Imaging, 2025

    2025

  20. [28]

    Tetgen, a delaunay-based quality tetrahedral mesh generator,

    S. Hang, “Tetgen, a delaunay-based quality tetrahedral mesh generator,” ACM Trans. Math. Softw, vol. 41, no. 2, p. 11, 2015

    2015

  21. [29]

    Intraoperative correction of liver deformation using sparse surface and vascular features via linearized iterative boundary reconstruction,

    J. S. Heiselman, W. R. Jarnagin, and M. I. Miga, “Intraoperative correction of liver deformation using sparse surface and vascular features via linearized iterative boundary reconstruction,”IEEE Trans. Medical Imaging, vol. 39, no. 6, pp. 2223–2234, 2020

    2020

  22. [30]

    Comparing regularized kelvinlet functions and the finite element method for registration of medical images to sparse organ data,

    M. J. Ringel, J. S. Heiselman, W. L. Richey, I. M. Meszoely, W. R. Jarnagin, and M. I. Miga, “Comparing regularized kelvinlet functions and the finite element method for registration of medical images to sparse organ data,”Medical Image Anal., vol. 96, p. 103221, 2024

    2024

  23. [31]

    Boundary constraint- free biomechanical model-based surface matching for intraoperative liver deformation correction,

    Z. Yang, R. Simon, K. Merrell, and C. A. Linte, “Boundary constraint- free biomechanical model-based surface matching for intraoperative liver deformation correction,”IEEE Trans. Medical Imaging, 2024

    2024

  24. [32]

    Optimal hyperparameter selection in deformable image registration using information criterion and band-limited modal recon- struction,

    J. S. Heiselman, M. J. Ringel, J. Chakraborty, W. R. Jarnagin, and M. I. Miga, “Optimal hyperparameter selection in deformable image registration using information criterion and band-limited modal recon- struction,” inMedical Imaging 2024: Image-Guided Procedures, Robotic Inte...

    2024

  25. [33]

    Learning preconditioners for conjugate gradient pde solvers,

    Y . Li, P. Y . Chen, T. Du, and W. Matusik, “Learning preconditioners for conjugate gradient pde solvers,” inICML 2023, vol. 202. PMLR, 2023, pp. 19 425–19 439

    2023

  26. [2021]

    1230–1234

    IEEE, 2021, pp. 1230–1234

    2021

  27. [2024]

    10 371–10 381

    IEEE, 2024, pp. 10 371–10 381

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.