Pith. sign in

REVIEW 3 major objections 5 minor 47 references

Deep operator network models for predicting post-burn contraction

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A neural operator predicts post-burn contraction on unseen wound shapes with R^2 = 0.99 and 128–235x speedup.

desk verdict A solid, honest surrogate-model paper with reproducible code, but the generalization claim rests on a shape encoding the authors admit is not injective; worth peer review with a request to fix the evidence. read the letter →

arxiv 2411.14555 v1 pith:BNU7QMJ6 submitted 2024-11-21 math.NA cs.LGcs.NAphysics.bio-phq-bio.TO

Selma Husanovic , Ginger Egberts , Alexander Heinlein , Fred Vermolen This is my paper · ORCID
classification math.NAcs.LGcs.NAphysics.bio-phq-bio.TO MSC 65M6068T0792C50
keywords operatorlearningDeepONetpost-burncontractionsurrogatemodelfiniteelementmethodmorphoelasticitywoundhealingneuralnetwork
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that a deep operator network (DeepONet) — a neural network that learns maps between function spaces — can replace slow finite-element simulations of the two-dimensional morphoelastic model that governs post-burn skin contraction. The authors train on three basic wound shapes (rectangle, rhombus, ellipse) and test on convex combinations of those shapes that were never seen during training, reporting $R^2 = 0.99$, an average relative root mean squared error of $0.075$, and speedups of about 128-fold on CPU and 235-fold on GPU. If the claim holds, clinicians could obtain patient-specific predictions of dermal displacement and wound-relative surface area in seconds rather than minutes, which is what interactive treatment planning would need.

What carries the argument

The load-bearing object is the DeepONet architecture with its branch/trunk split. The branch is a multi-layer perceptron that maps five scalar parameters — cell diffusion constant, chemotactic constant, signaling diffusion constant, cell differentiation rate, and half-maximum cell division enhancement rate — to coefficient vectors $b_1,\dots,b_{2p}$. The trunk is another multi-layer perceptron that maps $(t,x,y)$ together with the wound-shape quadruple $(y_{\mathrm{cut}}, x_m, y_m, x_{\mathrm{cut}})$ to basis values $c_1,\dots,c_p$; the outputs are $\hat{u}_1 = \sum_i b_i c_i$ and $\hat{u}_2 = \sum_i b_{i+p} c_i$. Sine augmentation then enforces the boundary conditions exactly by multiplying $\hat{u}_1$ by $\sin(\pi x/x_l)\cos(\pi y/(2y_l))$ and $\hat{u}_2$ by $\sin(\pi y/y_l)\cos(\pi x/(2x_l))$. The quadruple is the part that lets the trunk adapt its learned basis to a new wound geometry, and it is what carries the claimed generalization.

What would settle it

Take two convex combinations with the same quadruple $(y_{\mathrm{cut}}, x_m, y_m, x_{\mathrm{cut}})$ but different weights, run the finite-element solver on both, and compare the two displacement fields; if the DeepONet returns nearly identical predictions while the solver's outputs differ by more than the reported error band, the shape encoding is not carrying the deformation-relevant geometry and the generalization claim would be falsified.

Watch

Extended reading notes

Core claim

The central claim is that the solution operator of the post-burn contraction PDE system can be learned by a DeepONet whose branch network takes five patient-specific biophysical parameters and whose trunk network takes the evaluation point $(t,x,y)$ plus a four-number encoding $(y_{\mathrm{cut}}, x_m, y_m, x_{\mathrm{cut}})$ of the initial wound shape; the dot product of the two network outputs gives the displacement field $(u_1,u_2)$. A sine-augmentation layer multiplies the raw outputs by sinusoidal factors, so the homogeneous boundary conditions on the rectangular domain hold exactly rather than approximately. Trained on 750 finite-element simulations run to $t=100$ days, the model reaches $R^2 = 0.99437$, $\mathrm{aRRMSE} = 0.07497$, and $\mathrm{aRelErr} = 0.03429$ on an 18-million-point convex test set. Extending the horizon to one year by adding 50 simulations with $t_{\mathrm{end}}=365$ days and sampling 30 time steps per run brings the predicted relative surface area at $t=365$ to within roughly half a percent of the finite-element target. From the ablation comparisons the paper concludes that initial-shape information must enter through the trunk, and that adding new finite-element runs improves generalization more than adding extra time samples from existing runs.

Load-bearing premise

The load-bearing premise is that the four-number shape encoding $(y_{\mathrm{cut}}, x_m, y_m, x_{\mathrm{cut}})$ carries enough geometric information for the trunk network to distinguish the wound shapes that matter, even though the convex test set contains different shapes that this quadruple does not uniquely determine; if the encoding is insufficient, the high scores may reflect interpolation among the three training shapes rather than true generalization.

Editorial extensions

If this is right

  • Burn-contraction simulation can be reduced to a learned operator: given five patient parameters and a four-number wound-shape descriptor, the full two-dimensional displacement field over 100 days is obtained in seconds rather than minutes.
  • Because the trunk receives the shape encoding, the same trained network applies to any convex combination of the three training shapes without retraining, supporting generalization beyond the exact geometries in the training set.
  • Extending the time horizon to one year requires adding simulations that reach $t=365$; with 50 added runs sampled at 30 time steps, the predicted relative surface area stays within about 0.6 percent of the finite-element reference in the worst case.
  • Initial wound-shape information must be fed to the trunk network: moving it to the branch degrades $R^2$ from 0.99 to 0.90, and omitting sine augmentation along with shape information gives an aRRMSE around 0.32.
  • Adding new finite-element runs to the training set improves one-year predictions more than adding more time samples from existing runs, suggesting data diversity across patients matters more than density per patient.

Reading between the lines

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

  • Editorial inference: the convex test set is a low-dimensional slice of wound geometry, so the strongest reading of the claim is tied to shapes built from the same three prototypes; a genuinely harder test would use asymmetric or non-convex wounds, which the paper itself lists as future work.
  • Editorial inference: because the paper notes that the quadruple no longer uniquely identifies shapes in the convex set, feeding an image of the initial wound (the convolutional route the authors mention) would test whether the network has truly learned geometry or merely a smooth interpolation in the four-parameter encoding.
  • Editorial inference: the comparison of sampling strategies suggests a data-acquisition rule for clinical practice — include more patients with fewer samples per patient — that could be tested by training the same DeepONet on datasets with matched total sample size but different patient counts.
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. The paper trains a DeepONet as a surrogate for a two-dimensional morphoelastic finite-element model of post-burn dermal contraction. The branch network takes five patient-specific biological parameters, while the trunk network takes space-time coordinates together with a four-number wound-shape descriptor (ycut, xm, ym, xcut); sine augmentation is used to enforce homogeneous boundary conditions. The model is trained on 750 FE simulations of rectangle, rhombus, and ellipse wound shapes and tested on 150 FE simulations whose geometries are convex combinations of the three training shapes. On a pooled test set of about 18 million points the authors report R2 = 0.99437, aRRMSE = 0.07497, and aRelErr = 0.03429, with CPU/GPU speedups of about 128x and 235x, respectively. The one-year extension requires retraining on new FE simulations with tend = 365 days; the paper reports this honestly and compares two retraining data-sampling strategies.

Significance. If the generalization claim holds, this is a useful step toward geometry-flexible neural surrogates for wound-healing simulations, with potential clinical planning applications. The paper has clear strengths: the code and data are publicly available, the architecture ablation in Table 2 is informative, the one-year extrapolation failure is reported rather than hidden, and the speedup measurements are concrete. However, the main claim of generalization to convex combinations rests on a shape encoding that the paper itself admits is not injective for the test set, so the headline metrics do not yet establish the stated generalization. The work is therefore promising but requires additional evidence or an architectural change.

major comments (3)
  1. [Section 3.3 and Section 2.2] The trunk input quadruple is claimed in Section 2.2 to uniquely determine the initial wound shape, but Section 3.3 explicitly states that for the convex test set the quadruple "no longer uniquely define[s] the initial wound shapes." Since the trunk is the only channel carrying shape information, two different convex combinations with the same quadruple are mapped to the same input and thus to the same predicted displacement field. The paper does not describe how the quadruple is computed for a convex combination, and it provides no evidence that the unrepresented part of the shape has negligible effect on the FE displacement field. Consequently, the pooled R2 = 0.994 in Table 2 does not by itself establish generalization to unseen geometries; it is also compatible with interpolation within the training-shape family or with dominance of small-displacement points. Please quantify the severity of the non-injectivity: identify distinct convex combinations sharing a quadruple, compare their FE displacement fields, and report per-simulation or per-collision-group metrics. Alternatively, replace the four-number descriptor with an injective shape representation, such as values of a level-set or signed-distance function at fixed sensor points.
  2. [Section 4.1 and Eqs. (10)-(12)] The reported R2, aRRMSE, and aRelErr are computed by pooling about 18 million points from only 150 FE simulations, in which the points are highly correlated within each simulation across space and time. No uncertainty intervals, per-simulation score distributions, or effective sample sizes are given. A small number of shape collisions, or collisions whose differences occur mainly in low-displacement interior regions, would be masked by this aggregation. Please report the distribution of scores over the 150 simulations (e.g., per-run R2), and ideally identify how many distinct quadruples occur and how many collisions exist.
  3. [Abstract and Section 4.3] The abstract states that the model "provided reliable predictions over an extended period of up to one year," but Section 4.3 shows that the model trained on t in [0,100] fails after day 100 unless it is retrained on new FE simulations with tend = 365. The abstract and conclusions should state clearly that the one-year predictions require additional FE data and fine-tuning, and that the original model does not extrapolate temporally. This is an honest limitation already present in the body, but the abstract currently overstates the temporal scope.
minor comments (5)
  1. [Section 2.2] The sentence "The latter uniquely determines the initial wound shape at t = 0" should be restricted to the three training shapes, since Section 3.3 later revokes this property for the convex test set.
  2. [Section 3.3] The convex combination in Eq. (9) requires the parametrizations r_p(s) to share a common parameter interval and orientation, but the paper does not specify these parametrizations. Please add the explicit form of r_p(s) for the rectangle, rhombus, and ellipse used in the experiments.
  3. [Section 4.1, Eq. (12)] The relative error metric uses displacements rounded to one decimal place; please clarify the rationale and state whether the reported values are sensitive to this rounding.
  4. [Section 4.3, Fig. 10] The text labels the retraining strategies as "scenario 1" and "scenario 2," but the figure legend describes them by the number of FE simulations and time steps; please align the terminology so the reader can map the text to the figure.
  5. [General] There is a minor typo in Section 3.3: "The convex test set it created" should read "The convex test set is created."

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the DeepONet is trained on independently generated FE outputs and evaluated on separately generated FE outputs; the convex-test-shape descriptor limitation is a generalization concern, not a derivation that reduces to its inputs.

full rationale

The paper's derivation chain is a standard supervised surrogate construction. The labeled displacement targets u1 and u2 come from finite element simulations of the Koppenol morphoelastic model (Eqs. (1)-(3) and the Appendix parameter table), not from the DeepONet's own outputs. The branch inputs (DF, chiF, Dc, kF, aIc) and trunk inputs (t, x, y, ycut, xm, ym, xcut) are independent of the FE targets, and the sine augmentation is a fixed boundary-enforcement transform (Eq. (7)), not a fitted quantity. The convex test set is generated by 150 additional FE simulations with shapes formed by Eq. (9), and the reported R2 = 0.99437 and aRRMSE = 0.07497 are computed against those separately simulated targets (Tab. 2, Eqs. (10)-(12)). No step asserts that a fitted parameter is a prediction, and no uniqueness theorem or ansatz is imported from the authors' prior work; citations to [6], [9], [8], and [23] supply model context and parameter values, not the surrogate's predictive content. The paper itself flags in Section 3.3 that the quadruple (ycut, xm, ym, xcut) 'no longer uniquely define[s]' the convex test shapes, which is an internal limitation on the generalization claim: if two distinct test shapes share a descriptor, the trunk cannot distinguish them. However, this is a data-representation validity concern, not circularity, because the test targets are still externally generated and never feed back into the training loss or architecture choice. The one-year extension is retrained on new tend = 365 FE data (Section 4.3), not tuned to the test metric. Therefore the central claim, while subject to a legitimate generalization caveat, does not reduce to its inputs by construction.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the morphoelastic FE model as ground truth, on the low-dimensional shape encoding, and on hand-chosen parameter ranges and hyperparameters. No new physical entities are introduced.

free parameters (6)
  • q (logistic growth exponent) = -0.4151
    Listed as NC in Tab. 3; not independently measured, yet it changes cell kinetics and therefore the FE training data.
  • k_rho (collagen secretion rate) = 7.6e-8 g/(cells day)
    Marked NC in Tab. 3; set to be consistent with other parameters rather than from a direct measurement.
  • e_N (initial fibroblast density in wound) = 2e3 cells/cm3
    Marked NC; chosen for consistency with equilibrium values, and it sets the initial wound state used in all simulations.
  • e_rho (initial collagen density in wound) = 1.13e-2 g/cm3
    Marked NC; chosen to satisfy equilibrium collagen concentration.
  • Neural network hyperparameters (depth, width, p, learning rate, batch size, epochs) = 3 hidden layers, 50 neurons, p=50, lr=0.001, batch 100, 100 epochs
    Chosen by hand or from a master's thesis using a simplified one-shape model; these affect the reported accuracy.
  • Branch input parameter ranges (DF, chi_F, Dc, kF, aI_c) = Ranges in Tab. 1
    Uniform sampling ranges are chosen and are not derived from a patient cohort; the surrogate's generalization is conditional on them.
assumptions (6)
  • domain assumption The Koppenol morphoelastic PDE model (Eqs. 1-3 and constitutive laws) is an accurate description of post-burn contraction.
    Invoked throughout Section 2.1 as the ground truth that the DeepONet learns to reproduce.
  • domain assumption The quarter-domain symmetry reduction and the stated boundary conditions match the full problem.
    Used in Section 2.1 to define the computational domain and the sine augmentation targets.
  • domain assumption Finite element solutions with mass lumping, FCT limiting, and remeshing are sufficiently accurate reference data.
    The training and test labels come entirely from these FE simulations, described in Section 2.1.
  • domain assumption Convex combinations of the three training shape parametrizations produce valid, physically meaningful wound shapes.
    This is the basis for the convex test set in Section 3.3.
  • domain assumption The sine augmentation in Eq. (7) exactly reproduces the FE boundary conditions on all test shapes.
    The final outputs are multiplied by these sine and cosine factors, so the predicted solution is forced to satisfy the encoded boundary conditions.
  • standard math The operator universal approximation theorem guarantees existence of a DeepONet with sufficient capacity.
    Theorem 1 in Section 2.2 is used to justify the DeepONet framework, though it gives no convergence rate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Deep operator network models for predicting post-burn contraction." pith.science (2026). https://pith.science/paper/BNU7QMJ6

@misc{pith2026241114555,
  author       = {Pith},
  title        = {Pith review of: Deep operator network models for predicting post-burn contraction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BNU7QMJ6}},
  note         = {Machine review of arXiv:2411.14555}
}
abstract

Burn injuries present a significant global health challenge. Among the most severe long-term consequences are contractures, which can lead to functional impairments and disfigurement. Understanding and predicting the evolution of post-burn wounds is essential for developing effective treatment strategies. Traditional mathematical models, while accurate, are often computationally expensive and time-consuming, limiting their practical application. Recent advancements in machine learning, particularly in deep learning, offer promising alternatives for accelerating these predictions. This study explores the use of a deep operator network (DeepONet), a type of neural operator, as a surrogate model for finite element simulations, aimed at predicting post-burn contraction across multiple wound shapes. A DeepONet was trained on three distinct initial wound shapes, with enhancement made to the architecture by incorporating initial wound shape information and applying sine augmentation to enforce boundary conditions. The performance of the trained DeepONet was evaluated on a test set including finite element simulations based on convex combinations of the three basic wound shapes. The model achieved an $R^2$ score of $0.99$, indicating strong predictive accuracy and generalization. Moreover, the model provided reliable predictions over an extended period of up to one year, with speedups of up to 128-fold on CPU and 235-fold on GPU, compared to the numerical model. These findings suggest that DeepONets can effectively serve as a surrogate for traditional finite element methods in simulating post-burn wound evolution, with potential applications in medical treatment planning.

Figures

Figures reproduced from arXiv: 2411.14555 by the authors.

Figure 1
Figure 1. Nonlinear activation functions (a) and a feedforward neural network with two hidden layers (b) [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. (Unstacked) DeepONet architecture; based on [29] [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. The DeepONet architecture used for predicting dermal displacement. The legend at the top [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: The three initial wound shapes used for training: rectangle (a), rhombus (b), and ellipse (c). [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: Enforcement of boundary conditions as a result of sine augmentation (a) and a convex combination [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 4
Figure 4. Figure 4: We run 750 finite element simulations, each run uniformly selecting one of the three initial wound shapes. For the wound size, we uniformly sample xcut, ycut ∈ (0, 5) centimeters. The size of the complete domain is then determined as xl = 2.5 · xcut, yl = 2.5 · ycut, (…
Figure 6
Figure 6. Figure 6: Training and validation losses of the proposed DeepONet. Dataset is based on 750 finite element [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: Prediction vs. target for three samples from the convex test set, at their respective time of [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: True vs. predicted displacement in x- (a) and y-direction (b) on the convex test set 4.1 Performance on convex test set Our DeepONet predicts the x and y displacements in any given spatial coordinate in the domain, at any given time. We evaluate the trained model on ea…
Figure 9
Figure 9. Figure 9: Best (a) and worst (b) prediction on the convex test set in terms of the RSAW. Mean and [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: Best (a) and worst (b) prediction in terms of the RSAW on a small convex test set, comparing [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

47 extracted references · 41 canonical work pages

  1. [1]

    Alberts, D

    B. Alberts, D. Bray, J. Lewis, M. Raff, K. Roberts, and J. Watson. The molecular biology of the cell. Garland Publishing, second edition, 1989. 18

    work page 1989

  2. [2]

    Chen and H

    T. Chen and H. Chen. Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems. IEEE Transactions on Neural Networks, 6:911–917, 1995

    work page 1995

  3. [3]

    DelftBlue Supercomputer (Phase 1)

    Delft High Performance Computing Centre (DHPC). DelftBlue Supercomputer (Phase 1). https: //www.tudelft.nl/dhpc/ark:/44463/DelftBluePhase1, 2024

    work page 2024

  4. [4]

    Evaluation of empirical models for predicting monthly mean horizontal diffuse solar radiation

    Milan Despotovic, Vladimir Nedic, Danijela Despotovic, and Slobodan Cvetanovic. Evaluation of empirical models for predicting monthly mean horizontal diffuse solar radiation. Renewable and Sustainable Energy Reviews, 56:246–260, 2016

    work page 2016

  5. [5]

    U-DeepONet: U-Net Enhanced Deep Operator Network for Geologic Carbon Sequestration

    Waleed Diab and Mohammed Al-Kobaisi. U-DeepONet: U-Net Enhanced Deep Operator Network for Geologic Carbon Sequestration, December 2023. arXiv:2311.15288 [physics]

    work page Pith review arXiv 2023

  6. [6]

    A Bayesian finite- element trained machine learning approach for predicting post-burn contraction

    Ginger Egberts, Marianne Schaaphok, Fred Vermolen, and Paul van Zuijlen. A Bayesian finite- element trained machine learning approach for predicting post-burn contraction. Neural Computing and Applications, 34(11):8635–8642, 2022

    work page 2022

  7. [7]

    Some mathematical properties of morphoelasticity

    Ginger Egberts, Daan Smits, Fred Vermolen, and Paul van Zuijlen. Some mathematical properties of morphoelasticity. In Lecture Notes in Computational Science and Engineering, pages 1119–1127. Springer International Publishing, 2020

    work page 2020

  8. [8]

    Sensitivity and feasibility of a one-dimen- sional morphoelastic model for post-burn contraction

    Ginger Egberts, Fred Vermolen, and Paul van Zuijlen. Sensitivity and feasibility of a one-dimen- sional morphoelastic model for post-burn contraction. Biomechanics and Modeling in Mechanobiol- ogy, 20(6):2147–2167, 2021

    work page 2021

Show all 47 references
  1. [9]

    High-speed predictions of post-burn con- traction using a neural network trained on 2d-finite element simulations

    Ginger Egberts, Fred Vermolen, and Paul van Zuijlen. High-speed predictions of post-burn con- traction using a neural network trained on 2d-finite element simulations. Frontiers in Applied Mathematics and Statistics, 9, 2023

    2023

  2. [10]

    Deep learning

    Ian Goodfellow, Yoshua Bengio, and Aaron Courville. Deep learning. Adaptive Computation and Machine Learning. MIT Press, Cambridge, MA

  3. [11]

    K. Gosh, Z. Pan, E. Guan, S. Ge, T. Lio, Y. Nakamura, Z. Ren, M. Rafailovich, and R. Clark. Cell adaptation to a physiologically relevant ECM mimic with different viscoelastic properties. Biomaterials, 28(4):671–679, 2007

    2007

  4. [12]

    FEM-based Neural Networks for Solving Incompressible Fluid Flows and Related Inverse Problems, September 2024

    Franziska Griese, Fabian Hoppe, Alexander R¨ uttgers, and Philipp Knechtges. FEM-based Neural Networks for Solving Incompressible Fluid Flows and Related Inverse Problems, September 2024. arXiv:2409.04067 [physics]

    2024 arXiv

  5. [13]

    Modelling of some biological materials using continuum mechanics

    Cameron Luke Hall. Modelling of some biological materials using continuum mechanics. PhD thesis, Queensland University of Technology, 2008

    2008

  6. [14]

    Howard, Damien Beecroft, and Panos Stinis

    Alexander Heinlein, Amanda A. Howard, Damien Beecroft, and Panos Stinis. Multifidelity domain decomposition-based physics-informed neural networks for time-dependent problems, January 2024. arXiv:2401.07888 [cs, math]

    2024 arXiv

  7. [15]

    A user’s guide to pde models for chemotaxis

    Thomas Hillen and Kevin J Painter. A user’s guide to pde models for chemotaxis. Journal of mathematical biology, 58(1-2):183–217, 2009

    2009

  8. [16]

    Howard, Sarah H

    Amanda A. Howard, Sarah H. Murphy, Shady E. Ahmed, and Panos Stinis. Stacked networks improve physics-informed training: applications to neural networks and deep operator networks, November 2023. arXiv:2311.06483 [cs, math]

    2023 arXiv

  9. [17]

    The application of neural operators to predict skin evolution after burn trauma

    Selma Husanovic. The application of neural operators to predict skin evolution after burn trauma. Unpublished thesis, 2024

    2024

  10. [18]

    ICRP Publication 110

    Icrp. ICRP Publication 110. SAGE Publications Ltd, 2010. Table A.1 p. 51

    2010

  11. [19]

    Henry J. Kelley. Gradient theory of optimal flight paths. 30(10):947–954. Publisher: American Institute of Aeronautics and Astronautics eprint: https://doi.org/10.2514/8.5282

    doi

  12. [20]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. 19

  13. [21]

    DeepOnet Based Preconditioning Strategies For Solving Parametric Linear Systems of Equations, January 2024

    Alena Kopaniˇ c´ akov´ a and George Em Karniadakis. DeepOnet Based Preconditioning Strategies For Solving Parametric Linear Systems of Equations, January 2024. arXiv:2401.02016 [cs, math]

    2024 arXiv

  14. [22]

    Koppenol

    D. Koppenol. Biomedical implications from mathematical models for the simulation of dermal wound healing. PhD thesis, Delft University of Technology, 2017

    2017

  15. [23]

    Koppenol and F

    D. Koppenol and F. Vermolen. Biomedical implications from a morphoelastic continuum model for the simulation of contracture formation in skin grafts that cover excised burns. Biomechanics and Modeling in Mechanobiology, 16(4):1187–1206, 2017

    2017

  16. [24]

    A mathematical model for the simulation of the formation and the subsequent regression of hyper- trophic scar tissue after dermal wounding

    Dani¨ el C Koppenol, Fred J Vermolen, Frank B Niessen, Paul PM van Zuijlen, and Kees Vuik. A mathematical model for the simulation of the formation and the subsequent regression of hyper- trophic scar tissue after dermal wounding. Biomechanics and modeling in mechanobiology, 1...

    2017

  17. [25]

    Neural operator: Learning maps between function spaces

    Nikola Kovachki, Zongyi Li, Burigede Liu, Kamyar Azizzadenesheli, Kaushik Bhattacharya, Andrew Stuart, and Anima Anandkumar. Neural operator: Learning maps between function spaces

  18. [26]

    Lagaris, A

    I.E. Lagaris, A. Likas, and D.I. Fotiadis. Artificial neural networks for solving ordinary and partial differential equations. IEEE Transactions on Neural Networks, 9(5):987–1000, September 1998. Conference Name: IEEE Transactions on Neural Networks

    1998

  19. [27]

    Fourier neural operator for parametric partial differential equations

    Zongyi Li, Nikola Kovachki, Kamyar Azizzadenesheli, Burigede Liu, Kaushik Bhattacharya, An- drew Stuart, and Anima Anandkumar. Fourier neural operator for parametric partial differential equations

  20. [28]

    Liu and Jorge Nocedal

    Dong C. Liu and Jorge Nocedal. On the limited memory BFGS method for large scale optimization. 45(1):503–528

  21. [29]

    DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators

    Lu Lu, Pengzhan Jin, and George Em Karniadakis. DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators. 3(3):218–229

  22. [30]

    Maskarinec, C

    S. Maskarinec, C. Franck, D. Tirell, and G. Ravichandran. Quantifying cellular traction forces in three dimensions. Proceedings of the National Academy of Sciences, 106(52):22108–22113, 2009

    2009

  23. [31]

    Meethal, Anoop Kodakkal, Mohamed Khalil, Aditya Ghantasala, Birgit Obst, Kai-Uwe Bletzinger, and Roland W¨ uchner

    Rishith E. Meethal, Anoop Kodakkal, Mohamed Khalil, Aditya Ghantasala, Birgit Obst, Kai-Uwe Bletzinger, and Roland W¨ uchner. Finite element method-enhanced neural network for forward and inverse problems. Advanced Modeling and Simulation in Engineering Sciences, 10(1):6, May 2023

    2023

  24. [32]

    Physics-informed machine learning embedded into isogeometric analysis

    Matthias M¨ oller, Deepesh Toshniwal, and Frank van Ruiten. Physics-informed machine learning embedded into isogeometric analysis. Platform Wiskunde, Amsterdam, pages 57–59, 2021

    2021

  25. [33]

    M¨ oller, D

    M. M¨ oller, D. Kuzmin, and D. Kourounis. Implicit FEM-FCT algorithms and discrete Newton methods for transient convection problems. International Journal for Numerical Methods in Fluids, 57(6):761–792, 2008

    2008

  26. [34]

    Olsen, J

    L. Olsen, J. Sherratt, and P. Maini. A mechanochemical model for adult dermal wound contraction and the permanence of the contracted tissue displacement profile. Journal of Theoretical Biology, 177(2):113–128, 1995

    1995

  27. [35]

    Overall, J

    C. Overall, J. Wrana, and J. Sodek. Transcriptional and post-transcriptional regulation of 72-kda gelatinase/ type IV collagenase by transforming growth factor-beta in human fibroblasts. Journal of Biological Chemistry, 266(21):14061–14071, 1991

    1991

  28. [36]

    Raissi, P

    M. Raissi, P. Perdikaris, and G. E. Karniadakis. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 378:686–707, 2019

    2019

  29. [37]

    Convolutional neural operators for robust and accurate learning of PDEs

    Bogdan Raoni´ c, Roberto Molinaro, Tim De Ryck, Tobias Rohner, Francesca Bartolucci, Rima Alaifari, Siddhartha Mishra, and Emmanuel de B´ ezenac. Convolutional neural operators for robust and accurate learning of PDEs. 20

  30. [38]

    Roberts, M

    A. Roberts, M. Sporn, R. Assoian, J. Smith, N. Roche, L. Wakefield, U. Heine, L. Liotta, V. Falanga, and J. Kehrl. Transforming growth factor type beta: rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proceedings of the Nati...

    1986

  31. [39]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Nassir Navab, Joachim Hornegger, William M. Wells, and Alejandro F. Frangi, editors, Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015...

    2015

  32. [40]

    Rudolph and J

    R. Rudolph and J. Vande Berg. The myofibroblast in Dupuytren’s contracture. Journal of Hand Clinics, 7(4):683–692, 1991. Discussion 693-4

    1991

  33. [41]

    Strutz, M

    F. Strutz, M. Zeisberg, A. Renziehausen, B. Raschke, V. Becker, C. Van Kooten, and G. M¨ uller. TGF-β1 induces proliferation in human renal fibroblasts via induction of basic fibroblast growth factor (FGF-2). Kidney International, 59(2):579–592, 2001

    2001

  34. [42]

    DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-shot Transfer the Dynamic Response of Networked Systems, September

    Yixuan Sun, Christian Moya, Guang Lin, and Meng Yue. DeepGraphONet: A Deep Graph Operator Network to Learn and Zero-shot Transfer the Dynamic Response of Networked Systems, September

  35. [43]

    Vande Berg, R

    J. Vande Berg, R. Rudolph, W. Poolman, and D. Disharoon. Comparative growth dynamics and actin concentration between cultured human myofibroblasts from granulating wounds and dermal fibroblasts from normal skin. Lab Invest, 61(5):532–538, 1989

    1989

  36. [44]

    Learning the solution operator of parametric partial differential equations with physics-informed deeponets

    Sifan Wang, Hanwen Wang, and Paris Perdikaris. Learning the solution operator of parametric partial differential equations with physics-informed deeponets. Science advances, 7(40):eabi8605, 2021

    2021

  37. [45]

    Chung, Yalchin Efendiev, and Min Wang

    Yating Wang, Siu Wun Cheung, Eric T. Chung, Yalchin Efendiev, and Min Wang. Deep multiscale model learning. Journal of Computational Physics, 406:109071, April 2020

    2020

  38. [46]

    Wrobel, T

    L. Wrobel, T. Fray, J. Molloy, J. Adams, M. Armitage, and J. Sparrow. Contractility of single human dermal myofibroblasts and fibroblasts. Cell Motility and the Cytoskeleton, 52(2):82–90, 2002. 21 Appendix Tab. 3 lists the fixed parameter values used for the finite element sim...

    2002

  39. [2022]

    arXiv:2209.10622 [cs]

    arXiv

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.