Pith. sign in

REVIEW 3 major objections 6 minor 67 references

Mechanics and Design of Metastructured Auxetic Patches with Bio-inspired Materials

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read Neural networks trained on only 150 finite-element simulations can design auxetic silk-fibroin patches that match specified Poisson's-ratio and stress curves, and the inverse design network outperforms genetic-algorithm optimization.

desk verdict A solid surrogate-based design pipeline for auxetic silk patches whose headline claim about beating GA rests on circular evaluation through the same surrogate that supervises the design model. read the letter →

arxiv 2501.06233 v1 pith:DOL474TY submitted 2025-01-08 cs.LG cond-mat.mtrl-sci

classification cs.LGcond-mat.mtrl-sci
keywords auxeticpatchesmetastructuresilkfibroinneuralnetworksinversedesigncycle-consistentlossgreedysamplingPoisson'sratio
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 argues that a purely data-driven pipeline can replace the slow trial-and-error loop of designing auxetic patches—flexible sheets that widen sideways when pulled lengthwise, a negative-Poisson's-ratio behavior shared by many human tissues. The authors train two neural networks on just 150 finite-element simulations of a sinusoidal silk-fibroin metastructure, selected by greedy sampling, and both predict Poisson's ratio and stress across strain levels up to 15% with $R^2 > 0.995$. They then train a design network that takes desired mechanical response curves as input and outputs the three geometric parameters—amplitude, wavelength, ligament thickness—of a patch that should reproduce them, using a cycle-consistent loss to close the loop and a scaling penalty to generate diverse alternatives. On held-out tests the design network matches target properties with lower error than genetic-algorithm optimization, and it can even partially mimic the response of a differently shaped metastructure. If this works in practice, it would let medical-device engineers specify the mechanical behavior a patch must have and get a fabricable geometry quickly.

What carries the argument

The central machinery is the sinusoidal metastructure parameterized by three variables—amplitude $A$, wavelength $\lambda$, and thickness $t$—with the unit cell repeated into a $5 \times 5$ patch, plus the paired neural networks: two forward predictors (Poisson's ratio and stress over 30 strain levels) and an inverse design network whose loss is a cycle-consistent term (comparing predicted properties of proposed designs against targets) augmented by $L_{\text{scale}}$, which penalizes design groups that are proportional scalings of one another. Greedy sampling selects the 150 most informative FEM simulations to make training tractable. The forward networks serve as differentiable surrogates for FEM, and the design network exploits the scale invariance of the structure (designs $n\lambda, nt, nA$ have the same mechanical response) to recommend accurate, non-redundant designs.

What would settle it

Take a design recommended by the design network that lies outside the FEM training ranges (for example, $\lambda = 23.41$ mm, $t = 2.93$ mm, $A = 1.33$ mm from Section 4.3), run a fresh finite-element simulation with the same silk-fibroin material model, and compare the predicted Poisson's-ratio and stress curves against the network's outputs; a relative MAE much larger than the reported test-set errors (0.0015 for $\nu$, 0.0827 kPa for $\sigma$) would falsify the claim that the design model achieves those targets.

Watch

Extended reading notes

Core claim

The central claim is that a neural-network-based inverse design model can tailor the three geometric parameters of a re-entrant sinusoidal auxetic patch—amplitude $A$, wavelength $\lambda$, and ligament thickness $t$—to achieve specified Poisson's-ratio and stress curves, and that it does so more efficiently and precisely than genetic-algorithm optimization. The forward models predict 30 strain-level outputs from the three parameters; the design network uses these pre-trained predictors as a differentiable feedback loop, minimizing a loss that combines property-matching errors with a scale-diversity penalty. The authors demonstrate on held-out samples that the design model's mean absolute error for Poisson's ratio reaches as low as 0.0015 and for stress as low as 0.0827 kPa, with GA errors several times larger. They also show the same design network can propose three distinct designs for one target, and that it can approximately replicate the Poisson's-ratio-only or stress-only response of a different (oval voids) metastructure, though not both properties simultaneously.

Load-bearing premise

The predictive neural networks, trained on 150 simulations whose three design parameters lie inside fixed ranges, stay accurate for the new designs the inverse model proposes, even though some proposed parameters fall outside the training ranges and no proposed design is re-checked with a fresh simulation or a fabricated patch.

Editorial extensions

If this is right

  • Given a target mechanical response, the trained design network can output a valid sinusoidal patch geometry that matches Poisson's ratio and stress to within the reported MAE, so patch design no longer requires a fresh FEM solve per candidate.
  • Because the structure is scale-invariant, any recommended design can be rescaled to a convenient size without changing its predicted mechanical response, which the paper exploits to align $\lambda$ with the true design before comparison.
  • With the design network, multiple diverse design alternatives can be generated for the same target by setting $n > 1$ and activating the $L_{\text{scale}}$ penalty, giving engineers options beyond the single optimum.
  • The reliance on only 150 FEM samples suggests the same active-learning-plus-surrogate approach can transfer to other low-dimensional metastructure families with limited simulation budgets.
  • The design network can partially mimic the behavior of a different metastructure type (oval voids) for one property at a time, showing limited generalization beyond its training family, but a sinusoidal geometry with three parameters is insufficient to match both properties simultaneously.

Reading between the lines

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

  • Because the forward networks are differentiable, the same cycle-consistent design loop could be closed with physical experiments instead of FEM: fabricate a proposed patch, measure its response, and fine-tune the network weights, which would test whether the surrogate's accuracy survives real manufacturing defects.
  • The scale-invariance penalty $L_{\text{scale}}$ is a generic way to handle one-to-many inverse problems when the forward map is homogeneous; this loss design could be reused for other metamaterial families where proportional scaling preserves properties.
  • The failure to match both Poisson's ratio and stress of an oval-voids structure at once suggests a concrete diagnostic: when a proposed design's predicted curves are accurate but conflicted, the geometry family lacks enough shape degrees of freedom, so adding more parameters (or multi-material layers) is the principled fix rather than just adding more data.
  • A quick testable extension: train the same pipeline on a two-family dataset (sinusoidal plus oval voids) and check whether the design network learns to interpolate between families, which would indicate whether the limitation is data coverage or geometry expressiveness.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper presents a data-driven framework for predicting and inversely designing sinusoidal auxetic patches made of silk fibroin. After fabricating and testing the material, the authors validate a finite element (FEM) model against experiments up to 15% strain, use greedy sampling (GS) to select 150 FEM simulations, and train two neural network regressors that predict Poisson's ratio and stress curves across 30 strain levels. They then propose a design neural network that takes target mechanical property curves as input and outputs design parameters (λ, t, A), trained with a cycle-consistent-style loss that feeds the predicted designs through the same predictive networks, augmented with a diversity penalty. The design model is compared with a genetic algorithm (GA) on a few test cases, and the paper claims superior accuracy and efficiency. The paper concludes with a limitation analysis showing that the sinusoidal geometry cannot simultaneously match both Poisson's ratio and stress of an oval-voids structure.

Significance. If the inverse-design claims were independently validated, the framework would offer a fast, data-driven route to tailoring auxetic patch geometries for tissue-engineering applications. The paper has several genuine strengths: the FEM model is verified against experimental compression up to 15% strain; GS active learning reduces the number of required simulations to 150; the two predictive models report R² values above 0.995 on an unseen test set; SHAP and sensitivity analyses give interpretable insights into which geometric variables drive each mechanical property; and the authors honestly acknowledge a real limitation—the three-parameter sinusoidal design cannot reproduce both Poisson's ratio and stress of a different metastructure simultaneously. The core weakness is that the central 'superior performance' claim for the inverse design model rests entirely on evaluation through the same surrogate ANNs that supervise its training, with no fresh FEM or experimental confirmation.

major comments (3)
  1. [Section 4.3, Tables 1–2, Eqs. (5)–(7)] The design model's training loss and its evaluation both use the same fitted ANN predictive models. The MAE values in Tables 1 and 2 (e.g., MAE_ν = 0.0015, MAE_σ = 0.0827 kPa) are computed by feeding the proposed designs back into the Poisson's-ratio and stress networks that appear in the loss function, not by running new FEM simulations or experiments. This is a circular evaluation: any systematic bias or interpolation artifact in the surrogate is invisible, so the reported errors measure agreement with the surrogate, not with the physical response. The abstract's claim that the design model 'demonstrated superior performance' relative to GA is therefore not yet established. I request an independent check: run FEM simulations (or experiments) on at least a subset of the generated designs, including the three cases in Table 1 and the multi-design case in Table 2, and report the true Poisson's-ratio and stress errors alongside the surrogate-based MAEs.
  2. [Section 4.3 (text near Table 1) and Section 2.1] The design layer is unconstrained during training, and the paper reports a raw output of λ = 23.41 mm, t = 2.93 mm, A = 1.33 mm for the first test sample—outside the GS pool ranges (λ ∈ [2, 21] mm, t ∈ [0.2, 2.1] mm, A ∈ [0.2, 2.1] mm). The paper does not report the raw outputs for the other test samples, nor does it check whether any raw or rescaled design satisfies the physical validity condition d = λ/2 − 2A − t > 0 (Eq. 2). The rescaling step is also questionable: the text claims properties are 'consistent under proportional scaling,' but the out-of-plane shell thickness remains fixed at 1 mm while λ, t, and A are scaled, so the geometry is not truly self-similar. If the rescaling is necessary to bring designs into the training range, its validity should be demonstrated, and the design model should either be constrained or its raw outputs should be explicitly checked for physical feasibility.
  3. [Tables 1–2 and Figure 13] The comparison with GA is based on only three test samples (Table 1) and one sample (Table 2), with Figure 13 showing a single representative case. No confidence intervals, repeated runs, or significance tests are reported, and the GA is confined to the GS bounds while the design model is unconstrained during training—an asymmetry that biases the comparison in favor of the design model. The claim that the design model outperforms GA needs a systematic evaluation over the entire test set, with multiple independent GA runs and a matched feasible design region (e.g., constraining the design model to the GS pool or allowing GA to search beyond it), before such a conclusion can be drawn.
minor comments (6)
  1. [Section 2.2] The text contains a typo: 'Poission's ratios' should be 'Poisson's ratios'.
  2. [Figure 7] The caption says 'Comparion' instead of 'Comparison'.
  3. [Eq. (4) and Section 2.1] The nominal stress expression σ = F_R / (5λte) uses the symbol e without a prior definition; clarify that e denotes the out-of-plane shell thickness (stated as 1 mm earlier in the section).
  4. [Section 2.3 vs. Section 2.2] The design model input is described as Poisson's ratios and stresses at 30 strain levels 'ranging from 0% to 15% in 0.5% intervals,' while Section 2.2 states the predictive models cover '0.5% to 15%' in 0.5% intervals; reconcile the starting strain value.
  5. [Eq. (8)] The definition of L_scale is hard to read: the denominator '1 P × (N 2)' is ambiguous, and P is not defined before its first use. Please rewrite the equation with a clearly stated summation index and definition of P and N.
  6. [Introduction, first challenge paragraph] The phrase 'sign-point material property prediction less effective' is unclear; consider rephrasing to 'pointwise material property prediction'.

Circularity Check

1 steps flagged · score 6.0 of 10

The inverse-design evaluation is closed-loop: the same fitted ANN predictive models appear in the design-model loss (Eqs. 5-7) and in the reported MAE evaluation (Section 4.3), with no fresh FEM or experimental check on generated designs.

  1. fitted input called prediction [Section 2.3 (Eqs. 5-7) and Section 4.3 (Fig. 13, Tables 1-2)]
    "The model’s performance on the test dataset is evaluated using mean absolute error (MAE) as the metric. Specifically, the MAE for each design scheme, generated by the model’s design layer, is calculated by comparing the desired mechanical behaviors with the predicted values obtained from the predictive models."

    The design model’s training loss L_nu + L_sigma (Eqs. 6-7) is the MSE between target properties and outputs of the Poisson/stress predictive ANNs. Tables 1-2 and Fig. 13 report MAE between the same targets and the same predictive-ANN outputs for the generated designs, after rescaling. The reported “superior performance” is therefore the objective that was optimized, measured through the same fitted surrogates that supervise the design network; the surrogate is both teacher and examiner. No FEM simulation or experiment on any generated design is reported, and the raw output λ=23.41 mm, t=2.93 mm, A=1.33 mm lies outside the training pool (λ in [2,21], t in [0.2,2.1], A in [0.2,2.1] mm) and is rescaled to λ=9 mm before evaluation, bypassing the actual out-of-range output.

full rationale

The paper is not globally circular: the forward ANN predictive models are trained and R2-validated on 150 FEM simulations, and the comparison of the design model with GA is internally fair since both are scored by the same surrogate. However, the central claim of “precise design solutions” that “tailor patch designs to achieve specific mechanical properties” rests entirely on the fitted predictive models: they define the design objective (Eqs. 6-7) and they define the success metric (Section 4.3 MAE). No independent FEM or experimental verification of the optimized designs is provided, and the rescaling of out-of-range outputs before evaluation means even surrogate-level performance of the raw generated design is not shown. This is a partial circularity (score 6): the evaluated quantity reduces to the quantity optimized, although the optimization could still fail and the GA baseline adds some independent content. The GS self-citation [56] is not load-bearing for the central claim, since the method is described in the text and the forward models are validated against FEM test data.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a chain of modeling choices: the 5x5 patch approximation, the single selected material curve, the scale-invariance property, and the extrapolation of surrogate models to unconstrained design outputs. No new physical entities are introduced.

free parameters (5)
  • Loss weights alpha, beta, gamma in design model = alpha=beta=1.0, gamma=0 (single); alpha=beta=1.0, gamma=0.5 (multi); alpha=1,beta=0 or alpha=0,beta=1 (single-property)
    Balance property-matching and diversity terms in Eq (5); chosen by hand, no systematic sweep reported.
  • ANN hyperparameters (learning rate, epochs, batch size, layer sizes) = Layer sizes reported (predictive: 50-100-125-75; design: 90-125-150-100-50); learning rate and epochs not reported
    Tuned on the 9-sample validation set; omitted details prevent exact reproduction.
  • Friction coefficient in FEM validation = 1.0
    Set to prevent sliding between glass slides and patch in Section 3.3; no sensitivity study.
  • Number of FEM training samples selected by greedy sampling = 150
    Chosen to balance labeling cost and accuracy; selection criterion details are referenced to [56], not reproduced.
  • Strain levels and range (0.5% to 15% at 0.5% intervals) = 30 levels
    Defines the output dimension and the design targets; the upper bound is tied to the FEM validation limit.
assumptions (6)
  • domain assumption The central unit cell of a 5x5 patch under fixed-displacement boundary conditions approximates periodic boundary conditions.
    Section 2.1: 'This approach enables us to adopt periodic boundary conditions, ensuring data stability and consistency'; the 5x5 finite patch is not actually periodic.
  • domain assumption The material response of silk fibroin is identical in all simulated designs and matches the single selected experimental stress-strain curve.
    Section 3.2: one central curve selected among three; Sections 2.1 and 3.3 use it as the FEM material input.
  • domain assumption Proportional scaling of lambda, t, A leaves mechanical properties unchanged.
    Used in Section 4.3 to rescale design-model outputs to match the true lambda; no proof or error analysis provided.
  • ad hoc to paper The ANN predictive models generalize to designs outside the training parameter ranges explored during design-model optimization.
    Design layer outputs are unbounded (Section 2.3); Section 4.3 shows out-of-range outputs such as lambda=23.41 mm, which the paper then rescales.
  • ad hoc to paper Greedy sampling (GS) selects a representative and sufficient training set for the design space.
    Method referenced to the authors' prior work [56]; no replication details in this paper.
  • domain assumption FEM with SHELL181 and the chosen material model accurately represents the fabricated silk fibroin patches up to 15% strain.
    Validated against one experimental geometry in Section 3.3; the validation does not cover the full design space.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Mechanics and Design of Metastructured Auxetic Patches with Bio-inspired Materials." pith.science (2026). https://pith.science/paper/DOL474TY

@misc{pith2026250106233,
  author       = {Pith},
  title        = {Pith review of: Mechanics and Design of Metastructured Auxetic Patches with Bio-inspired Materials},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DOL474TY}},
  note         = {Machine review of arXiv:2501.06233}
}
abstract

Metastructured auxetic patches, characterized by negative Poisson's ratios, offer unique mechanical properties that closely resemble the behavior of human tissues and organs. As a result, these patches have gained significant attention for their potential applications in organ repair and tissue regeneration. This study focuses on neural networks-based computational modeling of auxetic patches with a sinusoidal metastructure fabricated from silk fibroin, a bio-inspired material known for its biocompatibility and strength. The primary objective of this research is to introduce a novel, data-driven framework for patch design. To achieve this, we conducted experimental fabrication and mechanical testing to determine material properties and validate the corresponding finite element models. Finite element simulations were then employed to generate the necessary data, while greedy sampling, an active learning technique, was utilized to reduce the computational cost associated with data labeling. Two neural networks were trained to accurately predict Poisson's ratios and stresses for strains up to 15\%, respectively. Both models achieved $R^2$ scores exceeding 0.995, which indicates highly reliable predictions. Building on this, we developed a neural network-based design model capable of tailoring patch designs to achieve specific mechanical properties. This model demonstrated superior performance when compared to traditional optimization methods, such as genetic algorithms, by providing more efficient and precise design solutions. The proposed framework represents a significant advancement in the design of bio-inspired metastructures for medical applications, paving the way for future innovations in tissue engineering and regenerative medicine.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 53 canonical work pages

  1. [1]

    Tomoda, K., Kimura, H., Osaki, S.: Distribution of collagen fiber orientation in the human lung. Anat. Rec. 296(5), 846–850 (2013) https://doi.org/10.1002/ar. 22649

  2. [2]

    Ferraiuoli, P., Fixsen, L.S., Kappler, B., Lopata, R.G., Fenner, J.W., Narra- cott, A.J.: Measurement of in vitro cardiac deformation by means of 3d digital image correlation and ultrasound 2d speckle-tracking echocardiography. Med. Eng. Phys. 74, 146–152 (2019) https://doi.org/10.1016/j.medengphy.2019.09.021

  3. [3]

    Neurogastroenterol

    Banerjee, S., Pal, A., Fox, M.: Volume and position change of the stomach during gastric accommodation and emptying: A detailed three-dimensional mor- phological analysis based on mri. Neurogastroenterol. Motil. 32(8), 13865 (2020) https://doi.org/10.1111/nmo.13865

  4. [4]

    Mir, M., Ali, M.N., Sami, J., Ansari, U.: Review of mechanics and applications of auxetic structures. Adv. Mater. Sci. Eng., 753496 (2014) https://doi.org/10. 1155/2014/753496 23

  5. [5]

    Li, T., Chen, Y., Hu, X., Li, Y., Wang, L.: Exploiting negative poisson’s ratio to design 3d-printed composites with enhanced mechanical properties. Mater. Des. 142, 247–258 (2018) https://doi.org/10.1016/j.matdes.2018.01.034

  6. [6]

    Kapnisi, M., Mansfield, C., Marijon, C., Guex, A.G., Perbellini, F., Bardi, I., Humphrey, E.J., Puetzer, J.L., Mawad, D., Koutsogeorgis, D.C., Stuckey, D.J., Terracciano, C.M., Harding, S.E., Stevens, M.M.: Auxetic cardiac patches with tunable mechanical and conductive properties toward treating myocardial infarc- tion. Adv. Funct. Mater.28(21), 1800618 (...

  7. [7]

    Chansoria, P., Etter, E.L., Nguyen, J.: Regenerating dynamic organs using biomimetic patches. Trends. Biotechnol. 40(3), 338–353 (2022) https://doi.org/ 10.1016/J.TIBTECH.2021.07.001

  8. [8]

    Olvera, D., Sohrabi, M.M., Hendy, G., Monaghan, M.G.: Electroconductive melt electrowritten patches matching the mechanical anisotropy of human myocardium. Adv. Funct. Mater. 30(44), 1909880 (2020) https://doi.org/10. 1002/ADFM.201909880

Show all 67 references
  1. [9]

    Brazhkina, O., Park, J.H., Park, H., Bheri, S., Maxwell, J.T., Hollister, S.J., Davis, M.E.: Designing a 3d printing based auxetic cardiac patch with hipsc-cms for heart repair. J. Cardiovasc. Dev. Dis. 8(12), 172 (2021) https://doi.org/10. 3390/jcdd8120172

  2. [10]

    Asulin, M., Michael, I., Shapira, A., Dvir, T.: One-step 3d printing of heart patches with built-in electronics for performance regulation. Adv. Sci. 8(9), 2004205 (2021) https://doi.org/10.1002/ADVS.202004205

  3. [11]

    Lecina-Tejero, O., Perez, M.A., Garcia-Gareta, E., Borau, C.: The rise of mechan- ical metamaterials: Auxetic constructs for skin wound healing. J. Tissue Eng. 14, 1–16 (2023) https://doi.org/10.1177/20417314231177838

  4. [12]

    JMMCE 11, 27–35 (2023) https://doi.org/10.4236/jmmce.2023.112003

    Rose, S., Siu, D., Zhu, J.D., Roufail, R.: Auxetics in biomedical applications: A review. JMMCE 11, 27–35 (2023) https://doi.org/10.4236/jmmce.2023.112003

  5. [13]

    Mei, X., Cheng, K.: Recent development in therapeutic cardiac patches. Front. Cardiovasc. Med. 7, 610364 (2020) https://doi.org/10.3389/fcvm.2020.610364

  6. [14]

    Chansoria, P., Chaudhari, A., Etter, E.L., Bonacquisti, E.E., Heavey, M.K., Le, J., Maruthamuthu, M.K., Kussatz, C.C., Blackwell, J., Jasiewicz, N.E., Sellers, R.S., Maile, R., Wallet, S.M., Egan, T.M., Nguyen, J.: Instantly adhesive and ultra-elastic patches for dynamic organ...

  7. [15]

    Biomater

    Mardling, P., Alderson, A., Jordan-Mahy, N., Maitre, C.L.: The use of auxetic materials in tissue engineering. Biomater. Sci. 8(8), 2074–2083 (2020) https:// 24 doi.org/10.1039/C9BM01928F

  8. [16]

    ACS Appl

    Pradhan, S., Ventura, L., Agostinacchio, F., Xu, M., Barbieri, E., Motta, A., Pugno, N.M., Yadavalli, V.K.: Biofunctional silk kirigami with engineered prop- erties. ACS Appl. Mater. Interfaces. 12(11), 12436–12444 (2020) https://doi.org/ 10.1021/acsami.9b20691

  9. [17]

    Shu, T., Zheng, K., Zhang, Z., Ren, J., Wang, Z., Pei, Y., Yeo, J., Gu, G.X., Ling, S.: Birefringent silk fibroin hydrogel constructed via binary solvent-exchange- induced self-assembly. Biol. Macromol. 22(5), 1955–1965 (2021) https://doi.org/ 10.1021/acs.biomac.1c00065

  10. [18]

    Open Biol

    Abascal, N.C., Regan, L.: The past, present and future of protein-based materials. Open Biol. 8(10), 180113 (2018) https://doi.org/10.1098/rsob.180113

  11. [19]

    Harrington, M.J., Fratzl, P.: Natural load-bearing protein materials. Prog. Mater. Sci. 120, 100767 (2021) https://doi.org/10.1016/j.pmatsci.2020.100767

  12. [20]

    Hu, X., Cebe, P., Weiss, A.S., Omenetto, F., Kaplan, D.L.: Protein-based com- posite materials. Mater. Today 15(5), 208–215 (2012) https://doi.org/10.1016/ S1369-7021(12)70091-3

  13. [21]

    Miserez, A., Yu, J., Mohammadi, P.: Protein-based biological materials: Molecular design and artificial production. Chem. Rev. 123(5), 2049–2111 (2023) https: //doi.org/10.1021/acs.chemrev.2c00621

  14. [22]

    Sorushanova, A., Delgado, L.M., Wu, Z., Shologu, N., Kshirsagar, A., Raghunath, R., Mullen, A.M., Bayon, Y., Pandit, A., Raghunath, M., Zeugolis, D.I.: The collagen suprafamily: From biosynthesis to advanced biomaterial development. Adv. Mater. 31(1), 1801651 (2019) https://do...

  15. [23]

    Mu, X., Fitzpatrick, V., Kaplan, D.L.: From silk spinning to 3d printing: Poly- mer manufacturing using directed hierarchical molecular assembly. Adv. Healthc. Mater. 9(15), 1901552 (2020) https://doi.org/10.1002/adhm.201901552

  16. [24]

    Rockwood, D.N., Preda, R.C., Yucel, T., Wang, X., Lovett, M.L., Kaplan, D.L.: Materials fabrication from bombyx mori silk fibroin. Nat. Protoc. 6(10), 1612– 1631 (2011) https://doi.org/10.1038/nprot.2011.379

  17. [25]

    Zhang, W., Chen, J., Qu, M., Backman, L.J., Zhang, A., Liu, H., Zhang, X., Zhou, Q., Danielson, P.: Sustained release of tpca-1 from silk fibroin hydrogels preserves keratocyte phenotype and promotes corneal regeneration by inhibiting interleukin-1beta signaling. Adv. Healthc....

  18. [26]

    Guo, C., Li, C., Vu, H.V., Hanna, P., Lechtig, A., Qiu, Y., Mu, X., Ling, S., Nazarian, A., Lin, S.J., Kaplan, D.L.: Thermoplastic moulding of regenerated silk. 25 Nat. Mater. 19(1), 102–108 (2020) https://doi.org/10.1038/s41563-019-0560-8

  19. [27]

    Li, C., Guo, C., Fitzpatrick, V., Ibrahim, A., Zwierstra, M.J., Hanna, P., Lechtig, A., Nazarian, A., Lin, S.J., Kaplan, D.L.: Design of biodegradable, implantable devices towards clinical translation. Nat. Rev. Mater. 5(1), 61–81 (2020) https: //doi.org/10.1038/s41578-019-0150-z

  20. [28]

    Holland, C., Numata, K., Rnjak-Kovacina, J., Seib, F.P.: The biomedical use of silk: Past, present, future. Adv. Healthc. Mater. 8(1), 1800465 (2019) https: //doi.org/10.1002/adhm.201800465

  21. [29]

    Guo, C., Li, C., Mu, X., Kaplan, D.L.: Engineering silk materials: From natural spinning to artificial processing. Appl. Phys. Rev. 7(1), 011313 (2020) https: //doi.org/10.1063/1.5091442

  22. [30]

    Mu, X., Yuen, J.S., Choi, J., Zhang, Y., Cebe, P., Jiang, X., Zhang, Y.S., Kaplan, D.L.: Conformation-driven strategy for resilient and functional protein materials. Proc. Natl. Acad. Sci. 119(4), 2115523119 (2022) https://doi.org/10.1073/pnas. 2115523119

  23. [32]

    Biomaterials 32(10), 2642–2650 (2011) https://doi.org/10.1016/j.biomaterials

    Guziewicz, N., Best, A., Perez-Ramirez, B., Kaplan, D.L.: Lyophilized silk fibroin hydrogels for the sustained local delivery of therapeutic monoclonal antibodies. Biomaterials 32(10), 2642–2650 (2011) https://doi.org/10.1016/j.biomaterials. 2010.12.023

  24. [33]

    Zhu, Z., Ling, S., Yeo, J., Zhao, S., Tozzi, L., Buehler, M.J., Omenetto, F., Li, C., Kaplan, D.L.: High-strength, durable all-silk fibroin hydrogels with versatile processability toward multifunctional applications. Adv. Funct. Mater. 28(10), 1704757 (2018) https://doi.org/10...

  25. [34]

    Gupta, S., Jiskoot, W., Schoneich, C., Rathore, A.S.: Oxidation and deamidation of monoclonal antibody products: Potential impact on stability, biological activity, and efficacy. J. Pharm. Sci. 111(4), 903–918 (2022) https://doi.org/10.1016/j. xphs.2021.11.024

  26. [35]

    Molecules 27(7), 2148 (2022) https://doi.org/10.3390/ molecules27072148

    Mu, X., Gonzalez-Obeso, C., Xia, Z., Sahoo, J.K., Li, G., Cebe, P., Zhang, Y.S., Kaplan, D.L.: 3d printing of monolithic proteinaceous cantilevers using regenerated silk fibroin. Molecules 27(7), 2148 (2022) https://doi.org/10.3390/ molecules27072148

  27. [36]

    Macromol

    Mu, X., Wang, Y., Guo, C., Li, Y., Ling, S., Huang, W., Cebe, P., Hsu, H., Ferrari, F., Jiang, X., Xu, Q., Balduini, A., Omenetto, F.G., Kaplan, D.L.: 3d printing of 26 silk protein structures by aqueous solvent-directed molecular assembly. Macromol. Biosci. 20(1), 1900191 (20...

  28. [37]

    Biomaterials 26(15), 2775–2785 (2005) https://doi.org/10.1016/j.biomaterials.2004.07.044

    Kim, U., Park, J., Joo Kim, H., Wada, M., Kaplan, D.L.: Three-dimensional aqueous-derived biomaterial scaffolds from silk fibroin. Biomaterials 26(15), 2775–2785 (2005) https://doi.org/10.1016/j.biomaterials.2004.07.044

  29. [38]

    Andersson, M., Johansson, J., Rising, A.: Silk spinning in silkworms and spiders. Int. J. Mol. Sci. 17(8), 1290 (2016) https://doi.org/10.3390/ijms17081290

  30. [39]

    Mu, X., Amouzandeh, R., Vogts, H., Luallen, E., Arzani, M.: A brief review on the mechanisms and approaches of silk spinning-inspired biofabrication. Front. Bio- eng. Biotechnol. 11, 1252499 (2023) https://doi.org/10.3389/fbioe.2023.1252499

  31. [40]

    (eds.): Nonlinear Finite Elements for Continua and Structures

    Belytschko, T., Liu, W.K., Moran, B., Elkhodary, K. (eds.): Nonlinear Finite Elements for Continua and Structures. Wiley, New York (2014)

  32. [41]

    Chansoria, P., Shirwaiker, R.: 3d bioprinting of anisotropic engineered tissue con- structs with ultrasonically induced cell patterning. Addit. Manuf. 32, 101042 (2020) https://doi.org/10.1016/J.ADDMA.2020.101042

  33. [42]

    Chansoria, P., Shirwaiker, R.: Characterizing the process physics of ultrasound- assisted bioprinting. Sci. Rep. 9(1), 13889 (2019) https://doi.org/10.1038/ S41598-019-50449-W

  34. [43]

    Biofabrication 11(3), 035015 (2019) https://doi.org/10.1088/ 1758-5090/AB15CF

    Chansoria, P., Narayanan, L.K., Schuchard, K., Shirwaiker, R.: Ultrasound- assisted biofabrication and bioprinting of preferentially aligned three-dimensional cellular constructs. Biofabrication 11(3), 035015 (2019) https://doi.org/10.1088/ 1758-5090/AB15CF

  35. [44]

    Chansoria, P., Blackwell, J., Etter, E.L., Bonacquisti, E.E., Jasiewicz, N., Neal, T., Kamal, S.A., Hoque, J., Varghese, S., Egan, T., Nguyen, J.: Rationally designed anisotropic and auxetic hydrogel patches for adaptation to dynamic organs. Adv. Funct. Mater. 32(43), 2207590 ...

  36. [45]

    Javadi, A.A., Faramarzi, A., Farmani, R.: Design and optimization of microstruc- ture of auxetic materials. Eng. Comput. 29(3), 260–276 (2012) https://doi.org/ 10.1108/02644401211212398

  37. [46]

    Npj Comput

    Meier, T., Li, R., Mavrikos, S., Blankenship, B., Vangelatos, Z., Yildizdag, M.E., Grigoropoulos, C.P.: Obtaining auxetic and isotropic metamaterials in counter- intuitive design spaces: an automated optimization approach and experimental characterization. Npj Comput. Mater. 1...

  38. [47]

    Thin-Walled Struct

    Zhang, W., Wang, Y., Li, Z., Wang, X., Scarpa, F.: Design, analysis and opti- mization of the uniaxial properties of a new auxetic anti-chiral parallelogram metamaterial. Thin-Walled Struct. 202, 112119 (2024) https://doi.org/10.1016/ j.tws.2024.112119

  39. [48]

    Xiao, S.P., Li, J., Bordas, S.P., Kim, T.Y.: Artificial neural networks and their applications in computational materials science: A review and a case study. Adv. Appl. Mech. 57, 1–33 (2023) https://doi.org/10.1016/bs.aams.2023.09.001

  40. [49]

    Gurbuz, F., Mudireddy, A., Mantilla, R., Xiao, S.: Using a physics-based hydro- logical model and storm transposition to investigate machine-learning algorithms for streamflow prediction. J. Hydrol.628, 130504 (2024) https://doi.org/10.1016/ j.jhydrol.2023.130504

  41. [50]

    Thin-Walled Struct

    Yan, H., Yu, H., Zhu, S., Wang, Z., Zhang, Y., Guo, L.: Nonlinear properties prediction and inverse design of a porous auxetic metamaterial based on neural networks. Thin-Walled Struct. 197, 111717 (2024) https://doi.org/10.1016/j.tws. 2024.111717

  42. [51]

    Shen, X., Yan, K., Zhu, D., Hu, Q., Wu, H., Qi, S., Yuan, M., Qian, X.: Inverse machine learning framework for optimizing gradient honeycomb structure under impact loading. Eng. Struct. 309, 118079 (2024) https://doi.org/10.1016/ j.engstruct.2024.118079

  43. [52]

    Thin-Walled Struct

    Mohammadnejad, M., Montazeri, A., Bahmanpour, E., Mahnama, M.: Artificial neural networks for inverse design of a semi-auxetic metamaterial. Thin-Walled Struct. 200, 111927 (2024) https://doi.org/10.1016/j.tws.2024.111927

  44. [53]

    Peng, X., Xu, B.: Data-driven inverse design of composite triangular lattice struc- tures. Int. J. Mech. Sci. 265, 108900 (2024) https://doi.org/10.1016/j.ijmecsci. 2023.108900

  45. [54]

    Npj Comput

    Kumar, S., Tan, S., Zheng, L., Kochmann, D.M.: Inverse-designed spinodoid metamaterials. Npj Comput. Mater. 6(1), 73 (2020) https://doi.org/10.1038/ s41524-020-0341-6

  46. [55]

    Bastek, J., Kumar, S., Telgen, B., Glaesener, R.N., Kochmann, D.M.: Inverting the structure–property map of truss metamaterials by deep learning. Proc. Natl. Acad. Sci. 119(1), 2111505119 (2022) https://doi.org/10.1073/pnas.2111505119

  47. [56]

    Chen, Y., Deierling, P., Xiao, S.: Exploring active learning strategies for predictive models in mechanics of materials. Appl. Phys. A 130(8), 588 (2024) https://doi. org/10.1007/s00339-024-07728-9

  48. [57]

    Gibson, L.J., Ashby, M.F., Schajer, G.S., Robertson, C.I.: The mechanics of two- dimensional cellular materials. Proc. R. soc. Lond. Ser. A-Contain. Pap. Math. Phys. 382(1782), 25–42 (1982) https://doi.org/10.1098/rspa.1982.0087 28

  49. [58]

    Grima, J.N., Evans, K.E.: Auxetic behavior from rotating squares. J. Mater. Sci. Lett. 19(17), 1563–1565 (2000) https://doi.org/10.1023/A:1006781224002

  50. [59]

    Lakes, R.: Deformation mechanisms in negative poisson’s ratio materials: struc- tural aspects. J. Mater. Sci. 26(9), 2287–2292 (1991) https://doi.org/10.1007/ BF01130170

  51. [60]

    Alderson, K., Alderson, A., Ravirala, N., Simkins, V., Davies, P.: Manufacture and characterisation of thin flat and curved auxetic foam sheets. Phys. Status Solidi B 249(7), 1315–1321 (2012) https://doi.org/10.1002/pssb.201084215

  52. [61]

    Grima, J.N., Gatt, R.: Perforated sheets exhibiting negative poisson’s ratios. Adv. Eng. Mater. 12(6), 460–464 (2010) https://doi.org/10.1002/adem.201000005

  53. [62]

    Burton, W.S., Noor, A.K.: Assessment of continuum models for sandwich panel honeycomb cores. Comput. Methods Appl. Mech. Eng. 145(3-4), 341–360 (1997) https://doi.org/10.1016/S0045-7825(96)01196-6

  54. [63]

    Grima, J.N., Gatt, R., Alderson, A., Evans, K.E.: On the potential of connected stars as auxetic systems. Mol. Model. 31(13), 925–935 (2005) https://doi.org/10. 1080/08927020500401139

  55. [64]

    Zhang, Y., Xu, X., Fang, J., Huang, W., Wang, J.: Load characteristics of trian- gular honeycomb structures with self-similar hierarchical features. Eng. Struct. 257, 114114 (2022) https://doi.org/10.1016/j.engstruct.2022.114114

  56. [65]

    Dolla, W.J., Fricke, B.A., Becker, B.R.: Structural and drug diffusion models of conventional and auxetic drug-eluting stents. J. Med. Devices 1(1), 47–55 (2007) https://doi.org/10.1115/1.2355691

  57. [66]

    2017 IEEE International Conference on Computer Vision (ICCV), 2242–2251 (2017)

    Zhu, J., Park, T., Isola, P., Efros, A.A.: Unpaired image-to-image translation using cycle-consistent adversarial networks. 2017 IEEE International Conference on Computer Vision (ICCV), 2242–2251 (2017)

  58. [67]

    (ed.): Evolutionary Algorithms and Neural Networks

    Mirjalili, S. (ed.): Evolutionary Algorithms and Neural Networks. Springer, Chambersburg (2019)

  59. [68]

    Xiao, S., She, B., Mehta, S., Kan, Z.: Design of controllable leader–follower networks via memetic algorithms. Adv. Complex Syst. 24(02), 2150004 (2021) https://doi.org/10.1142/S0219525921500041 29

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.