Integrating machine learning and finite element simulation for interpretable prediction of 3D-printed bone scaffold mechanics

Abstract

An integrated framework combining finite element analysis (FEA) and artificial neural networks (ANN) is presented to enhance the prediction and design of bioprinted scaffolds. By leveraging the strengths of data-driven learning and physics-based simulations, the hybrid approach (ANN + FEA) achieved superior predictive accuracy and generalization compared to standalone approaches. Validation against experimental results demonstrated that a single ANN model yields a relative error of 5.17% when predicting the scaffold’s Young’s modulus. Incorporating FEA simulation based on ANN-predicted geometry and material properties reduced the relative error to 4.72%, representing an 8.6% improvement. The framework also enables accurate simulation of unseen combinations of printing parameters located far from the experimental data manifold, reducing prediction errors from 14.2% (ANN-only) to 5.7% (hybrid). By integrating predictive modeling, simulation, and data augmentation, this approach offers an efficient pathway for optimizing scaffold designs and accelerating the development of biomaterials with tailored mechanical performance.

Keywords

3D printing / Artificial neural network / Bone scaffold / Data augmentation / Finite element analysis

Funding

This work was supported by the Engineering and Physical Sciences Research Council Impact Acceleration Account 2022–2026 (EP/X525674/1).

Conflict of Interest

Fengyuan Liu serves as the Editorial Board Member of the journal but was not involved in the editorial or peer-review process conducted for this paper, either directly or indirectly. The Other authors declare no competing interests.

References

[1] O’Brien FJ. Biomaterials and scaffolds for tissue engineering. Mater Today. 2011;14(3):88-95. doi: 10.1016/S1369-7021(11)70058-X

[2] Hutmacher DW. Scaffolds in tissue engineering bone and cartilage. Biomaterials. 2000;21(24):2529-2543. doi: 10.1016/S0142-9612(00)00121-6

[3] Bose S, Roy M, Bandyopadhyay A. Recent advances in bone tissue engineering scaffolds. Trends Biotechnol. 2012;30(10):546-554. doi: 10.1016/j.tibtech.2012.07.005

[4] Feng Y, Zhu S, Mei D, et al. Application of 3D printing technology in bone tissue engineering: a review. Curr Drug Deliv. 2021;18:847-861. doi: 10.2174/1567201817999201113100322

[5] Xu Y, Zhang F, Zhai W, Cheng S, Li J, Wang Y. Unraveling of advances in 3D-printed polymer-based bone scaffolds. Polymers (Basel). 2022;14:566. doi: 10.3390/polym14030566

[6] Hussain M, Khan SM, Shafiq M, Abbas N. A review on PLA-based biodegradable materials for biomedical applications. Giant. 2024;18(3):100261. doi: 10.1016/j.giant.2024.100261

[7] Afonso J, Alves JL, Caldas G, Gouveia BP, Santana L, Belinha J. Influence of 3D printing process parameters on the mechanical properties and mass of PLA parts and predictive models. Rapid Prototyp J. 2021;27(3):487-495. doi: 10.1108/RPJ-03-2020-0043

[8] Ma S, Tang Q, Liu Y, Feng Q. Prediction of mechanical properties of three-dimensional printed lattice structures through machine learning. J Comput Inf Sci Eng. 2022;22(3):031008. doi: 10.1115/1.4053077

[9] Sheoran AJ, Kumar H. Fused deposition modeling process parameters optimization and effect on mechanical properties and part quality: review and reflection on present research. Mater Today Proc. 2020;21:1659-1672. doi: 10.1016/j.matpr.2019.11.296

[10] Goh GD, Sing SL, Yeong WY. A review on machine learning in 3D printing: applications, potential, and challenges. Artif Intell Rev. 2021;54(1):63-94. doi: 10.1007/s10462-020-09876-9

[11] Goh GD, Sing SL, Lim YF, et al. Machine learning for 3D-printed multi-materials tissue-mimicking anatomical models. Mater Des. 2021;211:110125. doi: 10.1016/j.matdes.2021.110125

[12] Jayasudha M, Elangovan M, Mahdal M, Priyadarshini J. Accurate estimation of tensile strength of 3D-printed parts using machine learning algorithms. Processes. 2022;10(6):1158. doi: 10.3390/pr10061158

[13] Rodrigues JF, Florea L, de Oliveira MC, Diamond D, Oliveira ON. Big data and machine learning for materials science. Discover Mater. 2021;1:1-27. doi: 10.1007/s43939-021-00012-0

[14] Oviedo F, Ferres JL, Buonassisi T, Butler KT. Interpretable and explainable machine learning for materials science and chemistry. Acc Mater Res. 2022;3(6):597-607. doi: 10.1021/accountsmr.1c00244

[15] Kalina KA, Linden L, Brummund J, Kästner M. FE-ANN: an efficient data-driven multiscale approach based on physics-constrained neural networks and automated data mining. Comput Mech. 2023;71(5):827-851. doi: 10.1007/s00466-022-02260-0

[16] Peng B, Wei Y, Qin Y, et al. Machine learning-enabled constrained multi-objective design of architected materials. Nat Commun. 2023;14(1):6630. doi: 10.1038/s41467-023-42415-y

[17] Oladipo B, Matos H, Krishnan NA, Das SJ. Integrating experiments, finite element analysis, and interpretable machine learning to evaluate the auxetic response of 3D-printed re-entrant metamaterials. J Mater Res Technol. 2023;25:1612-1625. doi: 10.1016/j.jmrt.2023.06.038

[18] Drakoulas G, Gortsas T, Polyzos E, et al. An explainable machine learning-based probabilistic framework for the design of scaffolds in bone tissue engineering. Biomech Model Mechanobiol. 2024;23(3):987-1012. doi: 10.1007/s10237-024-01817-7

[19] Dong H, Liu F, Ye L, et al. Process optimization and mechanical property investigation of Inconel 718 manufactured by selective electron beam melting. Mater Sci Addit Manuf. 2022;1(4):23. doi: 10.18063/msam.v1i4.23

[20] Rodrigues N, Benning M, Ferreira AM, Dixon L, Dalgarno K. Manufacture and characterisation of porous PLA scaffolds. Procedia CIRP. 2016;49:33-38. doi: 10.1016/j.procir.2015.07.025

[21] Frunzaverde D, Cojocaru V, Ciubotariu CR, et al. The influence of the printing temperature and the filament color on the dimensional accuracy, tensile strength, and friction performance of FFF-printed PLA specimens. Polymers (Basel). 2022;14(10):1978. doi: 10.3390/polym14101978

[22] Lei M, Wei Q, Li M, Zhang J, Yang R, Wang Y. Numerical simulation and experimental study of the effects of process parameters on filament morphology and mechanical properties of FDM 3D-printed PLA/GNPs nanocomposite. Polymers (Basel). 2022;14(15):3081. doi: 10.3390/polym14153081

[23] Wickramasinghe S, Do T, Tran P. FDM-based 3D printing of polymer and associated composite: a review on mechanical properties, defects, and treatments. Polymers (Basel). 2020;12(7):1529. doi: 10.3390/polym12071529

[24] Heidari-Rarani M, Ezati N, Sadeghi P, Badrossamay M. Optimization of FDM process parameters for tensile properties of polylactic acid specimens using Taguchi design of experiment method. J Thermoplast Compos Mater. 2022;35(12):2435-2452. doi: 10.1177/0892705720964560

[25] Chokshi H, Shah DB, Patel KM, Joshi SJ. Experimental investigations of process parameters on mechanical properties for PLA during processing in FDM. Adv Mater Process Technol. 2022;8(suppl 2):696-709. doi: 10.1080/2374068X.2021.1946756

[26] Farah S, Anderson DG, Langer R. Physical and mechanical properties of PLA and their functions in widespread applications—a comprehensive review. Adv Drug Deliv Rev. 2016;107:367-392. doi: 10.1016/j.addr.2016.06.012

[27] Zienkiewicz OC, Taylor RL, Zhu JZ. Chapter 10: Incompressible problems, mixed methods, and other procedures of solution. In: The Finite Element Method: Its Basis and Fundamentals. 7th ed.; 2013:315-359. doi: 10.1016/B978-1-85617-633-0.00010-1

[28] Gomes MS, Carmo GP, Ptak M, Fernandes FA, Alves de Sousa RJ. Accuracy and efficiency of finite element head models: the role of finite element formulation and material laws. Int J Numer Methods Biomed Eng. 2024;40(9):e3851. doi: 10.1002/cnm.3851

[29] Zienkiewicz OC, Taylor RL, Zhu JZ. Chapter 16: Adaptive finite element refinement. In: The Finite Element Method: Its Basis and Fundamentals. 7th ed.; 2013:545-572. doi: 10.1016/B978-1-85617-633-0.00016-2

[30] Demir M, Seki Y. Interfacial adhesion strength between FDM-printed PLA parts and surface-treated cellulosic-woven fabrics. Rapid Prototyp J. 2023;29(6):1166-1174. doi: 10.1108/RPJ-10-2022-0369

[31] Rankouhi B, Javadpour S, Delfanian F, Letcher T. Failure analysis and mechanical characterization of 3D-printed ABS with respect to layer thickness and orientation. J Fail Anal Prev. 2016;16(3):467-481. doi: 10.1007/s11668-016-0113-2

[32] Sagias V, Giannakopoulos K, Stergiou C. Mechanical properties of 3D-printed polymer specimens. Procedia Struct Integr. 2018;10:85-90. doi: 10.1016/j.prostr.2018.09.013

[33] Xu L, Leguillon D. Dual-notch void model to explain the anisotropic strengths of 3D-printed polymers. J Eng Mater Technol. 2020;142(3):031001. doi: 10.1115/1.4044282

[34] Ahn SH, Montero M, Odell D, Roundy S, Wright PK. Anisotropic material properties of fused deposition modeling ABS. Rapid Prototyp J. 2002;8(4):248-257. doi: 10.1108/13552540210441166

[35] Chen J, Liu X, Tian Y, et al. 3D-printed anisotropic polymer materials for functional applications. Adv Mater. 2022;34(5):2102877. doi: 10.1002/adma.202102877

[36] Castelló-Pedrero P, García-Gascón C, García-Manrique J. Multiscale numerical modeling of large-format additive manufacturing processes using carbon fiber reinforced polymer for digital twin applications. Int J Mater Form. 2024;17(2):15. doi: 10.1007/s12289-024-01811-5

[37] He K, Zhang X, Ren S, Sun J. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV); 2015:1026-1034. doi: 10.1109/ICCV.2015.123

[38] Varma S, Simon R. Bias in error estimation when using cross-validation for model selection. BMC Bioinformatics. 2006;7(1):91. doi: 10.1186/1471-2105-7-91

[39] Banerjee S, Lio P, Jones PB, Cardinal RN. A class-contrastive human-interpretable machine learning approach to predict mortality in severe mental illness. npj Schizophr. 2021;7(1):60. doi: 10.1038/s41537-021-00191-y

[40] Schnaubelt M. A comparison of machine learning model validation schemes for non-stationary time series data. Preprint. 2019.

[41] Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks. In: JMLR Workshop Conf Proc. 2011;315-323.

[42] Vakalopoulou M, Christodoulidis S, Burgos N, Colliot O, Lepetit VJ. Deep Learning: Basics and Convolutional Neural Networks (CNNs); 2023:77-115. doi: 10.1007/978-1-0716-3195-9_3

[43] Cantero-Chinchilla S, Simpson CA, Ballisat A, Croxford AJ, Wilcox PD. Convolutional neural networks for ultrasound corrosion profile time series regression. NDT E Int. 2023;133:102756. doi: 10.1016/j.ndteint.2022.102756

[44] Rahaman R, Thiery AH. Uncertainty quantification and deep ensembles. Adv Neural Inf Process Syst. 2021;34:20063-20075. doi: 10.48550/arXiv.2007.08792

[45] Ovadia Y, Fertig E, Ren J, et al. Can you trust your model’s uncertainty? Evaluating predictive uncertainty under dataset shift. Adv Neural Inf Process Syst. 2019;32. doi: 10.48550/arXiv.1906.02530

[46] Fort S, Hu H, Lakshminarayanan B. Deep ensembles: a loss landscape perspective. In: International Conference on Learning Representations (ICLR); 2020. doi: 10.

[47] Pepelnjak T, Sevšek L, Movrin D, Milutinović M. Influence of process parameters on the characteristics of additively manufactured parts made from advanced biopolymers. Polymers (Basel). 2023;15(3):716. doi: 10.3390/polym15030716

[48] Wang C, Tan XP, Tor SB, Lim CJ. Machine learning in additive manufacturing: state-of-the-art and perspectives. Addit Manuf. 2020;36:101538. doi: 10.1016/j.addma.2020.101538

[49] Wu H, Sulkis M, Driver J, Saade-Castillo A, Thompson A, Koo JH. Multi-functional ULTEM™ 1010 composite filaments for additive manufacturing using fused filament fabrication (FFF). Addit Manuf. 2018;24:298-306. doi: 10.1016/j.addma.2018.10.014

[50] Zhao X, Li S, Zhang M, et al. Comparison of the microstructures and mechanical properties of Ti–6Al–4V fabricated by selective laser melting and electron beam melting. Mater Des. 2016;95:21-31. doi: 10.1016/j.matdes.2015.12.135

[51] Dejene ND, Lemu HG. Characterisation and prediction of mechanical properties in laser powder bed fusion-printed parts: a comparative analysis using machine learning. Mater Tehnol. 2024;39(1):2419228. doi: 10.1080/10667857.2024.2419228

[52] Ibrahimi S, D’Andrea L, Gastaldi D, Rivolta MW, Vena P. Machine learning approaches for the design of biomechanically compatible bone tissue engineering scaffolds. Comput Methods Appl Mech Eng. 2024;423:116842. doi: 10.1016/j.cma.2024.116842

[53] Rudolph M, Kurz S, Rakitsch J. Hybrid modeling design patterns. J Math Ind. 2024;14(1):3. doi: 10.1186/s13362-024-00141-0

[54] Lamane H, Mouhir L, Moussadek R, Baghdad B, Kisi O, El Bilali A. Interpreting machine learning models based on SHAP values in predicting suspended sediment concentration. Int J Sediment Res. 2025;40(1):91-107. doi: 10.1016/j.ijsrc.2024.10.002

[55] Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst. 2017;30. doi: 10.48550/arXiv.1705.07874

[56] Rooks NB, Besier TF, Schneider MT. A parameter sensitivity analysis on multiple finite element knee joint models. Front Bioeng Biotechnol. 2022;10:841882. doi: 10.3389/fbioe.2022.841882

[57] Gal Y, Ghahramani Z. Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning. In: Proceedings of the 33rd International Conference on Machine Learning (ICML 2016). Proc Mach Learn Res. 2016;48:1050-1059. https://proceedings.mlr.press/v48/gal16.html

[58] Kendall A, Gal Y. What uncertainties do we need in Bayesian deep learning for computer vision? Adv Neural Inf Process Syst. 2017;30:5575-5585.

[59] Bock FE, Keller S, Huber N, Klusemann B. Hybrid modelling by machine learning corrections of analytical model predictions towards high-fidelity simulation solutions. Materials (Basel). 2021;14(8):1883. doi: 10.3390/ma14081883

[60] Aldakheel F, Elsayed ES, Heider Y, Weeger O. Physics-based machine learning for computational fracture mechanics. Mach Learn Comput Sci Eng. 2025;1(1):18. doi: 10.1007/s44379-025-00019-x

[61] Napolitano F, Cozzolino E, Papa I, Astarita A, Squillace A. Experimental integrated approach for mechanical characteristic optimization of FDM-printed PLA in an energy-saving perspective. Int J Adv Manuf Technol. 2022;121(5):3551-3565. doi: 10.1007/s00170-022-09535-z

[62] Yao H, Gao Y, Liu Y. FEA-Net: A physics-guided data-driven model for efficient mechanical response prediction. Comput Methods Appl Mech Eng. 2020;363:112892. doi: 10.1016/j.cma.2020.112892

[63] Bahraminasab M. Challenges on optimization of 3D-printed bone scaffolds. BioMed Eng Online. 2020;19(1):69. doi: 10.1186/s12938-020-00810-2

[64] Akbari P, Zamani M, Mostafaei A. Machine learning prediction of mechanical properties in metal additive manufacturing. Addit Manuf. 2024;91:104320. doi: 10.1016/j.addma.2024.104320

[65] Ng WL, Goh GL, Goh GD, Ten JSJ, Yeong WY. Progress and opportunities for machine learning in materials and processes of additive manufacturing. Adv Mater. 2024;36(34):2310006. doi: 10.1002/adma.202310006

[66] Mohamed AP, Lee B, Zhang Y, et al. Simulation-enhanced data augmentation for machine learning pathloss prediction. In: ICC 2024 - IEEE International Conference on Communications; 2024: 4863-4868.