Skip to main content

Advertisement

SpringerLink
Log in

Recurrent neural network to predict hyperelastic constitutive behaviors of the skeletal muscle

  • Original Article
  • Published:
Medical & Biological Engineering & Computing Aims and scope Submit manuscript

Abstract

Hyperelastic constitutive laws have been commonly used to model the passive behavior of the human skeletal muscle. Despite many efforts, the use of accurate finite element formulations of hyperelastic constitutive laws is still time-consuming for a real-time medical simulation system. The objective of the present study was to develop a deep learning model to predict the hyperelastic constitutive behaviors of the skeletal muscle toward a fast estimation of the muscle tissue stress.

A finite element (FE) model of the right psoas muscle was developed. Neo-Hookean and Mooney-Rivlin laws were used. A tensile test was performed with an applied body force. A learning database was built from this model using an automatic probabilistic generation process. A long-short term memory (LSTM) neural network was implemented to predict the stress evolution of the skeletal muscle tissue. A hyperparameter tuning process was conducted. Root mean square error (RMSE) and associated relative error was quantified to evaluate the precision of the predictive capacity of the developed deep learning model. Pearson correlation coefficients (R) was also computed.

The nodal displacements and the maximal stresses range from 70 to 227 mm and from 2.79 to 5.61 MPa for Neo-Hookean and Monney-Rivlin laws, respectively. Regarding the LSTM predictions, the RMSE ranges from 224.3 ± 3.9 Pa (8%) to 227.5 \(\pm\) 5.7 Pa (4%) for Neo-Hookean and Monney-Rivlin laws, respectively. Pearson correlation coefficients (R) of 0.78 \(\pm\) 0.02 and 0.77 \(\pm\) 0.02 were obtained for Neo-Hookean and Monney-Rivlin laws, respectively.

The present study showed that, for the first time, the use of a deep learning model can reproduce the time-series behaviors of the complex FE formulations for skeletal muscle modeling. In particular, the use of a LSTM neural network leads to a fast and accurate surrogate model for the in silico prediction of the hyperelastic constitutive behaviors of the skeletal muscle. As perspectives, the developed deep learning model will be integrated into a real-time medical simulation of the skeletal muscle for prosthetic socket design and childbirth simulator.

Graphical abstract

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7

Similar content being viewed by others

References

  1. Hill AV (1938) The heat of shortening and the dynamic constants of muscle. Proc R Soc Lond B 126(843):136–195

    Article  Google Scholar 

  2. Tsugorka A, Rios E, Blatter L (1995) Imaging elementary events of calcium release in skeletal muscle cells. Science 269(5231):1723–1726

    Article  CAS  Google Scholar 

  3. Wu H, Kanatous SB, Thurmond FA, Gallardo T, Isotani E, Bassel-Duby R, Williams RS (2002) Regulation of mitochondrial biogenesis in skeletal muscle by CaMK. Science 296(5566):349–352

    Article  CAS  Google Scholar 

  4. Dao TT, Ho Ba Tho MC (2018) A systematic review of continuum modeling of skeletal muscles: current trends, limitations and recommendations. Appl Bionics Biomech 2018:7631818

    Article  Google Scholar 

  5. Sengupta S, Basak S, Saikia P, Paul S, Tsalavoutis V, Atiah F, Ravi V, Peters A (2020) A review of deep learning with special emphasis on architectures, applications and recent trends. Knowl Based Syst 194:105596

    Article  Google Scholar 

  6. Shrestha A, Mahmood A (2019) Review of deep learning algorithms and architectures. IEEE Access 7:53040–53065. https://doi.org/10.1109/ACCESS.2019.2912200

    Article  Google Scholar 

  7. LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444

    Article  CAS  Google Scholar 

  8. Kruger N, Janssen P, Kalkan S, Lappe M, Leonardis A, Piater J, Rodriguez-Sanchez AJ, Wiskott L (2013) Deep hierarchies in the primate visual cortex: what can we learn for computer vision?. IEEE Trans Pattern Anal Mach Intell 35(8):1847–1871

    Article  Google Scholar 

  9. LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proceedings of IEEE 86(11):2278–2324

    Article  Google Scholar 

  10. Mukherjee P, Zhou M, Lee E, Schicht A, Balagurunathan Y, Napel S, Gillies R, Wong S, Thieme A, Leung A, Gevaert O (2020) A shallow convolutional neural network predicts prognosis of lung cancer patients in multi-institutional computed tomography image datasets. Nat Mach Intell 2:274–282

    Article  Google Scholar 

  11. Kim H, Jung J, Kim J, Cho B, Kwak J, Jang JY, Lee SW, Lee JG, Yoon SM (2020) Abdominal multi-organ auto-segmentation using 3D-patch-based deep convolutional neural network. Sci Rep 10:6204

    Article  CAS  Google Scholar 

  12. Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9(8):1735–1780

    Article  CAS  Google Scholar 

  13. Dao TT (2019) From deep learning to transfer learning for the prediction of skeletal muscle forces. Med Biol Eng Compu 57(5):1049–1058

    Article  Google Scholar 

  14. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y (2014) Generative adversarial networks. Proceedings of the International Conference on Neural Information Processing Systems (NIPS 2014), pp 2672–2680

  15. Kaelbling LP, Littman ML, Moore AW (1996) Reinforcement learning: a survey. J Artif Intell Res 4:237–285

    Article  Google Scholar 

  16. Schmidt J, Marques MRG, Botti S, Marques MAL (2019) Recent advances and applications of machine learning in solid-state materials science. Npj Comput Mater 83. https://doi.org/10.1038/s41524-019-0221-0

  17. Batra R (2021) Accurate machine learning in materials science facilitated by using diverse data sources. Nature 589:524–525

    Article  CAS  Google Scholar 

  18. Chen C, Zuo Y, Ye W, Li X, Ong SP (2021) Learning properties of ordered and disordered materials from multi-fidelity data. Nat Comput Sci 1:46–53. https://doi.org/10.1038/s43588-020-00002-x

    Article  Google Scholar 

  19. Liang L, Liu M, Sun W (2017) A deep learning approach to estimate chemically-treated collagenous tissue nonlinear anisotropic stress-strain responses from microscopy images. Acta Biomater 63:227–235

    Article  CAS  Google Scholar 

  20. Halilaj E, Rajagopal A, Fiterau M, Hicks JL, Hastie TJ, Delp SL (2018) Machine learning in human movement biomechanics: best practices, common pitfalls, and new opportunities. J Biomech 81:1–11

    Article  Google Scholar 

  21. Liang L, Mao W, Sun W (2020) A feasibility study of deep learning for predicting hemodynamics of human thoracic aorta. J Biomech 99:109544. https://doi.org/10.1016/j.jbiomech.2019.109544

    Article  PubMed  Google Scholar 

  22. Bustamante-Orellana C, Guachi R, Guachi-Guachi L, Novelli S, Campana F, Bini F, Marinozzi F (2020) Biomechanics of soft tissues: the role of the mathematical model on material behavior. In: Botto-Tobar M, León-Acurio J, Díaz Cadena A, Montiel Díaz P (eds) Advances in Emerging Trends and Technologies. ICAETT 2019. Advances in intelligent systems and computing, vol 1066. Springer. https://doi.org/10.1007/978-3-030-32022-5_29

  23. Affagard J, Bensamoun SF, Feissel P (2014) Development of an inverse approach for the characterization of in vivo mechanical properties of the lower limb muscles. J Biomech Eng 136(11):111012. https://doi.org/10.1115/1.4028490

    Article  Google Scholar 

  24. Avril S, Bouten L, Dubuis L, Drapier S, Pouget J (2010) Mixed experimental and numerical approach for characterizing the biomechanical response of the human leg under elastic compression. J Biomech Eng 132(3):031006

    Article  Google Scholar 

  25. Lapeer R, Gerikhanov Z, Sadulaev SM, Audinis V, Rowland R, Crozier K, Morris E (2019) A computer-based simulation of childbirth using the partial Dirichlet-Neumann contact method with total Lagrangian explicit dynamics on the GPU. Biomech Model Mechanobiol 18:681–700. https://doi.org/10.1007/s10237-018-01109-x

    Article  PubMed  Google Scholar 

  26. Linder-Ganz E, Shabshin N, Itzchak Y, Gefen A (2007) Assessment of mechanical conditions in sub-dermal tissues during sitting: a combined experimental-MRI and finite element approach. Biomech Model Mechanobiol 40(7):1443–1454

    Google Scholar 

  27. Wu T, Hung A, Mithraratne K (2014) Generating facial expressions using an anatomically accurate biomechanical model. IEEE Trans Visual Comput Graphics 20(11):1519–1529

    Article  Google Scholar 

  28. Lee W, Won BH, Cho SW (2016) Finite element modeling for predicting the contact pressure between a foam mattress and the human body in a supine position. Comput Methods Biomech Biomed Eng 20(1):104–117

    Article  Google Scholar 

  29. Noakes KF, Pullan AJ, Bissett IP, Cheng LK (2008) Subject specific finite elasticity simulations of the pelvic floor. J Biomech 41(14):3060–3065

    Article  Google Scholar 

  30. Sadler Z, Scott J, Drost J, Chen S, Roccabianca S, Bush TR (2018) Initial estimation of the in vivo material properties of the seated human buttocks and thighs. Int J Non-Linear Mech 107:77–85

    Article  Google Scholar 

  31. Silva E, Parente M, Brandão S, Mascarenhas T, Natal RJ (2019) Characterizing the biomechanical properties of the pubovisceralis muscle using a genetic algorithm and the finite element method. J Biomech Eng 141(1)

  32. CoxSL, Mithraratne K, Smith NP (2007) An anatomically based finite element model of the lower limbs in the seated posture. 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp 6326–6329. https://doi.org/10.1109/IEMBS.2007.4353802.

  33. Dao TT, Marin F, Pouletaut P, Aufaure P, Charleux F, Ho Ba Tho MC (2012) Estimation of accuracy of patient specific musculoskeletal modeling: case study on a post polio residual paralysis subject. Comput Method Biomech Biomed Eng 15(7):745–751

    Article  CAS  Google Scholar 

  34. Dao TT (2017) Advanced computational workflow for the multi-scale modeling of the bone metabolic processes. Med Biol Eng Comput 55(6):923–933

    Article  Google Scholar 

  35. Dao TT, Pouletaut P, Charleux F, Lazáry Á, Eltes P, Varga PP, Ho Ba Tho MC (2015) Multimodal medical imaging (CT and dynamic MRI) data and computer-graphics multi-physical model for the estimation of patient specific lumbar spine muscle forces. Data Knowl Eng 96–97:3–18

    Article  Google Scholar 

  36. Meister F, Passerini T, Mihalef V, Tuysuzoglu A, Maier A, Mansi T (2020) Deep learning acceleration of Total Lagrangian explicit dynamics for soft tissue mechanics. Comput Methods Appl Mech Eng 358:112628

    Article  Google Scholar 

  37. ter Horst R, van Weert H, Loonen T, Bergé S, Vinayahalingam S, Baan F, Maal T, de Jong G, Xi T (2021) Three-dimensional virtual planning in mandibular advancement surgery: soft tissue prediction based on deep learning. J Craniomaxillofac Surg. https://doi.org/10.1016/j.jcms.2021.04.001

    Article  PubMed  Google Scholar 

  38. Medvedev AV, Agoureeva GI, Murro AM (2019) A Long short-term memory neural network for the detection of epileptiform spikes and high frequency oscillations. Sci Rep 9:19374. https://doi.org/10.1038/s41598-019-55861-w

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Kumar S, Sharma A, Tsunoda T (2019) Brain wave classification using long short-term memory network based OPTICAL predictor. Sci Rep 9:9153. https://doi.org/10.1038/s41598-019-45605-1

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Qiu J, Wang B, Zhou C (2020) Forecasting stock prices with long-short term memory neural network based on attention mechanism. PLoS ONE 15(1):e0227222

    Article  CAS  Google Scholar 

Download references

Funding

This present study was funded by the Métropole Européenne de Lille (MEL) and the I-SITE ULNE (Université Lille Nord Europe).

Author information

Authors and Affiliations

  1. Univ. Lille, CNRS, Centrale Lille, UMR 9013, LaMcube, Laboratoire de Mécanique, Multiphysique, Multiéchelle, 59655, Villeneuve d’Ascq Cedex, F-59000, Lille, France

    Abbass Ballit & Tien-Tuan Dao

Authors
  1. Abbass Ballit
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Tien-Tuan Dao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tien-Tuan Dao.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ballit, A., Dao, TT. Recurrent neural network to predict hyperelastic constitutive behaviors of the skeletal muscle. Med Biol Eng Comput 60, 1177–1185 (2022). https://doi.org/10.1007/s11517-022-02541-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11517-022-02541-z

Keywords

Search

Navigation