{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T16:27:20Z","timestamp":1781886440477,"version":"3.54.5"},"posted":{"date-parts":[[2023,3,28]]},"group-title":"Bioengineering","reference-count":61,"publisher":"openRxiv","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2023,3,28]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>Finite-element analysis (FEA) is widely used as a standard tool for stress and deformation analysis of solid structures, including human tissues and organs. For instance, FEA can be applied at a patient-specific level to assist in medical diagnosis and treatment planning, such as risk assessment of thoracic aortic aneurysm rupture\/dissection. These FEA-based biomechanical assessments often involve both forward and inverse mechanics problems. Current commercial FEA software packages (e.g., Abaqus) and inverse methods exhibit performance issues in either accuracy or speed.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>In this study, we propose and develop a new library of FEA code and methods, named PyTorch-FEA, by taking advantage of autograd, an automatic differentiation mechanism in PyTorch. We develop a class of PyTorch-FEA functionalities to solve forward and inverse problems with improved loss functions, and we demonstrate the capability of PyTorch-FEA in a series of applications related to human aorta biomechanics. In one of the inverse methods, we combine PyTorch-FEA with deep neural networks (DNNs) to further improve performance.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>We applied PyTorch-FEA in four fundamental applications for biomechanical analysis of human aorta. In the forward analysis, PyTorch-FEA achieved a significant reduction in computational time without compromising accuracy compared with Abaqus, a commercial FEA package. Compared to other inverse methods, inverse analysis with PyTorch-FEA achieves better performance in either accuracy or speed, or both if combined with DNNs.<\/jats:p>\n                <\/jats:sec>","DOI":"10.1101\/2023.03.27.533816","type":"posted-content","created":{"date-parts":[[2023,3,28]],"date-time":"2023-03-28T21:10:30Z","timestamp":1680037830000},"source":"Crossref","is-referenced-by-count":2,"title":["PyTorch-FEA: Autograd-enabled Finite Element Analysis Methods with Applications for Biomechanical Analysis of Human Aorta"],"prefix":"10.64898","author":[{"given":"Liang","family":"Liang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Minliang","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John","family":"Elefteriades","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2023033015300552000_2023.03.27.533816v1.1","doi-asserted-by":"publisher","DOI":"10.1080\/10255842.2012.746676"},{"key":"2023033015300552000_2023.03.27.533816v1.2","doi-asserted-by":"publisher","DOI":"10.1007\/s11517-012-0864-1"},{"key":"2023033015300552000_2023.03.27.533816v1.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcin.2016.01.003"},{"key":"2023033015300552000_2023.03.27.533816v1.4","doi-asserted-by":"publisher","DOI":"10.1016\/j.jtcvs.2009.02.057"},{"key":"2023033015300552000_2023.03.27.533816v1.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2015.10.048"},{"key":"2023033015300552000_2023.03.27.533816v1.6","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2014.06.007"},{"key":"2023033015300552000_2023.03.27.533816v1.7","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-bioeng-071813-104517"},{"key":"2023033015300552000_2023.03.27.533816v1.8","doi-asserted-by":"publisher","DOI":"10.1007\/s10237-014-0583-7"},{"key":"2023033015300552000_2023.03.27.533816v1.9","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbiomech.2012.05.008"},{"key":"2023033015300552000_2023.03.27.533816v1.10","doi-asserted-by":"publisher","DOI":"10.1152\/ajpheart.00908.2014"},{"issue":"12","key":"2023033015300552000_2023.03.27.533816v1.11","doi-asserted-by":"crossref","first-page":"9392","DOI":"10.1016\/j.actbio.2013.07.044","article-title":"Predictive biomechanical analysis of ascending aortic aneurysm rupture potential","volume":"9","year":"2013","journal-title":"Acta Biomaterialia"},{"key":"2023033015300552000_2023.03.27.533816v1.12","unstructured":"CDC. \u201cCenters for Disease Control and Prevention, National Center for Injury Prevention and Control, WISQARS Leading Causes of Death Reports, 1999 - 2018: https:\/\/webappa.cdc.gov\/cgi-bin\/broker.exe.\u201c (accessed Sept 15, 2020)."},{"key":"2023033015300552000_2023.03.27.533816v1.13","doi-asserted-by":"crossref","unstructured":"T. Faggion Vinholo , M. A. Zafar , B. A. Ziganshin , and J. A. Elefteriades , \u201cNonsyndromic Thoracic Aortic Aneurysms and Dissections\u2014Is Screening Possible?,\u201d Seminars in Thoracic and Cardiovascular Surgery, 2019\/06\/15\/ 2019, doi: https:\/\/doi.org\/10.1053\/j.semtcvs.2019.05.035.","DOI":"10.1053\/j.semtcvs.2019.05.035"},{"key":"2023033015300552000_2023.03.27.533816v1.14","doi-asserted-by":"publisher","DOI":"10.1038\/nrcardio.2016.211"},{"key":"2023033015300552000_2023.03.27.533816v1.15","doi-asserted-by":"publisher","DOI":"10.1088\/2516-1091\/ab9a29"},{"key":"2023033015300552000_2023.03.27.533816v1.16","doi-asserted-by":"crossref","unstructured":"T. C. Gasser , \u201cThe Biomechanical Rupture Risk Assessment of Abdominal Aortic Aneurysms\u2014Method and Clinical Relevance,\u201d in Biomedical Technology: Modeling, Experiments and Simulation, P. Wriggers and T. Lenarz Eds. Cham: Springer International Publishing, 2018, pp. 233\u2013253.","DOI":"10.1007\/978-3-319-59548-1_13"},{"key":"2023033015300552000_2023.03.27.533816v1.17","doi-asserted-by":"publisher","DOI":"10.12945\/j.aorta.2015.15.030"},{"key":"2023033015300552000_2023.03.27.533816v1.18","doi-asserted-by":"publisher","DOI":"10.1016\/j.actbio.2019.08.017"},{"issue":"5","key":"2023033015300552000_2023.03.27.533816v1.19","doi-asserted-by":"crossref","first-page":"1519","DOI":"10.1007\/s10237-017-0903-9","article-title":"A Machine Learning Approach to Investigate the Relationship between Shape Features and Numerically Predicted Risk of Ascending Aortic Aneurysm","volume":"16","year":"2017","journal-title":"Biomechanics and Modeling in Mechanobiology"},{"key":"2023033015300552000_2023.03.27.533816v1.20","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.jmbbm.2013.03.014","article-title":"In vivo determination of elastic properties of the human aorta based on 4D ultrasound data","volume":"27","year":"2013","journal-title":"Journal of the Mechanical Behavior of Biomedical Materials"},{"key":"2023033015300552000_2023.03.27.533816v1.21","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1016\/j.jmbbm.2015.09.022","article-title":"A finite element updating approach for identification of the anisotropic hyperelastic properties of normal and diseased aortic walls from 4D ultrasound strain imaging","volume":"58","year":"2016","journal-title":"Journal of the Mechanical Behavior of Biomedical Materials"},{"key":"2023033015300552000_2023.03.27.533816v1.22","doi-asserted-by":"crossref","first-page":"649","DOI":"10.1016\/j.jmbbm.2017.10.022","article-title":"Estimation of in vivo mechanical properties of the aortic wall: A multiresolution direct search approach","volume":"77","year":"2018","journal-title":"Journal of the Mechanical Behavior of Biomedical Materials"},{"key":"2023033015300552000_2023.03.27.533816v1.23","doi-asserted-by":"crossref","unstructured":"M. Liu et al., \u201cOn the Identification of Heterogeneous Nonlinear Material Properties of the Aortic Wall from Clinical Gated CT Scans,\u201d International Conference on Biomechanics andMedical Engineering, 2019.","DOI":"10.32604\/mcb.2019.07387"},{"issue":"1","key":"2023033015300552000_2023.03.27.533816v1.24","doi-asserted-by":"crossref","first-page":"12983","DOI":"10.1038\/s41598-019-49438-w","article-title":"Identification of in vivo nonlinear anisotropic mechanical properties of ascending thoracic aortic aneurysm from patient-specific CT scans","volume":"9","year":"2019","journal-title":"Scientific Reports"},{"key":"2023033015300552000_2023.03.27.533816v1.25","unstructured":"A. Paszke et al., \u201cPyTorch: An Imperative Style, High-Performance Deep Learning Library,\u201d Neural Information Processing Systems, pp. 8024\u20138035, 2019."},{"key":"2023033015300552000_2023.03.27.533816v1.26","unstructured":"M. Fey and J. E. Lenssen , \u201cFast Graph Representation Learning with PyTorch Geometric,\u201d ICLR Workshop on Representation Learning on Graphs and Manifolds, 2019."},{"key":"2023033015300552000_2023.03.27.533816v1.27","doi-asserted-by":"publisher","DOI":"10.3949\/ccjm.89a.21101"},{"key":"2023033015300552000_2023.03.27.533816v1.28","doi-asserted-by":"publisher","DOI":"10.1002\/jbm.a.34884"},{"key":"2023033015300552000_2023.03.27.533816v1.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcmg.2017.04.005"},{"key":"2023033015300552000_2023.03.27.533816v1.30","doi-asserted-by":"publisher","DOI":"10.1098\/rsif.2005.0073"},{"key":"2023033015300552000_2023.03.27.533816v1.31","doi-asserted-by":"crossref","first-page":"104539","DOI":"10.1016\/j.jmps.2021.104539","article-title":"A probabilistic and anisotropic failure metric for ascending thoracic aortic aneurysm risk assessment","volume":"155","year":"2021","journal-title":"Journal of the Mechanics and Physics of Solids"},{"key":"2023033015300552000_2023.03.27.533816v1.32","doi-asserted-by":"crossref","unstructured":"J. Bonet and R. D. Wood , Nonlinear Continuum Mechanics for Finite Element Analysis. Cambridge University Press, 2008.","DOI":"10.1017\/CBO9780511755446"},{"key":"2023033015300552000_2023.03.27.533816v1.33","doi-asserted-by":"crossref","first-page":"874","DOI":"10.1108\/02644400010355871","article-title":"On volumetric locking of low-order solid and solid-shell elements for finite elastoviscoplastic deformations and selective reduced integration","volume":"17","year":"2000","journal-title":"Engineering Computations"},{"key":"2023033015300552000_2023.03.27.533816v1.34","doi-asserted-by":"publisher","DOI":"10.1007\/BF01589116"},{"key":"2023033015300552000_2023.03.27.533816v1.35","unstructured":"A. Haas , \u201cPyPardiso,\u201d GitHub repository https:\/\/github.com\/haasad\/PyPardisoProject, 2023."},{"key":"2023033015300552000_2023.03.27.533816v1.36","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1016\/j.cam.2012.10.034","article-title":"A computational method to assess the in vivo stresses and unloaded configuration of patient-specific blood vessels","volume":"246","year":"2013","journal-title":"Journal of Computational and Applied Mathematics"},{"key":"2023033015300552000_2023.03.27.533816v1.37","doi-asserted-by":"crossref","first-page":"104794","DOI":"10.1016\/j.compbiomed.2021.104794","article-title":"Computation of a probabilistic and anisotropic failure metric on the aortic wall using a machine learning-based surrogate model","volume":"137","year":"2021","journal-title":"Computers in Biology and Medicine"},{"key":"2023033015300552000_2023.03.27.533816v1.38"},{"key":"2023033015300552000_2023.03.27.533816v1.39"},{"key":"2023033015300552000_2023.03.27.533816v1.40","doi-asserted-by":"publisher","DOI":"10.1080\/01616412.1996.11740404"},{"key":"2023033015300552000_2023.03.27.533816v1.41"},{"key":"2023033015300552000_2023.03.27.533816v1.42","unstructured":"L. Pineda et al., \u201cTheseus: A Library for Differentiable Nonlinear Optimization,\u201d Advances in Neural Information Processing Systems, 2022."},{"key":"2023033015300552000_2023.03.27.533816v1.43","unstructured":"I. J. Goodfellow et al., \u201cGenerative Adversarial Nets,\u201d Advances in Neural Information Processing Systems, pp. 2672\u20132680, 2014."},{"key":"2023033015300552000_2023.03.27.533816v1.44","unstructured":"J. Sohl-Dickstein , E. A. Weiss , N. Maheswaranathan , and S. Ganguli , \u201cDeep unsupervised learning using nonequilibrium thermodynamics,\u201d presented at the Proceedings of the 32nd International Conference on International Conference on Machine Learning-Volume 37, Lille, France, 2015."},{"key":"2023033015300552000_2023.03.27.533816v1.45","doi-asserted-by":"crossref","first-page":"113402","DOI":"10.1016\/j.cma.2020.113402","article-title":"A generic physics-informed neural network-based constitutive model for soft biological tissues","volume":"372","year":"2020","journal-title":"Computer Methods in Applied Mechanics and Engineering"},{"key":"2023033015300552000_2023.03.27.533816v1.46","doi-asserted-by":"publisher","DOI":"10.1007\/s00366-022-01733-3"},{"key":"2023033015300552000_2023.03.27.533816v1.47","doi-asserted-by":"crossref","first-page":"103993","DOI":"10.1016\/j.mechrescom.2022.103993","article-title":"Polyconvex neural networks for hyperelastic constitutive models: A rectification approach","volume":"125","year":"2022","journal-title":"Mechanics Research Communications"},{"key":"2023033015300552000_2023.03.27.533816v1.48","doi-asserted-by":"crossref","first-page":"115248","DOI":"10.1016\/j.cma.2022.115248","article-title":"Data-driven tissue mechanics with polyconvex neural ordinary differential equations","volume":"398","year":"2022","journal-title":"Computer Methods in Applied Mechanics and Engineering"},{"key":"2023033015300552000_2023.03.27.533816v1.49","doi-asserted-by":"crossref","first-page":"114160","DOI":"10.1016\/j.cma.2021.114160","article-title":"Predicting the mechanical properties of biopolymer gels using neural networks trained on discrete fiber network data","volume":"387","year":"2021","journal-title":"Computer Methods in Applied Mechanics and Engineering"},{"key":"2023033015300552000_2023.03.27.533816v1.50","doi-asserted-by":"publisher","DOI":"10.1007\/s00466-021-02061-x"},{"key":"2023033015300552000_2023.03.27.533816v1.51","doi-asserted-by":"crossref","first-page":"104394","DOI":"10.1016\/j.compbiomed.2021.104394","article-title":"Machine learning approaches to surrogate multifidelity Growth and Remodeling models for efficient abdominal aortic aneurysmal applications","volume":"133","year":"2021","journal-title":"Computers in Biology and Medicine"},{"key":"2023033015300552000_2023.03.27.533816v1.52","doi-asserted-by":"crossref","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","article-title":"Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations","volume":"378","year":"2019","journal-title":"Journal of Computational Physics"},{"issue":"7","key":"2023033015300552000_2023.03.27.533816v1.53","doi-asserted-by":"crossref","first-page":"1585","DOI":"10.1002\/nme.7176","article-title":"Enhanced physics-informed neural networks for hyperelasticity","volume":"124","year":"2023","journal-title":"International Journal for Numerical Methods in Engineering"},{"key":"2023033015300552000_2023.03.27.533816v1.54","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1016\/j.neucom.2022.09.004","article-title":"Adversarial attacks and defenses in deep learning for image recognition: A survey","volume":"514","year":"2022","journal-title":"Neurocomputing"},{"key":"2023033015300552000_2023.03.27.533816v1.55","unstructured":"A.-K. Dombrowski , M. Alber , C. J. Anders , M. Ackermann , K.-R. M\u00fcller , and P. Kessel , \u201cExplanations can be manipulated and geometry is to blame,\u201d in Proceedings of the 33rd International Conference on Neural Information Processing Systems: Curran Associates Inc., 2019, p. Article 1217."},{"key":"2023033015300552000_2023.03.27.533816v1.56","unstructured":"J. Heo , S. Joo , and T. Moon , \u201cFooling neural network interpretations via adversarial model manipulation,\u201d in Proceedings of the 33rd International Conference on Neural Information Processing Systems: Curran Associates Inc., 2019, p. Article 263."},{"key":"2023033015300552000_2023.03.27.533816v1.57","doi-asserted-by":"publisher","DOI":"10.1115\/1.4005694"},{"key":"2023033015300552000_2023.03.27.533816v1.58","unstructured":"O. C. Zienkiewicz and R. L. Taylor , The Finite Element Method: Its Basis and Fundamentals. Elsevier, 2013."},{"key":"2023033015300552000_2023.03.27.533816v1.59","doi-asserted-by":"crossref","unstructured":"A. Logg , K.-A. Mardal , and G. Wells , Automated Solution of Differential Equations by the Finite Element Method: The FEniCS Book. Springer Berlin, Heidelberg, 2012.","DOI":"10.1007\/978-3-642-23099-8"},{"key":"2023033015300552000_2023.03.27.533816v1.60","doi-asserted-by":"crossref","first-page":"106620","DOI":"10.1016\/j.compstruc.2021.106620","article-title":"An open-source FEniCS-based framework for hyperelastic parameter estimation from noisy full-field data: Application to heterogeneous soft tissues","volume":"255","year":"2021","journal-title":"Computers & Structures"},{"key":"2023033015300552000_2023.03.27.533816v1.61","doi-asserted-by":"crossref","first-page":"110651","DOI":"10.1016\/j.jcp.2021.110651","article-title":"Hybrid FEM-NN models: Combining artificial neural networks with the finite element method","volume":"446","year":"2021","journal-title":"Journal of Computational Physics"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2023.03.27.533816","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T04:28:11Z","timestamp":1768451291000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2023.03.27.533816"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,3,28]]},"references-count":61,"URL":"https:\/\/doi.org\/10.1101\/2023.03.27.533816","relation":{"is-preprint-of":[{"id-type":"doi","id":"10.1016\/j.cmpb.2023.107616","asserted-by":"subject"}]},"subject":[],"published":{"date-parts":[[2023,3,28]]},"subtype":"preprint"}}