Synergistic integration of deep neural networks and finite element method with applications of nonlinear large deformation biomechanics
Liang L, Liu M, Elefteriades J, Sun W. Synergistic integration of deep neural networks and finite element method with applications of nonlinear large deformation biomechanics. Computer Methods In Applied Mechanics And Engineering 2023, 416: 116347. PMID: 38370344, PMCID: PMC10871671, DOI: 10.1016/j.cma.2023.116347.Peer-Reviewed Original ResearchFinite element methodInverse problemPhysics-informed loss functionTime-sensitive clinical applicationsTraditional finite element methodElement methodMaterial parameter identificationForward problemSurrogate modelParameter identificationEquilibrium equationsOptimal solutionDNN solutionsTest casesOutput mappingParameter identification accuracyNeural networkDeep neural networksNon-negligible errorsFEM frameworkPatient-specific geometryFinite element analysisInverse methodInverse approachForward modelPyTorch-FEA: Autograd-enabled finite element analysis methods with applications for biomechanical analysis of human aorta
Liang L, Liu M, Elefteriades J, Sun W. PyTorch-FEA: Autograd-enabled finite element analysis methods with applications for biomechanical analysis of human aorta. Computer Methods And Programs In Biomedicine 2023, 238: 107616. PMID: 37230048, PMCID: PMC10330852, DOI: 10.1016/j.cmpb.2023.107616.Peer-Reviewed Original ResearchConceptsDeep neural networksFinite element analysisFEA codeImproved loss functionInverse methodFinite element analysis methodInverse problemCommercial FEA software packageElement analysis methodFEA software packageCommercial FEA packageNew libraryNeural networkPerformance issuesFEA packageNew inverse methodBiomechanical analysisComputational timeFEA methodLoss functionSolid mechanicsDeformation analysisSoftware packageInverse analysisSeries of applications