Biomedical simulations play an important role in aiding doctors with the diagnosis and treatment of many diseases. Such simulations are often based on patient medical imaging data. Medical imaging data is typically noisy which presents a challenge when generating geometric models and meshes of the patient’s organs. We have developed a high-order mesh generation scheme which can be employed to generate patient-specific meshes from medical images. The meshes can be used to more efficiently represent curved geometries present in the human body. In this talk, we present our method for generating high-order patient-specific biomedical meshes from medical images. A preliminary version of our methodology appeared in [1]. To this end, low-order surface meshes are first obtained from the medical images. Next, these overrefined meshes are simplified and converted to high-order surface meshes. Our method then directly generates a high-order tetrahedral volume mesh using an advancing front approach. Finally, the mesh is smoothed to take advantage of the high-order degrees of freedom and to improve its quality. We present several examples of high-order patient-specific biomedical meshes generated using our method. In addition, we present a dynamic cardiac application, where the heart is modeled as an isotropic, incompressible, and hyperelastic material using the two-parameter incompressible Mooney-Rivlin model. We apply our high-order tetrahedral mesh warping algorithm, which is based on a finite element formulation for hyperelastic materials, to yield the deforming heart meshes. For this application, the deep learning framework proposed in [2] was used to segment the cardiac chambers and estimate the cardiac motion from the cardiac cine magnetic resonance images. REFERENCES [1] Mohammadi, F. and Shontz, S.M., A direct method of generating quadratic curvilinear tetrahedral meshes using an advancing front approach, Proc. of the 29th International Meshing Roundtable, Zenodo, p. 74-91, October 2021. [2] Upendra, R.R., Wentz, B.J., Simon, R., Shontz, Suzanne M., and Linte, C.A. CNN-based cardiac motion extraction to generate deformable geometric left ventricle myocardial models from cine MRI. Proc. of the 11th International Conference on Functional Imaging and Modeling of the Heart (FIMH 2021), Vol. 12738, p. 253-263, June 2021.