svMorph: Interactive Geometry-Editing Tools for Virtual Patient-Specific Vascular Anatomies.

We propose svMorph, a framework for interactive virtual sculpting of patient-specific vascular anatomic models. Our framework includes three tools for the creation of tortuosity, aneurysms, and stenoses in tubular vascular geometries. These shape edits are performed via geometric operations on the surface mesh and vessel centerline curves of the input model. The tortuosity tool also uses the physics-based Oriented Particles method, coupled with linear blend skinning, to achieve smooth, elastic-like deformations. Our tools can be applied separately or in combination to produce simulation-suitable morphed models. They are also compatible with popular vascular modeling software, such as simvascular. To illustrate our tools, we morph several image-based, patient-specific models to create a range of shape changes and simulate the resulting hemodynamics via three-dimensional, computational fluid dynamics. We also demonstrate the ability to quickly estimate the hemodynamic effects of the shape changes via the automated generation of associated zero-dimensional lumped-parameter models.

Digital twinning of cardiac electrophysiology for congenital heart disease.

In recent years, blending mechanistic knowledge with machine learning has had a major impact in digital healthcare. In this work, we introduce a computational pipeline to build certified digital replicas of cardiac electrophysiology in paediatric patients with congenital heart disease. We construct the patient-specific geometry by means of semi-automatic segmentation and meshing tools. We generate a dataset of electrophysiology simulations covering cell-to-organ level model parameters and using rigorous mathematical models based on differential equations. We previously proposed Branched Latent Neural Maps (BLNMs) as an accurate and efficient means to recapitulate complex physical processes in a neural network. Here, we employ BLNMs to encode the parametrized temporal dynamics of in silico 12-lead electrocardiograms (ECGs). BLNMs act as a geometry-specific surrogate model of cardiac function for fast and robust parameter estimation to match clinical ECGs in paediatric patients. Identifiability and trustworthiness of calibrated model parameters are assessed by sensitivity analysis and uncertainty quantification.

Digital twinning of cardiac electrophysiology for congenital heart disease.

In recent years, blending mechanistic knowledge with machine learning has had a major impact in digital healthcare. In this work, we introduce a computational pipeline to build certified digital replicas of cardiac electrophysiology in pediatric patients with congenital heart disease. We construct the patient-specific geometry by means of semi-automatic segmentation and meshing tools. We generate a dataset of electrophysiology simulations covering cell-to-organ level model parameters and utilizing rigorous mathematical models based on differential equations. We previously proposed Branched Latent Neural Maps (BLNMs) as an accurate and efficient means to recapitulate complex physical processes in a neural network. Here, we employ BLNMs to encode the parametrized temporal dynamics of in silico 12-lead electrocardiograms (ECGs). BLNMs act as a geometry-specific surrogate model of cardiac function for fast and robust parameter estimation to match clinical ECGs in pediatric patients. Identifiability and trustworthiness of calibrated model parameters are assessed by sensitivity analysis and uncertainty quantification.

Automated generation of 0D and 1D reduced-order models of patient-specific blood flow.

Three-dimensional (3D) cardiovascular fluid dynamics simulations typically require hours to days of computing time on a high-performance computing cluster. One-dimensional (1D) and lumped-parameter zero-dimensional (0D) models show great promise for accurately predicting blood bulk flow and pressure waveforms with only a fraction of the cost. They can also accelerate uncertainty quantification, optimization, and design parameterization studies. Despite several prior studies generating 1D and 0D models and comparing them to 3D solutions, these were typically limited to either 1D or 0D and a singular category of vascular anatomies. This work proposes a fully automated and openly available framework to generate and simulate 1D and 0D models from 3D patient-specific geometries, automatically detecting vessel junctions and stenosis segments. Our only input is the 3D geometry; we do not use any prior knowledge from 3D simulations. All computational tools presented in this work are implemented in the open-source software platform SimVascular. We demonstrate the reduced-order approximation quality against rigid-wall 3D solutions in a comprehensive comparison with N = 72 publicly available models from various anatomies, vessel types, and disease conditions. Relative average approximation errors of flows and pressures typically ranged from 1% to 10% for both 1D and 0D models, measured at the outlets of terminal vessel branches. In general, 0D model errors were only slightly higher than 1D model errors despite requiring only a third of the 1D runtime. Automatically generated ROMs can significantly speed up model development and shift the computational load from high-performance machines to personal computers.

On the Periodicity of Cardiovascular Fluid Dynamics Simulations.

Three-dimensional cardiovascular fluid dynamics simulations typically require computation of several cardiac cycles before they reach a periodic solution, rendering them computationally expensive. Furthermore, there is currently no standardized method to determine whether a simulation has yet reached that periodic state. In this work, we propose the use of an asymptotic error measurement to quantify the difference between simulation results and their ideal periodic state using open-loop lumped-parameter modeling. We further show that initial conditions are crucial in reducing computational time and develop an automated framework to generate appropriate initial conditions from a one-dimensional model of blood flow. We demonstrate the performance of our initialization method using six patient-specific models from the Vascular Model Repository. In our examples, our initialization protocol achieves periodic convergence within one or two cardiac cycles, leading to a significant reduction in computational cost compared to standard methods. All computational tools used in this work are implemented in the open-source software platform SimVascular. Automatically generated initial conditions have the potential to significantly reduce computation time in cardiovascular fluid dynamics simulations.

Impact of removal of asymptomatic third molars on periodontal pathology.

PURPOSE: This study assessed the impact of third molar removal on periodontal pathology in subjects with third molars asymptomatic at enrollment. PATIENTS AND METHODS: Subjects in whom at least 2 third molars were removed were a subsample of healthy young subjects enrolled with 4 asymptomatic third molars in an institutional review board-approved longitudinal study. Full-mouth periodontal probing (PD) data, 6 sites per tooth, were obtained as a measure of periodontal status at each of 3 visits: enrollment, before removal of third molars, and after removal of third molars. Data were aggregated to subject and jaw levels. The oral cavity was divided by jaw into segments: the third molar region including the third molar (12 probing sites), distal to the second molar (4 probing sites), and non-third molars (80 probing sites). A PD >or=4 mm was considered an indicator variable for periodontal pathology. The number and percent of sites with a PD >or=4 mm were calculated from the total number of probing sites across all subjects. The frequency of subjects with at least one PD >or=4 mm and all third molars removed were compared with the frequency of subjects retaining at least 1 mandibular third molar using Fisher's exact test, with significance set at 0.05. RESULTS: Sixty-nine subjects had third molars removed: 57% were female, and 77% were Caucasian. The median age at surgery was 26.3 years (interquartile range, 23.3-31.5 yr). The median interval from enrollment to surgery was 2.4 years (interquartile range, 1.5-4.2 yr). The median follow-up after surgery was 9 months (interquartile range, 6.7-15.4 mo). All third molars were removed in 56 subjects; 13 retained at least 1 mandibular third molar. More subjects had at least 1 PD >or=4 mm around their mandibular third molars before surgery compared with enrollment (52% vs 45%, respectively). Of the total possible mandibular third molar probing sites, 18% had PD >or=4 mm presurgery compared with 12% at enrollment. Significantly fewer subjects who had all third molars removed had a PD >or=4 mm on the distal of their mandibular second molars after surgery, compared with those retaining at least 1 mandibular third molar (20% vs 69%, respectively, P= .001). The number of PDs >or=4 mm in the mandible was less after surgery if all third molars had been removed (1.4% vs 6.6%, respectively). CONCLUSION: Removal of the mandibular third molars significantly improved the periodontal status on the distal of second molars, positively affecting overall periodontal health.

Retained third molars with orthodontics and orthognathic surgery.

PURPOSE: The aim of this study was to document the prevalence of retained third molars after orthodontics and orthognathic surgery. PATIENTS AND METHODS: Inclusion criteria for these retrospective analyses included all subjects in a longitudinal trial at least 18 years old at enrollment with Class II skeletal problems, treated presurgery with orthodontics followed by orthognathic surgery. Panoramic or lateral cephalometric radiographs were analyzed to assess the presence and relationship to the occlusal plane of third molars and the presence or absence of premolars, recorded at enrollment, presurgery, and postsurgery for each subject. The primary outcome measure was the presence of third molars postsurgery. Explanatory variables included third molar position at the occlusal plane and missing premolars. Because of the few retained third molars postsurgery, analyses are limited to descriptive statistics only. RESULTS: The majority of the 372 subjects were female (80%) and Caucasian (91%). Median age at enrollment was 32.3 years (interquartile range, 27.0-39.6). At entry 145 subjects had at least 1 third molar; 57% of third molars present were at the occlusal plane, and 27% of quadrants in the 145 subjects had at least 1 missing premolar. Sixty subjects had at least 1 third molar postsurgery, 84% of third molars present were at the occlusal plane, and 44% of quadrants had at least 1 missing premolar. CONCLUSIONS: Third molars retained after treatment for dentofacial deformity with orthodontics and orthognathic surgery were more likely to be at the occlusal plane and tended to be in quadrants with missing premolars.

Mechanobiology of mandibular distraction osteogenesis: finite element analyses with a rat model.

Three-dimensional finite element (FE) analyses were performed to characterize the local mechanical environment created within the tissue regenerate during mandibular distraction osteogenesis (DO) in a rat model. Finite element models were created from three-dimensional computed tomography image data of rat hemi-mandibles at four different time points during an optimal distraction osteogenesis protocol (i.e., most successful protocol for bone formation): end latency (post-operative day (POD) 5), distraction day 2 (POD 7), distraction day 5 (POD 10), and distraction day 8 (POD 13). A 0.25 mm distraction was simulated and the resulting hydrostatic stresses and maximum principal tensile strains were determined within the tissue regenerate. When compared to previous histological findings, finite element analyses showed that tensile strains up to 13% corresponded to regions of new bone formation and regions of periosteal hydrostatic pressure with magnitudes less than 17 kPa corresponded to locations of cartilage formation. Tensile strains within the center of the gap were much higher, leading us to conclude that tissue damage would occur there if the tissue was not compliant enough to withstand such high strains, and that this damage would trigger formation of new mesenchymal tissue. These data were consistent with histological evidence showing mesenchymal tissue present in the center of the gap throughout distraction. Finite element analyses performed at different time points during distraction were instrumental in determining the changes in hydrostatic stress and tensile strain fields throughout distraction, providing a mechanical environment rationale for the different levels of bone formation in end latency, and distraction day 2, 5, and 8 specimens.

Morphometry-based impedance boundary conditions for patient-specific modeling of blood flow in pulmonary arteries.

Patient-specific computational models could aid in planning interventions to relieve pulmonary arterial stenoses common in many forms of congenital heart disease. We describe a new approach to simulate blood flow in subject-specific models of the pulmonary arteries that consists of a numerical model of the proximal pulmonary arteries created from three-dimensional medical imaging data with terminal impedance boundary conditions derived from linear wave propagation theory applied to morphometric models of distal vessels. A tuning method, employing numerical solution methods for nonlinear systems of equations, was developed to modify the distal vasculature to match measured pressure and flow distribution data. One-dimensional blood flow equations were solved with a finite element method in image-based pulmonary arterial models using prescribed inlet flow and morphometry-based impedance at the outlets. Application of these methods in a pilot study of the effect of removal of unilateral pulmonary arterial stenosis induced in a pig showed good agreement with experimental measurements for flow redistribution and main pulmonary arterial pressure. Next, these methods were applied to a patient with repaired tetralogy of Fallot and predicted insignificant hemodynamic improvement with relief of the stenosis. This method of coupling image-based and morphometry-based models could enable increased fidelity in pulmonary hemodynamic simulation.

Time-resolved three-dimensional phase-contrast MRI.

PURPOSE: To demonstrate the feasibility of a four-dimensional phase contrast (PC) technique that permits spatial and temporal coverage of an entire three-dimensional volume, to quantitatively validate its accuracy against an established time resolved two-dimensional PC technique to explore advantages of the approach with regard to the four-dimensional nature of the data. MATERIALS AND METHODS: Time-resolved, three-dimensional anatomical images were generated simultaneously with registered three-directional velocity vector fields. Improvements compared to prior methods include retrospectively gated and respiratory compensated image acquisition, interleaved flow encoding with freely selectable velocity encoding (venc) along each spatial direction, and flexible trade-off between temporal resolution and total acquisition time. RESULTS: The implementation was validated against established two-dimensional PC techniques using a well-defined phantom, and successfully applied in volunteer and patient examinations. Human studies were performed after contrast administration in order to compensate for loss of in-flow enhancement in the four-dimensional approach. CONCLUSION: Advantages of the four-dimensional approach include the complete spatial and temporal coverage of the cardiovascular region of interest and the ability to obtain high spatial resolution in all three dimensions with higher signal-to-noise ratio compared to two-dimensional methods at the same resolution. In addition, the four-dimensional nature of the data offers a variety of image processing options, such as magnitude and velocity multi-planar reformation, three-directional vector field plots, and velocity profiles mapped onto selected planes of interest.