A5057909836- Model Reduction and Neural Networks
- Elasticity and Material Modeling
- Advanced Mathematical Modeling in Engineering
- Coronary Interventions and Diagnostics
- Fluid Dynamics and Turbulent Flows
- Cardiovascular Health and Disease Prevention
- ECG Monitoring and Analysis
- Lattice Boltzmann Simulation Studies
- Cardiac Imaging and Diagnostics
- Neural Networks and Applications
- Composite Material Mechanics
- Numerical methods in engineering
- Seismic Imaging and Inversion Techniques
- Nuclear Engineering Thermal-Hydraulics
- Probabilistic and Robust Engineering Design
- Cardiovascular Function and Risk Factors
- Rheology and Fluid Dynamics Studies
- Enhanced Oil Recovery Techniques
- Numerical methods for differential equations
- Medical Imaging and Analysis
- Aortic aneurysm repair treatments
- Mathematical Biology Tumor Growth
- Control Systems and Identification
- Cardiac electrophysiology and arrhythmias
- Dam Engineering and Safety
Johns Hopkins University
2024
Brown University
2019-2023
John Brown University
2021-2023
Northwestern Polytechnical University
2016
Aortic dissection progresses mainly via delamination of the medial layer wall. Notwithstanding complexity this process, insight has been gleaned by studying in vitro and silico progression driven quasi-static pressurization intramural space fluid injection, which demonstrates that differential propensity along aorta can be affected spatial distributions structurally significant interlamellar struts connect adjacent elastic lamellae. In particular, diverse histological microstructures may...
We apply Physics-Informed Neural Networks (PINNs) for solving identification problems of nonhomogeneous materials. focus on the problem with a background in elasticity imaging, where one seeks to identify mechanical properties soft tissue based full-field displacement measurements under quasi-static loading. In our model, we two independent neural networks, approximating solution corresponding forward problem, and other unknown material parameter field. As proof concept, validate model...
The solution of a PDE over varying initial/boundary conditions on multiple domains is needed in wide variety applications, but it computationally expensive if the computed de novo whenever domain change. We introduce general operator learning framework, called DIffeomorphic Mapping Operator learNing (DIMON) to learn approximate solutions family $\{\Omega_{\theta}}_\theta$, that learns map from and $\Omega_\theta$ PDE, or specified functionals thereof. DIMON based transporting given problem...
Solving partial differential equations (PDEs) using numerical methods is a ubiquitous task in engineering and medicine. However, the computational costs can be prohibitively high when many-query evaluations of PDE solutions on multiple geometries are needed. Here we aim to address this challenge by introducing Diffeomorphic Mapping Operator Learning (DIMON), generic artificial intelligence framework that learns geometry-dependent solution operators different types variety geometries. We...
Focused ultrasound (FUS) therapy is a promising tool for optimally targeted treatment of spinal cord injuries (SCI), offering submillimeter precision to enhance blood flow at injury sites while minimizing impact on surrounding tissues. However, its efficacy highly sensitive the placement source, as cord's complex geometry and acoustic heterogeneity distort attenuate FUS signal. Current approaches rely computer simulations solve governing wave propagation equations compute patient-specific...
A computed approximation of the solution operator to a system partial differential equations (PDEs) is needed in various areas science and engineering. Neural operators have been shown be quite effective at predicting these generators after training on high-fidelity ground truth data (e.g. numerical simulations). However, order generalize well unseen spatial domains, neural must trained an extensive amount geometrically varying samples that may not feasible acquire or simulate certain...
Multiscale modeling is an effective approach for investigating multiphysics systems with largely disparate size features, where models different resolutions or heterogeneous descriptions are coupled together predicting the system's response. The solver lower fidelity (coarse) responsible simulating domains homogeneous whereas expensive high-fidelity (fine) model describes microscopic features refined discretization, often making overall cost prohibitively high, especially time-dependent...
Quantifying biomechanical properties of the human vasculature could deepen our understanding cardiovascular diseases. Standard nonlinear regression in constitutive modeling requires considerable high-quality data and an explicit form model as prior knowledge. By contrast, we propose a novel approach that combines generative deep learning with Bayesian inference to efficiently infer families relationships data-sparse regimes. Inspired by concept functional priors, develop adversarial network...