I am a fifth-year PhD student in Medical Engineering and Medical Physics at the Harvard-MIT Program in Health Sciences and Technology, specializing in computer science. I am advised by Elazer Edelman and Farhad Nezami. I have a M.Sc in Bioengineering from EPFL and dual bachelor’s degree in Mechanical Engineering and Physics from the American University in Cairo.

I have previously interned at Calico, where I developed protein generative models, General Electric Healthcare, where I worked on generative models for medical imaging; and Novostia, where I focused on medical device development.

I’m interested in how machine learning can unlock the full potential of computational simulation for science, engineering, and medicine. My current research focuses on alleviating key bottlenecks in using physical simulation for designing medical devices such as coronary stents and heart valves. I am specifically developing deep-learning and computational geometry tools that process high-dimensional patient-specific anatomy to automate the reconstruction of anatomical digital twins, augment these datasets with generative models, and accelerate the prediction of virtual intervention outcomes with deep-learning surrogate models.

Generative Diffusion Models of Cardiovascular Anatomy

CardioComposer: Flexible and Compositional Anatomical Structure Generation with Disentangled Geometric Guidance
Paper | Download Here
We propose a generalized compositional framework for guiding unconditional diffusion models of human anatomy using interpretable ellipsoidal primitives embedded in 3D space.
A Diffusion Model for Simulation Ready Coronary Anatomy with Morpho-skeletal Control
Paper | Download Here | Code
We introduce a family of diffusion model conditioning and guidance techniques based on geometric and skeletal constraints that can control generative models of multi-tissue coronary anatomy.
Probing the Limits and Capabilities of Diffusion Models for the Anatomic Editing of Digital Twins
Paper | Download Here
We develop a framework for studying how the natural editing capabilities of diffusion models can edit anatomic digital twins to create 'digital siblings', potentially revealing the effect of subtle anatomic variation on simulated device outcomes.

Automatic Reconstruction of Anatomic Digital Twins from Multimodal Datasets

Morphology-based non-rigid registration of coronary computed tomography and intravascular images through virtual catheter path optimization
Paper | Download Here
We introduce a geometric multimodal alignment algorithm where a set of virtual catheter frames are bent, twisted, and stretched to align the tissue morphology found in computed tomography and intravascular images.
Fully automated construction of three-dimensional finite element simulations from Optical Coherence Tomography
Paper | Download Here
We leverage morphological processing and multi-material meshing techniques to fully automate the process of creating simulation-ready digital twins for coronary artery biomechanics simulations.
A platform for high-fidelity patient-specific structural modelling of atherosclerotic arteries: from intravascular imaging to three-dimensional stress distributions
Paper | Download Here
We leverage convolutional neural networks and physics-based adaptive meshing to create simulation ready models of coronary biomechanics from intravascular images.

Numerical Simulation for Understanding Pathophysiology and Device Interventions

Impact of lesion preparation-induced calcified plaque defects in vascular intervention for atherosclerotic disease: in silico assessment
Paper | Download Here
We leverage computational mechanics simulations and parametric modelling to study the effect of calcium fractures induced by intravascular lithotripsy on stent deployment.
Biomechanics of Diastolic Dysfunction: A One-Dimensional Computational Modeling Approach
Paper | Download Here
We use a 0D-1D coupled model of the cardiovascular system to reveal how electromechanical delay and passive stiffness of the left ventricular wall affect the hemodynamic signatures of diastolic dysfunction.