Projects

Longitudinal Single-Center Study with Rapid Quantitative Assessment of Knee Joint with Compressed Sensing

This study uses compressive sensing (CS) techniques to accelerate knee MRI and track osteoarthritis (OA) progression by measuring T2 and T1rho relaxation times, enabling faster detection of cartilage degeneration and changes associated with advancing disease.

MSK   QUANTITATIVE MRI

Noninvasive Estimation of Electrical Properties from Magnetic Resonance Measurements

This project develops a three-pronged strategy—combining global inverse modeling, physics-informed neural networks, and vision transformers—to generate high-resolution maps of tissue electrical properties using MRI, with broad applications in medical imaging and treatment.

QUANTITATIVE MRI   SIMULATIONS

Prediction Models of Knee Osteoarthritis Incidence and Progression using Deep Learning

This project develops and validates deep-learning models that analyze clinical and imaging data to predict individuals' five-year risk of knee osteoarthritis progression and total knee replacement, aiming to enable early intervention and personalized treatment.

AI   MSK

Brain Microstructure Imaging

This project develops and validates biophysical models of diffusion and nuclear magnetic resonance relaxation to quantify tissue microstructure in the brain—such as axonal integrity, inflammation, fiber orientation dispersion, axonal diameters, axonal beading, undulations, and water exchange—using diffusion methods and a leading-edge Siemens Connectom.X gradient system.

BRAIN   DIFFUSION MRI   MICROSTRUCTURE

Data-Driven Learning Framework for Fast Quantitative Knee Joint Mapping

This project leverages advanced machine learning and deep learning to optimize T2 and T1rho mapping for knee osteoarthritis (OA), aiming to accelerate MRI techniques and enable earlier detection of cartilage degeneration.

AI   MSK   QUANTITATIVE MRI
1 2 3 7