Farhan is a Graduate Teaching Assistant at the Weldon School of Biomedical Engineering, where his research focuses on developing advanced AI and deep learning methods for medical imaging. His work includes motion correction in high-resolution computed tomography (HR-pQCT) using generative models, simulation-to-real domain adaptation, and vision-language foundation models for automated motion scoring. He also investigates non-rigid motion correction in ultrashort echo-time (UTE) MRI using the PETAL-UTE sequence and advanced free-breathing dynamic reconstruction methods such as XD-GRASP-Pro. His research further includes implementing PETAL-UTE for musculoskeletal imaging and quantitative bone water and pore-water characterization, as well as radiomics-based feature analysis and machine learning classification for pulmonary nontuberculous mycobacterial (NTM) disease. Farhan has also worked as an AI/ML Summer Intern with the Computer Vision team at Johnson & Johnson Innovative Medicine, where he applied advanced AI techniques to accelerate clinical trial workflows and improve the efficiency of clinical data analysis.