Dynamic Imaging

Dynamic imaging focuses on reconstructing sequences of images representing time-varying processes, aiming for high spatiotemporal resolution despite often limited data. Current research emphasizes developing robust and efficient reconstruction algorithms, leveraging techniques like deep learning (including deep unrolling and generative adversarial networks), online primal-dual methods, and manifold learning to handle undersampled data and motion artifacts. These advancements are crucial for improving various applications, such as medical imaging (MRI, PET, radiography), materials science, and fluid dynamics, by enabling more accurate and timely analysis of dynamic systems.