Abstract:
AI-based reconstruction for fast MRI is the most up-to-date way to make magnetic resonance imaging (MRI) faster. This method uses deep learning techniques to build high-quality MRI pictures from under-sampled k-space data. This cuts down on the time it takes to get an MRI scan. The thesis discusses the technical details of the AI-based rebuilding method, such as convolutional neural networks (CNNs), generative adversarial networks (GANs), and the fast MRI dataset. The results show that the AI-based rebuilding method produces images with the same quality as standard MRI scans but in much less time. This opens the door to future faster and more efficient MRI scans.
