Low Dose Computed Tomography

Low-dose computed tomography (LDCT) aims to reduce radiation exposure during CT scans while maintaining diagnostic image quality. Current research heavily utilizes deep learning, employing architectures like U-Nets and transformers, often coupled with techniques such as denoising diffusion probabilistic models and iterative reconstruction algorithms incorporating regularization methods (e.g., total variation) to mitigate noise and artifacts inherent in low-dose scans. These advancements are crucial for improving patient safety and enabling wider access to CT imaging, particularly in applications requiring frequent or longitudinal scans. The field is also exploring self-supervised and unsupervised learning approaches to reduce reliance on large paired datasets of low- and normal-dose images.