k-space Based Signal Processing Approach for Automated Detection of Streaky Coils in Radial k-space Sampling MRI

Motivation: Radial k-space sampling is preferred over cartesian k-space sampling due to its many advantages. One major drawback of radial k-space is the streaking artifacts that arise from the non-linear gradients from the peripheral of FOV. Goal(s): The goal of this study is to reduce streaky artifacts. Approach: We developed an algorithm that aims to detect the streaky coils based on post-processing of the coils’ images and then remove these streaky coils for a cleaner image. Results: Our results show that the proposed method precisely predicts streaky coils and improves the appearance of streaky artifacts in the CS reconstruction after removing the selected coils. Impact: Radial k-space sampling is improved using our new precise streaky coils detection algorithm that effectively removes them to produce a clean image with no streaky artifacts. An unmet need in the radial k-space sampling in the MRI field in general.