State Estimation of Time-Varying MRI with Radial Golden Angle Sampling

Abstract We propose a state estimation approach to time-varying magnetic resonance imaging utilizing priori information. In estimation, the time-dependent image reconstruction problem is modeled by separate evolution and observation models. our method, we compute estimates using Kalman filter steady-state smoother data-driven estimate for process noise covariance matrix, constructed from conventional sliding window estimates. The proposed evaluated radially golden angle sampled simulated experimental small animal data rat brain. are updated after each new spoke of radial becomes available, leading faster frame rate compared with approaches. results method. show that can improve both spatial temporal resolution.