Utilizing Bochners Theorem for Constrained Evaluation of Missing Fourier Data

A method is presented for estimating unknown Fourier domain (k-space) data using a small number of samples in that space. The method is derived from Bochners Theorem, and is termed: Bochner Inequality Completion of K-Space (BICKS). It is suitable for filling the k-space of a real and nonnegative unknown quantity, and applicable even when the sampling rate is substantially lower than the Nyquist sampling rate. The BICKS method is demonstrated in the context of medical imaging, but it is also applicable to many other scientific areas that utilize signal processing in Fourier domain. The results indicate that filling a highly undersampled k-space using BICKS enables high quality image reconstruction.