A rebinning-type backprojection-filtration algorithm for region of interest reconstruction in fan-beam CT with improved noise properties

We propose an alternative backprojection-filtration(BPF)-based reconstruction algorithm for fan-beam CT, which reconstructs images by first converting the fan-beam data to fan-parallel-beam data and then using a modified parallel-beam BPF algorithm to obtain the reconstruction. This proposed algorithm retains the properties of the original fan-beam BPF algorithm in that it can reconstruct exact region of interest (ROI) images from truncated data and/or super-short-scan data. The major advantage of this algorithm is its improved noise properties because of the elimination of the spatially-variant weighting factor. In addition, the proposed algorithm is computationally more efficient.