<p>C-arm CT is movable and plays an important role in clinical surgical treatment. This paper proposes a new image reconstruction method for C-arm computed tomography (CT) using the advantages of backprojection filtration (BPF) algorithm. It first collects the projection data in a short scan, then rearranges the collected projection data into parallel projection data, and finally achieves the 3D image reconstruction in C-arm CT based on the two-step Hilbert transform method. At the same projection data acquisition frequency, compared with the commonly used FDK algorithm in C-arm CT, the advantage of the proposed method is that it only requires the projection data within a short scan, reducing radiation dose by nearly half. In addition, unlike the original BPF algorithm, the scanning angle integration interval in the backprojection of the proposed method is fixed, which improves the reconstruction efficiency and parallelism. Its reconstruction efficiency is nearly twice that of the FDK algorithm. The results of simulation and real data experiments have demonstrated the effectiveness of this proposed method.</p>