Three-Dimensional (3D) Image Reconstruction in a Truncated Archimedean-Spiral Scan Geometry Based on the Compressed-Sensing (CS) Theory: Simulation Study

In this work, a feasibility study was performed on the three-dimensional (3D) reconstruction in a truncated Archimedean-spiral geometry with a narrow detector (e.g., 90 x 392 pixels) for application to high-accurate, cost-effective dental X-ray imaging. Here an X-ray tube and a narrow detector rotate together around the rotational axis several times and, concurrently, the detector moves horizontally in the detector coordinate to cover the whole imaging volume during the projection data acquisition. A compressed-sensing (CS) based algorithm, rather than a common filtered-backprojection (FBP), was employed for more accurate image reconstruction. A systematic simulation was performed to demonstrate the viability of the proposed approach to volumetric dental X-ray imaging and the image characteristics was investigated in terms of the image value profile, the universal-quality index (UQI), and the contrast-to-noise ratio (CNR). The 3D images of considerably high accuracy were successfully reconstructed by using the proposed method and thus it was expected to be applicable to developing a cost-effective, two-in-one dental imaging system that combines the panoramic and computed tomography features.