Evaluation of tuned-aperture computed tomography depth discrimination for image series acquired variously with linear horizontal, linear vertical, and conical beam projection arrays

Objective. To compare depth discrimination by using Tuned-Aperture Computed Tomography (TACT) variously with linear horizontal, linear vertical, combined linear horizontal and linear vertical, and conical beam projection arrays. Study design. The first test object was a metallic mesh angled at 30 degrees to the surface of a computed dental radiography size No. 1 x-ray sensor. The second test object was a dry human mandible. The sensor was mounted on an optical bench constructed to permit free and precise geometrical settings for the horizontal and vertical angulations of the x-ray beam. The extent of blurring of horizontal and vertical wires in each TACT reconstructed image slice was observed for each of the tested beam projection arrays. Results. With a linear horizontal beam projection array, it was not possible to determine the depth of structures parallel to the horizontal dimension, such as the mandibular canal. With a linear vertical beam projection, it was not possible to determine the depth of structures parallel to the vertical dimension. A conical array of beam projections was best suited to the task of depth discrimination of objects in ail planes. The best second alternative was a combination of linear vertical and linear horizontal projections. Conclusions. Beam projection geometry is important for the accurate depth discrimination of TACT reconstructed images. A conical beam projection array is ideal.