Quantitative analysis of effects of the grid specifications on the quality of digital radiography images.

Abstract

A grid is one of the key components of a digital radiography (DR) system because it removes scattered radiation, which arises when X-rays penetrate an object and improves diagnostic accuracy by enhancing image quality. With the widespread use of DR systems, demand for grids with high precision has simultaneously increased. Because unsuitable grids may decrease image quality and lead to misdiagnosis, using optimised grids for DR systems is critical. In this study, we aimed to analyse the quality of X-ray images acquired using grids with different specifications and proposed standardised criteria for grid use on the basis of our results. We measured modulation transfer function (MTF), normalised noise power spectrum (NNPS) and detective quantum efficiency (DQE) using grids with different ratios (10:1, 12:1 and 15:1) with or without implementing poly methyl methacrylate (PMMA) phantoms (0-20 cm). Pixel pitch of the detector used in this experiment was 143 μm. Based on this, a grid with a line frequency of 85 line pairs/cm was selected to prevent distortion caused by implementing unoptimised grids. As a result, the NNPS was found to increase when using the grid, and the difference in MTF and DQE was only measured when the scattered X-ray was generated by stacking the PMMA phantom. However, grids showed a positive effect MTF and DQE when the PMMA phantom was implemented. Specifically, MTF and DQE improved with increase in grid ratio. Thus, it is desirable to use a high-ratio grid to improve image quality.