A comparison of scatter correction software and physical grid in supine chest radiography using a mobile X-ray system...

Scatter radiation in mobile chest radiography reduces image contrast and diagnostic confidence. Although physical grids improve image quality, they increase radiation dose and require precise alignment. Software-based scatter correction offers a potential alternative, but its performance across varying body habitus remains uncertain.

An experimental study was conducted using an anthropomorphic chest phantom at three thicknesses (23, 28, and 33 cm) to simulate patients of varying sizes, including medium, large, and extra-large. Radiographs were obtained with a physical grid and scatter-correction software using matched exposure indices. Radiation dose was evaluated via dose-area product (DAP), entrance surface air kerma (ESAK), and exposure index (EI). Quantitative image analysis included noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). Qualitative image quality was assessed by three radiologic technologists using a 5-point Likert scale for visual grading.

Compared with the physical grid, the scatter-correction software reduced DAP and ESAK by 51-61 % while maintaining comparable EI values. Noise and SNR were similar or improved for medium and large phantoms, but CNR decreased significantly with increasing thickness. Visual grading scores confirmed acceptable diagnostic quality at 23 cm and 28 cm, but not at 33 cm, where physical grids remained superior.

Scatter-correction software substantially reduces the radiation dose and provides diagnostically acceptable image quality for patients with medium to large body habitus. However, in larger patients, conventional grids continue to offer superior contrast and diagnostic reliability.

The implementation of scatter-correction software offers significant potential for radiation dose optimisation in mobile chest radiography, specifically among patients presenting with medium to large body habitus.