Real time implementation of anti-scatter grid artifact elimination method for high resolution x-ray imaging CMOS detectors using Graphics Processing Units (GPUs)

Abstract

Scatter is one of the most important factors effecting image quality in radiography. One of the best scatter reduction methods in dynamic imaging is an anti-scatter grid. However, when used with high resolution imaging detectors these grids may leave grid-line artifacts with increasing severity as detector resolution improves. The presence of such artifacts can mask important details in the image and degrade image quality. We have previously demonstrated that, in order to remove these artifacts, one must first subtract the residual scatter that penetrates through the grid followed by dividing out a reference grid image; however, this correction must be done fast so that corrected images can be provided in real-time to clinicians. In this study, a standard stationary Smit-Rontgen x-ray grid (line density - 70 lines/cm, grid ratio - 13:1) was used with a high-resolution CMOS detector, the Dexela 1207 (pixel size - 75 micron) to image anthropomorphic head phantoms. For a 15 × 15 cm field-of-view (FOV), scatter profiles of the anthropomorphic head phantoms were estimated then iteratively modified to minimize the structured noise due to the varying grid-line artifacts across the FOV. Images of the head phantoms taken with the grid, before and after the corrections, were compared, demonstrating almost total elimination of the artifact over the full FOV. This correction is done fast using Graphics Processing Units (GPUs), with 7-8 iterations and total time taken to obtain the corrected image of only 87 ms, hence, demonstrating the virtually real-time implementation of the grid-artifact correction technique.

Keywords: CMOS detector; GPUs; anthropomorphic head phantoms; anti-scatter grid; grid artifacts; high resolution detector; iterative method; real time correction; scatter estimation; x-ray imaging.

Fig 1

Experimental set-up

Fig 2

External view of CMOS detector

Fig 3

RS-240T

Fig 4

PBU-50

Fig 5

Flowchart showing iteration process for finding correct scatter profile using GPU

Fig 6

RS-240T anthropomorphic head phantom imaged with the FPD. The selected region shows the portion of the head phantom imaged by the CMOS detector as shown in Fig 7.

Fig 7

RS-240T phantom imaged with grid attached to CMOS detector.

Fig 8

Final scatter distribution that was subtracted from Fig 7 to remove the residual anti-scatter grid lines (brighter indicates higher scatter).

Fig 9

(a) Object with grid, (b) Object with grid: Table and Grid corrected, (c) Object with grid – Table, Grid & Scatter corrected (d) Object without grid

Fig 10

(a) Object with grid, (b) Object with grid: Table and Grid corrected, (c) Object with grid – Table, Grid & Scatter corrected (d) Object without grid

Fig 11

PBU-50 anthropomorphic head phantom imaged with the FPD. The selected region shows the portion of the head phantom imaged by the CMOS detector as shown in Fig 12.

Fig 12

PBU-50 phantom imaged with grid attached to CMOS detector.

Fig 13

(a) Object with grid, (b) Object with grid: Table and Grid corrected, (c) Object with grid – Table, Grid & Scatter corrected (d) Object without grid

Fig 14

(a) Object with grid, (b) Object with grid: Table and Grid corrected, (c) Object with grid – Table, Grid & Scatter corrected (d) Object without grid