Experimental Characterization and Mitigation of Scatter Induced Spectral Contamination In Photon Counting CBCT for Radiotherapy
Abstract
Purpose
Photon-counting detector (PCD)–based CBCT imaging has the potential to improve tissue visualization, tissue characterization, and dose calculation accuracy in radiotherapy. However, the energy spectrum of PCDs is contaminated by X-ray scatter, degrading the utility of PCD-CBCT. This study characterized the effect of scatter on the PCD energy spectrum and investigated a hardware-driven approach to mitigate scatter contamination in PCD-CBCT.
Methods
A benchtop CBCT system incorporating a PCD with a CdTe sensor and pixel gain correction algorithms was developed. Scatter-induced bias in the energy spectrum and energy-dependent scatter-to-primary ratio (SPR) were characterized under scenarios relevant to image guided radiotherapy. The effect of scatter fluence on pulse pile-up was measured. A 2D antiscatter grid (2D ASG) was integrated with the PCD, and its effect on the energy spectrum and energy specific CT numbers was evaluated in PCD-CBCT scans.
Results
Scatter bias increased counts by up to 96% in the 25–50 keV range, shifting the mean energy by up to 8 keV. In PCD-CBCT scans, the average SPR reached 4 in the 25–60 keV range. The SPR was up to a factor of 5 higher at 35 keV than at 100 keV. Count underestimation due to pulse pile-up attributable to scatter was <3%. The 2D ASG reduced the SPR by 90–94% across the 25–120 keV range. The 2D ASG decreased CT number loss from 453 and 390 HU in the 25–60 and 60–120 keV bins to 34 and 22 HU in the respective bins.
Conclusion
Scatter contamination in PCD-CBCT for radiotherapy is substantially higher at lower energies, which may challenge spectral imaging applications. The effect of scatter fluence on pulse pile-up is considered small. The 2D ASG was highly effective in reducing scatter contamination across the entire energy spectrum, which will be crucial for future PCD-CBCT imaging applications in radiotherapy.