The impact of off-centered positioning on cone beam CT image quality for image-guided brain SRS treatment

Cone beam computed tomography (CBCT) is widely used in image-guided positioning for patients with brain metastases during stereotactic radiosurgery (SRS). A major issue with CBCT is the decrease in image quality due to scatter, especially when the isocenter is off-centered for treating peripheral-metastases. This affects image clarity and CT numbers in skull-reconstruction. The study aims to measure the scatter-to-primary ratio in projection-images based on gantry-angle for off-centered isocenter. It also examines the effects of scatter on CBCT image quality and alignment-accuracy for off-centered isocenter.

A Catphan-600 phantom was imaged using the head protocol across five different FOV(2-18 cm in SI dimension) to analyze the effect of scatter on CBCT image quality in both centered and off-centered isocenter Positions. Scatter-fractions were measured for the Catphan-600 relative to Field of View (FOV) size and for a head-phantom in relation to gantry rotation at an off-centered isocenter. Computed tomography (CT) and CBCT images alignment of the head phantom was compared between centered and off-centered isocenter placements.

Increasing FOV from 2 to 18 cm significantly reduced image quality; the contrast-to-noise ratio (CNR) decreased by 1.7 times with this FOV increase. CNR decreased by 3 times at a 6 cm off-center position and the skull CT-number decreased by 350 HU. For the Catphan-600 at an 18 cm FOV, the scatter-to-primary ratio reached 0.67. The scatter-fraction for the head phantom was 1.8 times higher when fully in-field compared to partially in-field. The mean alignment accuracy for the head phantom was 0.3 ± 0.2 mm for isocenter and 0.8 ± 0.3 mm for off-center isocenter.

Quantitative analysis of CBCT image quality in phantoms reveals substantial image degradation with larger FOV and off-centered isocenter positioning. The methods in this study provide insights and a framework for refining scatter correction algorithms to achieve precise scatter adjustments. These developments can be used to enhance CBCT image quality and improve target-alignment-accuracy.