Evaluation and Clinical Application of a Commercially Available Iterative Reconstruction Algorithm for CBCT-Based...

Area of Science:

• Medical Imaging
• Radiology
• Image Reconstruction

Background:

• Cone-beam computed tomography (CBCT) is crucial for image-guided interventions.
• Standard filtered back-projection (FBP) reconstruction in CBCT can be limited by noise and artifacts.
• Iterative reconstruction algorithms offer potential for improved image quality.

Purpose of the Study:

• To quantitatively evaluate the image quality of a new iterative CBCT reconstruction algorithm.
• To compare the performance of the iterative algorithm against standard FBP reconstruction.
• To assess the impact of the iterative algorithm on noise, resolution, and artifacts.

Main Methods:

• A new iterative CBCT reconstruction pipeline incorporating scatter correction and statistical reconstruction was developed.
• Image quality was assessed using phantoms (QA and anthropomorphic head) and clinical patient data (head/neck and prostate).
• Metrics included uniformity, HU constancy, spatial resolution, low contrast detection, noise level, and contrast-to-noise ratio.

Main Results:

• The iterative CBCT algorithm significantly reduced noise across all tested protocols (Head and Pelvis scans).
• Noise reduction ranged from 28.8 HU to as low as 5.9 HU in phantom studies.
• Artifacts were reduced, and noise in patient data decreased by 62% ± 15%.

Conclusions:

• The iterative CBCT reconstruction algorithm demonstrates superior performance in reducing noise and artifacts compared to FBP.
• Significant improvements in noise and contrast-to-noise ratio were observed.
• These enhancements are expected to improve the accuracy of CBCT-guided applications.