A quantitative CBCT pipeline based on 2D antiscatter grid and grid-based scatter sampling for image-guided radiation...

Area of Science:

• Medical Imaging
• Radiation Oncology
• Image Processing

Background:

• Quantitative accuracy in cone beam CT (CBCT) is crucial for radiation therapy, enabling treatment monitoring and plan adaptation.
• Current CBCT methods still have a quantitative accuracy gap compared to multi-detector CT (MDCT).

Purpose of the Study:

• To investigate a physics-driven approach, quantitative CBCT (qCBCT), combining scatter rejection, raw data correction, and iterative reconstruction.
• The goal is to enhance CBCT image quality and quantitative accuracy for improved clinical applications.

Main Methods:

• Implemented a tungsten 2D antiscatter grid and grid-based scatter sampling for residual scatter correction.
• Applied image lag and beam hardening correction for linac-mounted CBCT with offset detector geometry.
• Utilized iterative image reconstruction to reduce noise and evaluated qCBCT using phantoms, benchmarking against clinical CBCT and MDCT.

Main Results:

• qCBCT demonstrated statistically significant improvements in CT number accuracy and reduced artifacts compared to clinical CBCT.
• Mean HU errors for qCBCT and clinical CBCT were 17 ± 9 HU and 38 ± 29 HU, respectively, versus MDCT.
• Iterative reconstruction improved contrast-to-noise ratio by 25% over standard CBCT protocols.

Conclusions:

• The qCBCT approach achieves CT number accuracy comparable to MDCT through advanced scatter suppression and data correction.
• This enhanced quantitative accuracy holds potential for critical tasks in image-guided radiation therapy, including online dose calculations and treatment response assessment.