Quantitative cone-beam CT reconstruction with polyenergetic scatter model fusion

Area of Science:

• Medical Physics
• Image Reconstruction
• Computed Tomography

Background:

• Scatter in cone-beam CT (CBCT) causes significant errors, particularly with its wide field of view.
• Compensating for scatter is challenging due to its complex nature.
• Iterative polyenergetic reconstruction algorithms for CT are often incompatible with scatter-corrupted data.

Purpose of the Study:

• To develop and evaluate a novel scatter compensation method for CBCT.
• To integrate scatter modeling directly into the iterative reconstruction process.
• To improve quantitative imaging accuracy and reduce artifacts in CBCT.

Main Methods:

• Introduced a polyenergetic convolutional scatter model fused directly into the iterative reconstruction.
• Exploited iterative information for scatter estimation at a low computational cost.
• Evaluated the method using numerical and real CBCT measurements.

Main Results:

• Significantly enhanced electron density estimation compared to fast adaptive scatter kernel superposition (fASKS).
• Demonstrated superior artifact mitigation over pre-calculated scatter correction methods.
• Showed reduced bias from pre-reconstruction scatter estimation.

Conclusions:

• The integrated convolutional scatter model improves CBCT quantitative accuracy.
• The method adapts to specimen-specific spectral and spatial properties.
• This approach offers a computationally efficient and effective solution for scatter correction in CBCT.