A practical and robust method for beam blocker-based cone beam CT scatter correction.

Area of Science:

• Medical Imaging
• Radiological Physics
• Computational Imaging

Background:

• Cone beam CT (CBCT) imaging is susceptible to scatter radiation, which degrades image quality.
• Traditional scatter correction methods using beam-blockers require tedious optimization of correction factors.
• A need exists for a more efficient and objective scatter correction method in CBCT.

Purpose of the Study:

• To develop and validate an indirect scatter estimation method for beam-blocker based CBCT scatter correction.
• To eliminate the need for optimization of correction factors in CBCT scatter correction.
• To improve the accuracy and robustness of CBCT imaging.

Main Methods:

• An indirect scatter estimation method using a beam-blocker with lead strips was developed.
• Primary signal was calculated by subtracting measured scatter, then used to estimate scatter in unblocked regions.
• Scatter signals were smoothed and applied for projection correction; CBCT was reconstructed using the Feldkamp-Davis-Kress algorithm.

Main Results:

• CT number errors were significantly reduced in phantom studies with the proposed method.
• Without a bow-tie filter, errors were reduced to 3.97±2.27 HU (Catphan, full-blocker) and 4.01±2.18 HU (head phantom, full-blocker).
• With a bow-tie filter, errors were further reduced to 2.29±1.42 HU (Catphan, full-blocker) and 2.35±1.25 HU (head phantom, full-blocker).

Conclusions:

• The proposed indirect scatter estimation method effectively corrects scatter in CBCT without tedious optimization.
• This method avoids self-influence from the beam blocker on scatter estimation.
• It offers a practical and robust solution for beam-blocker based scatter correction in CBCT.