Modified fast adaptive scatter kernel superposition (mfASKS) correction and its dosimetric impact on CBCT-based...

Area of Science:

• Medical Physics
• Radiotherapy Physics
• Image Processing

Background:

• Cone beam computed tomography (CBCT) provides anatomical data but suffers from scatter contamination, degrading accuracy in proton therapy dose estimation.
• Accurate dose distribution is crucial for effective proton therapy and requires high-quality imaging data.

Purpose of the Study:

• To evaluate the impact of a modified fast adaptive scatter kernel superposition (mfASKS) method on proton dose distribution accuracy using CBCT.
• To combine point-spread function (PSF)-based scatter kernel derivation with the fASKS model for improved scatter correction in CBCT.

Main Methods:

• Developed mfASKS by modifying scatter kernel properties (central intensity, constant intensity, amplitude) derived from PSFs.
• Optimized a key fASKS parameter for uniformity in mfASKS-corrected images.
• Quantitatively compared Hounsfield Unit (HU) and proton stopping power ratio (SPR) images and assessed dose calculation accuracy using proton treatment plans.

Main Results:

• mfASKS correction significantly improved the accuracy of HU and SPR intensity quantifications.
• Mean absolute water-equivalent path length difference decreased from 10.3 to 0.934 mm (Gammex phantom, simulation).
• Mean absolute relative error of proton range improved from 5.03% to 2.57% (pelvic phantom, experiment), particularly at the distal fall-off region.

Conclusions:

• The mfASKS technique enables more accurate CBCT-based proton dose calculations.
• This advancement holds significant implications for image-guided radiotherapy and dose verification in adaptive proton therapy.