An X-ray spectrum estimation method from transmission measurement combined with scatter correction

Area of Science:

• Medical Physics
• Radiological Imaging

Background:

• X-ray spectrum estimation is crucial for accurate medical imaging.
• X-ray scatter degrades the accuracy of conventional spectrum estimation methods.
• Accurate spectrum estimation is vital for dose monitoring and image quality.

Purpose of the Study:

• To apply a weighted L1-norm scatter correction algorithm for improved x-ray spectrum estimation.
• To reduce residual differences between estimated and true x-ray spectra in the presence of scatter.
• To enhance the accuracy of spectrum estimation in radiographic imaging.

Main Methods:

• Developed a scatter correction algorithm based on a radiographic scattering model.
• Estimated scattered X-ray intensity directly from transmission measurements.
• Applied scatter-corrected measurements to a spectrum estimation method using weighted linear combinations of predefined spectra.

Main Results:

• The scatter-corrected method significantly improved spectrum estimation accuracy.
• Normalized-root-mean-square-error decreased from 5.8% to 3.2%.
• Mean energy difference reduced from 1.33 keV to 0.73 keV.

Conclusions:

• The proposed method offers superior accuracy for X-ray spectrum acquisition compared to methods without scatter correction.
• Successful spectrum estimation is achievable irrespective of unknown filter materials and thicknesses.
• This technique holds potential for various diagnostic X-ray imaging applications.