Segmentation-free x-ray energy spectrum estimation for computed tomography using dual-energy material decomposition

Area of Science:

• Medical Imaging
• Physics
• Computational Science

Background:

• X-ray energy spectrum is crucial for computed tomography (CT) imaging.
• Existing spectrum estimation methods are often indirect and limited.
• High photon flux in clinical CT scanners complicates direct measurement.

Purpose of the Study:

• To develop a segmentation-free, indirect transmission measurement-based method for CT energy spectrum estimation.
• To utilize dual-energy material decomposition for spectrum calibration.
• To provide an accurate CT spectrum estimation without dedicated phantoms or prolonged workflows.

Main Methods:

• Employs dual-energy material decomposition to obtain material-specific images.
• Minimizes quadratic error between polychromatic forward projection and raw projection.
• Calibrates unknown weights to represent the spectrum using model spectra.

Main Results:

• The developed method accurately estimates the CT spectrum, closely matching reference spectra.
• Evaluated using numerical simulations, phantom data, and patient data, demonstrating robustness.
• Achieved accurate spectrum estimation without specialized physical phantoms or extended procedures.

Conclusions:

• The proposed method offers a robust and accurate approach to CT energy spectrum estimation.
• Applicable to various CT applications including dose calculation, artifact reduction, and image reconstruction.
• Facilitates improved CT imaging by providing precise spectral information.