Photon counting spectral CT: improved material decomposition with K-edge-filtered x-rays.

Area of Science:

• Medical Imaging
• Physics
• Materials Science

Background:

• Photon counting spectral computed tomography (PCSCT) enables material-selective imaging.
• Current PCSCT methods face limitations in material decomposition due to energy bin width and separation.
• Narrow energy bins increase noise and data loss, while wide bins lead to suboptimal decomposition.

Purpose of the Study:

• To investigate the use of selective K-edge filtration in PCSCT for improved material decomposition.
• To experimentally evaluate the impact of K-edge filtration on contrast-to-noise ratio (CNR).

Main Methods:

• PCSCT system with a cadmium zinc telluride detector and five energy bins was used.
• A 14 cm CT phantom with iodine, gold, and calcification contrast agents was scanned.
• K-edge filters (Ba and Gd) and conventional Al filters were employed at 60, 90, and 120 kVp.
• Half-value layers and mean entrance skin exposure were matched between filter types.

Main Results:

• K-edge filtration significantly improved CNR in material-decomposed images by 30%-50% compared to Al filtration.
• Mean entrance skin exposure remained consistent across all tested conditions (280 mR).

Conclusions:

• Selective K-edge filtration is a promising method to enhance material-selective PCSCT.
• This approach offers substantial improvements in CNR without increasing radiation dose.
• Further optimization could lead to even greater improvements in decomposed image quality.