Reduction of CT number location dependency using photon-counting detector CT and virtual monoenergetic imaging

Area of Science:

• Medical Imaging
• Radiotherapy Physics
• Detector Technology

Background:

• CT number accuracy is crucial for precise radiotherapy planning.
• Location-dependent CT number variations in energy-integrated-detector (EID) CT can impact treatment accuracy.
• Photon-counting-detector (PCD) CT offers potential for improved CT number stability.

Purpose of the Study:

• To compare the location dependency of CT numbers between PCD CT and EID CT across various imaging modes, tissue types, and phantom sizes.
• To evaluate the potential of PCD CT virtual-monoenergetic images (VMIs) to mitigate CT number variations.

Main Methods:

• Scans were performed on phantoms simulating different patient sizes (40x30 cm² and 50 cm width) with tissue inserts (cortical bone, liver, lung, adipose).
• Utilized PCD CT, dual-source EID CT, and dual-layer EID CT scanners, acquiring data at matched CTDIvol.
• Reconstructed single-energy and 70 keV VMIs, quantifying CT number variations (std, CV, max-to-min, central-to-peripheral differences) across seven locations.

Main Results:

• PCD-VMI demonstrated significantly lower CT number locational variation compared to EID CTs across most tissue types and phantom sizes.
• For cortical bone in the standard phantom, PCD-VMI achieved a 70.3% reduction in standard deviation compared to single-energy PCD.
• PCD-VMI exhibited coefficient of variation below 1% and 3% for small and large phantoms, respectively, with notable improvements for high-Z materials and larger patients.

Conclusions:

• PCD-VMI effectively reduces CT number locational dependency compared to conventional EID CT imaging modes.
• The enhanced stability of CT numbers with PCD-VMI is particularly beneficial for high-Z materials and in scenarios simulating larger patients.
• PCD-VMIs show significant promise for improving accuracy and reliability in radiation therapy planning.