Task-based automatic keV selection: leveraging routine virtual monoenergetic imaging for dose reduction on clinical...

Area of Science:

• Medical Imaging
• Radiology
• Photon-Counting Detector Technology

Background:

• Photon-counting detector (PCD) CT allows for routine virtual-monoenergetic image (VMI) reconstruction.
• Automatic selection tools for VMI energy levels (keV) on PCD-CT and tube potential (kV) on energy-integrating detector (EID) CT are crucial for optimizing imaging protocols.

Purpose of the Study:

• To evaluate the performance of an automatic VMI energy level selection tool on a clinical PCD-CT system.
• To compare its performance against an automatic tube potential selection tool on an EID-CT system from the same manufacturer.

Main Methods:

• Four torso phantoms of varying sizes (20-50 cm) with iodine and calcium were scanned using both PCD-CT and EID-CT.
• Task-based automatic selection tools (CARE keV for PCD-CT, CARE kV for EID-CT) were enabled.
• CT numbers, image noise, and dose-normalized contrast-to-noise ratio (CNR_d) were compared.

Main Results:

• PCD-CT automatically selected task-specific VMIs (55-70 keV) and a fixed 120 kV tube potential, achieving consistent dose reductions (9-39%) across phantom sizes.
• EID-CT utilized a range of tube potentials (80-150 kV) with variable dose reduction factors ( -65% to +21%) due to size adaptation.
• PCD-CT demonstrated equivalent or superior CNR_d compared to EID-CT, except in specific vascular imaging scenarios.

Conclusions:

• PCD-CT provides more consistent CT numbers than EID-CT due to standardized VMI output.
• This consistency aids in standardization efforts within medical imaging.
• PCD-CT facilitates effective radiation dose reduction while maintaining or improving image quality.