Performance of four dual-energy CT platforms for abdominal imaging: a task-based image quality assessment based on...

Area of Science:

• Medical Imaging
• Radiology
• Computed Tomography

Background:

• Dual-energy CT (DECT) enables material differentiation and spectral analysis.
• Task-based image quality assessment provides a clinically relevant evaluation of CT performance.

Purpose of the Study:

• To compare the spectral performance of four DECT platforms using task-based image quality assessment.
• To evaluate lesion detectability across a range of virtual monoenergetic images (40-140 keV).

Main Methods:

• Four DECT platforms (KVSCT, SFCT, DSCT, DLCT) were tested using two phantoms.
• Noise power spectrum (NPS) and task-based transfer function (TTF) were analyzed.
• Detectability index (d') was computed for simulated contrast-enhanced lesions.

Main Results:

• Noise magnitude decreased with increasing keV for all platforms.
• Dual-layer CT (DLCT) and dual-source CT (DSCT) exhibited lower noise magnitude.
• DLCT and DSCT demonstrated the highest detectability for simulated lesions, particularly at 40 keV.

Conclusions:

• DLCT provides the highest detectability at low energy levels (40 keV).
• Task-based assessment confirms the advantage of low keV for contrast-enhanced lesion detection with DECT.
• Image noise and texture are affected by energy levels in DECT acquisitions.