An experimental comparison of detector performance for computed radiography systems

Area of Science:

• Medical Imaging Physics
• Radiography Technology

Background:

• Digital computed radiography (CR) systems are widely used in medical imaging.
• Understanding the performance characteristics of different CR systems is crucial for optimizing image quality and diagnostic accuracy.

Purpose of the Study:

• To compare the intrinsic resolution, noise, and signal-to-noise transfer characteristics of eight commercial digital CR systems.
• To evaluate the impact of pixel size, beam quality, and screen type on CR system performance.

Main Methods:

• Identical experimental methods were used to evaluate eight reader/screen combinations across five CR systems.
• Modulation transfer functions (MTFs) were measured using an edge method.
• Noise power spectra (NPS) were determined by 2D Fourier analysis.
• Frequency-dependent detective quantum efficiencies (DQEs) were computed.

Main Results:

• Spatial frequencies at 0.2 MTF varied from 1.9 to 3.5 cycles/mm at 70 kVp.
• Detective quantum efficiency at 0.15 cycles/mm (DQE(0.15)) ranged from 13.8% to 30.0% at 70 kVp and 15.3% to 23.1% at 115 kVp.
• Standard-resolution screens showed similar resolution but varied DQE, while a high-resolution screen had reduced DQE and increased high-frequency MTF.

Conclusions:

• Image quality in CR systems is significantly influenced by pixel size, beam quality, screen type, and reader.
• CR system performance, particularly DQE, exhibits notable variations.
• The choice of screen type impacts both resolution and DQE, with high-resolution screens potentially sacrificing DQE.