Optimization of x-ray imaging geometry (with specific application to flat-panel cone-beam computed tomography)

Area of Science:

• Medical Physics
• Radiological Imaging
• Image Quality Assessment

Background:

• X-ray imaging systems involve complex geometries influencing image quality.
• Factors like x-ray scatter and detector performance (Detective Quantum Efficiency - DQE) significantly impact diagnostic accuracy.
• Optimizing imaging geometry is crucial for maximizing diagnostic performance.

Purpose of the Study:

• To develop a theoretical method for identifying optimal x-ray imaging geometry.
• To incorporate x-ray scatter and imager DQE into image quality metrics for optimization.
• To determine optimal system configurations, particularly for flat-panel imagers (FPIs) in cone-beam computed tomography (CBCT).

Main Methods:

• Extended cascaded systems analysis for FPIs to include x-ray scatter effects on DQE.
• Incorporated x-ray source distribution, imaging task, scatter, and DQE into the ICRU-defined detectability index.
• Maximized the detectability index to determine optimal system configurations and magnification.

Main Results:

• X-ray scatter degrades DQE as an additive noise source.
• Optimal magnification for FPI-CBCT configurations is approximately 1.4-1.6, dependent on scatter fraction.
• Sensitivity analysis showed results vary with focal spot size, imaging task, scatter, and detector resolution.

Conclusions:

• The presented theoretical method allows for the identification of optimal x-ray imaging geometry.
• The methodology is general and applicable to various FPI applications beyond CBCT.
• Quantified the trade-off between increased exposure and scatter degradation compensation.