Quantification of scattered radiation in projection mammography: four practical methods compared

Area of Science:

• Medical Physics
• Radiological Imaging
• Digital Mammography

Background:

• Accurate quantification of scattered radiation is crucial for image quality in digital mammography.
• Traditional beam stop techniques can be time-consuming and less reproducible.
• Development of alternative, practical methods for scatter estimation is needed.

Purpose of the Study:

• To compare four practical methodologies for quantifying scattered radiation in digital mammography systems.
• To evaluate methods as alternatives to the beam stop technique.
• To assess reproducibility and applicability across different antiscatter grid types (linear and cellular) and geometries (grid in/out).

Main Methods:

• Quantified scattered radiation using scatter to primary ratio (SPR) and scatter fraction (SF).
• Applied four scatter estimation methods: beam stop, hybrid (MTF + Monte Carlo), low-frequency drop from MTF, and edge spread function (ESF) from PMMA measurements.
• Assessed repeatability error for all methods under varying PMMA thickness, grid position, and beam quality.

Main Results:

• Beam stop SPR varied with PMMA thickness and grid type, ranging from 0.012 to 1.124.
• The direct MTF method showed a maximum difference of 24% compared to beam stop data, except for cellular grid-in geometry (164%).
• Repeatability errors were lower for the direct MTF method (3-12%) compared to beam stop (1-87%) and ESF methods.

Conclusions:

• Three alternative methods provided reasonable SPR estimates without a grid.
• The direct MTF method is a viable, quick, and reproducible alternative to the beam stop technique for scatter estimation in digital mammography, especially when SPR is not very low.
• The direct MTF method demonstrated good agreement for linear grids but showed higher differences for cellular grids with low SPR.