Evaluation of algorithmic requirements for clinical application of material decomposition using a multi-layer flat...

Area of Science:

• Medical Imaging
• Radiological Physics
• Image Processing

Background:

• Multi-layer flat panel detectors (FPDTs) combined with physics-based algorithms enable anatomical structure removal in X-ray imaging.
• The reliability of material decomposition can be compromised by unaddressed factors like scattered radiation.

Purpose of the Study:

• To investigate the two-material decomposition performance of a multi-layer FPDT in 2D chest radiography.
• To evaluate the impact of anti-scatter grids on decomposition accuracy.
• To compare different material decomposition algorithms.

Main Methods:

• Evaluated matrix-based material decomposition (MBMD), MBMD with polynomial beam hardening pre-correction (MBMD-PBC), and projection domain decomposition.
• Assessed decomposition accuracy using simulated data and the structural similarity index measure (SSIM).
• Validated simulation results with experiments on a triple-layer FPDT.

Main Results:

• Uncorrected scatter caused negative bone estimates and visible bone structures in soft tissue images, regardless of the decomposition algorithm.
• Even with a 13:1 anti-scatter grid, negative bone estimates persisted, and bone structures remained in soft tissue images.
• Adipose tissue had a negligible effect on decomposition accuracy.

Conclusions:

• Scattered radiation leads to negative bone contrast estimates and residual bone contrast in soft tissue images.
• Accurate scatter estimation and correction algorithms are essential for reliable material decomposition with multi-layer FPDTs.