Evaluation of quantitative accuracy among different scatter corrections for quantitative bone SPECT/CT imaging

Area of Science:

• Nuclear Medicine
• Medical Imaging Physics

Background:

• Scatter correction is crucial for improving SPECT image quality and quantitative accuracy.
• The impact of various scatter correction techniques on quantitative bone SPECT/CT imaging remains incompletely understood.

Purpose of the Study:

• To empirically determine the optimal scatter correction conditions for quantitative bone SPECT/CT imaging.
• To evaluate the trade-offs between quantitative accuracy and image uniformity for different scatter correction methods.

Main Methods:

• Scatter correction was applied using dual energy windows (DEW) and triple energy windows (TEW) with varying sub-window widths, alongside CT-based scatter correction (ESSE).
• Quantitative accuracy was assessed using scatter fraction (SF) and normalized mean square error (NMSE) in phantoms simulating bone tissue with different backgrounds (air, water, K2HPO4 solution).
• Image uniformity was evaluated using the coefficient of variation (CV) in a uniform cylinder phantom.

Main Results:

• Dual energy window with a 20% sub-window (DEW20%) demonstrated the lowest SF and NMSE in K2HPO4 backgrounds, indicating superior quantitative accuracy for bone SPECT.
• Triple energy window with a 3% sub-window (TEW3%) showed better results in water backgrounds.
• Larger DEW and smaller TEW sub-windows generally led to more effective scatter correction.
• The DEW20% method resulted in slightly decreased image uniformity (higher CV) compared to other methods.

Conclusions:

• The quantitative accuracy of bone SPECT/CT images is significantly influenced by the chosen scatter correction method.
• Dual energy window with a 20% sub-window (DEW20%) is recommended as the optimal method for quantitative bone SPECT, balancing accuracy and acceptable uniformity.