Experimental implementation of a polyenergetic statistical reconstruction algorithm for a commercial fan-beam CT...

Area of Science:

• Medical Physics
• Image Reconstruction
• Computed Tomography (CT)

Background:

• Accurate data characterization is crucial for implementing advanced CT reconstruction algorithms.
• Polyenergetic statistical reconstruction methods offer potential improvements over traditional algorithms.
• The Alternating Minimization (AM) algorithm requires precise polyenergetic data for optimal performance.

Purpose of the Study:

• To establish a framework for characterizing data essential for the polyenergetic Alternating Minimization (AM) algorithm on a commercial fan-beam CT scanner.
• To introduce a novel method for evaluating the accuracy of the commissioned data model used in AM reconstruction.
• To assess the performance of the AM algorithm against Filtered-Backprojection (FBP) for quantitative CT applications.

Main Methods:

• X-ray spectra were estimated using a semi-empirical model fitted to transmission measurements for three tube potentials.
• Spectral variations from the bowtie filter were computationally modeled, and central-axis scatter was measured.
• AM reconstruction with a matched object-model was used to assess data model accuracy, with FBP used for comparison.

Main Results:

• The spectrum model achieved a root-mean-square-error of 1.20%-1.34% when fitting transmission curves.
• Polyenergetic AM reconstruction of test cylinders achieved accuracy within 0.5% of expected values.
• Compared to FBP, polyenergetic AM demonstrated superior uniformity and reduced object-size dependence.

Conclusions:

• The polyenergetic AM algorithm's data model was successfully commissioned to within 0.5% of ground truth using a matched object-model.
• These findings validate the framework for data characterization and accuracy assessment.
• The results support the use of polyenergetic AM reconstruction for quantitative CT applications.