Characterisation of fan-beam collimators.

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Physics

Background:

• Fan-beam collimators are crucial in single-photon emission tomography (SPET) for balancing image resolution and noise.
• Incorporating collimator response into iterative reconstruction algorithms can enhance SPET resolution.
• Accurate characterization of collimator focal regions and response is essential for this improvement.

Purpose of the Study:

• To evaluate a method for characterizing fan-beam collimators.
• To determine the real focal region and assess collimator convergence accuracy.
• To model collimator resolution and sensitivity using analytical functions.

Main Methods:

• Calculation of the real focal region and collimator convergence accuracy.
• Verification of Gaussian distribution fitting for collimator responses.
• Development of analytical functions to model resolution and sensitivity, with parameter fitting via non-linear regression.

Main Results:

• Identified a 1.43% difference between nominal and real focal lengths.
• Quantified focal convergence with standard deviations of 2.5 mm (x-direction) and 7.1 mm (y-direction).
• Achieved high correlation coefficients (0.994 for resolution, 0.991 for sensitivity) between experimental and predicted values.

Conclusions:

• The proposed method effectively characterizes fan-beam collimators.
• Gaussian distributions adequately represent collimator responses, excluding scattering and penetration effects.
• Analytical modeling provides accurate predictions of resolution and sensitivity, aiding SPET image reconstruction.