Filtered backprojection reconstruction and redundancy in Compton camera imaging.

Area of Science:

• Nuclear physics
• Medical imaging
• Signal processing

Background:

• Compton cameras measure intensity integrals on conical surfaces, known as Compton projections.
• Image reconstruction from Compton projections relies on principles similar to the Fourier-Slice theorem.

Purpose of the Study:

• To propose and evaluate a filtered backprojection algorithm for image reconstruction using planar Compton camera data.
• To analyze the relationship between different projections and their combination in tomographical reconstruction.

Main Methods:

• Development of a filtered backprojection algorithm tailored for Compton camera data.
• Simulation of a Compton imaging system to test the algorithm.
• Analysis of projection truncation effects and data selection strategies.

Main Results:

• The proposed algorithm achieves accurate image reconstruction for simple sources in simulations.
• Observed source elongation orthogonal to the camera is attributed to projection truncation from finite device extent.
• Data selection can mitigate truncation but may increase image noise.

Conclusions:

• The filtered backprojection algorithm is a viable method for Compton camera image reconstruction.
• Projection truncation is a significant challenge affecting image fidelity in Compton imaging systems.
• The redundancy in Compton transform data is beneficial for noise reduction, despite practical limitations.