Experimental task-based optimization of a four-camera variable-pinhole small-animal SPECT system.

Area of Science:

• Medical Imaging
• Nuclear Medicine
• Biomedical Engineering

Background:

• Hardware optimization in small-animal SPECT imaging is crucial for improving diagnostic accuracy.
• Previous work utilized lumpy object models and simulated imaging for optimization figures of merit.
• Experimental validation methods and phantoms were needed to confirm simulation-based optimizations.

Purpose of the Study:

• To develop and validate experimental methods and phantoms for small-animal SPECT system hardware optimization.
• To compare the performance of different system configurations using a realistic imaging task.
• To establish a basis for future hardware optimization studies in SPECT imaging.

Main Methods:

• A four-camera small-animal SPECT system with interchangeable pinhole plates was used.
• A novel small-animal phantom was developed to generate random backgrounds for image sequences.
• The channelized Hotelling observer (CHO) with Laguerre-Gauss channels was employed to assess signal-to-noise ratio (SNR).
• The task involved detecting a 2mm diameter sphere within a random background using 138 projection images.

Main Results:

• Experimental data showed agreement with simulated data, validating the optimization approach.
• Higher detectability rates were observed for systems with multiple cameras, multiple pinholes, and high magnification.
• Mixed magnifications generally outperformed single-magnification systems for this detection task.

Conclusions:

• The developed methods and phantoms enable experimental validation of SPECT hardware optimization.
• The study confirms that specific system configurations enhance lesion detectability in small animals.
• This research provides a foundation for optimizing SPECT hardware for improved preclinical imaging performance.