Comparison of conventional, model-based quantitative planar, and quantitative SPECT image processing methods for...

Area of Science:

• Nuclear Medicine
• Medical Imaging Physics
• Radiopharmaceutical Dosimetry

Background:

• Accurate quantification of organ radionuclide uptake is crucial for patient-specific dosimetry in nuclear medicine.
• Conventional planar imaging methods suffer from limited quantitative accuracy due to projection overlap, scatter, attenuation, and background activity.
• Existing methods struggle to overcome inherent limitations in planar imaging for precise organ activity measurement.

Purpose of the Study:

• To propose and validate a novel quantitative planar (QPlanar) processing method for improved organ radionuclide uptake quantification.
• To compare the accuracy and precision of the QPlanar method against conventional planar (CPlanar) and quantitative SPECT (QSPECT) methods.
• To assess the impact of 3D organ VOIs and accurate physical modeling on planar imaging quantification.

Main Methods:

• Developed a quantitative planar (QPlanar) method using maximum likelihood (ML) estimation with 3D organ VOIs and a projector modeling image degrading effects.
• Evaluated the QPlanar method using both a physical phantom experiment (RSD Torso phantom) and Monte Carlo simulations (MCS) with the 3D NCAT phantom.
• Compared QPlanar results with CPlanar methods (with various corrections) and QSPECT methods using experimental and simulated projection and CT data.

Main Results:

• Even with ideal corrections, CPlanar methods demonstrated limited quantitative accuracy in MCS studies.
• The QPlanar method, incorporating accurate physical factor modeling and 3D organ VOIs, significantly increased quantitative accuracy.
• QPlanar achieved accuracy comparable to QSPECT but required substantially less acquisition and computation time, confirmed in both phantom and simulation studies.

Conclusions:

• The proposed QPlanar method offers a substantial improvement in the accuracy of organ activity estimates from planar images compared to CPlanar processing.
• QPlanar leverages 3D organ VOIs and accurate projection models to overcome limitations of conventional planar imaging.
• The QPlanar method provides accuracy approaching that of QSPECT, presenting a more time-efficient alternative for patient-specific dosimetry.