Normal Dual Isotope V/Q SPECT Model for Monte-Carlo Studies

Area of Science:

• Nuclear Medicine
• Medical Imaging
• Computational Modeling

Background:

• Quantifying functional lung volumes using ventilation/perfusion (V/Q) SPECT/CT lacks validated tools.
• Challenges include physiological lung function non-uniformity and the absence of ground truth for algorithm assessment.
• Monte Carlo simulations offer a potential solution for creating realistic models.

Purpose of the Study:

• To develop a realistic dual-isotope lung V/Q SPECT/CT Monte Carlo simulation model.
• To integrate the anterior-to-posterior gradient of lung perfusion into the simulation.
• To provide a tool for testing delineation algorithms of functional lung volumes.

Main Methods:

• A patient's CT scan was used to build a realistic lung model.
• The model incorporated an anterior-to-posterior gradient of radioactivity concentration across sixteen coronal planes.
• Simulated projections were reconstructed with and without attenuation correction (AC) and compared to a normal database using Z-score maps.

Main Results:

• Simulations demonstrated realistic results with low mean Z-scores for both ventilation and perfusion images, with or without attenuation correction.
• Mean Z-scores for ventilation images were -0.2 (0.7) for AC and -0.2 (0.7) for noAC.
• Mean Z-scores for perfusion images were -0.2 (0.6) for AC and -0.1 (0.6) for noAC.

Conclusions:

• A validated Monte Carlo simulation model for dual-isotope lung V/Q SPECT/CT was successfully developed.
• The model realistically integrates the anterior-to-posterior perfusion gradient.
• This model can be utilized to create a catalog of clinical scenarios for testing functional lung volume delineation methods.