PATIENT-SPECIFIC DOSE ESTIMATES IN DYNAMIC COMPUTED TOMOGRAPHY MYOCARDIAL PERFUSION EXAMINATION

Area of Science:

• Medical Physics
• Radiological Dosimetry
• Cardiovascular Imaging

Background:

• Computed Tomography (CT) myocardial perfusion imaging is crucial for diagnosing coronary artery disease.
• Accurate patient-specific dose estimation is essential for radiation protection in dynamic CT examinations.
• Previous dose estimations often relied on generic phantoms, potentially limiting accuracy.

Purpose of the Study:

• To implement realistic source models of a CT scanner for Monte Carlo simulations.
• To calculate patient-specific organ and effective dose estimates for dynamic CT myocardial perfusion.
• To evaluate the impact of using actual patient data versus generic phantoms on dose estimation accuracy.

Main Methods:

• Developed realistic source models (bowtie filter, tube output, X-ray spectra) for a Siemens Somatom Definition Flash scanner.
• Merged 20 CT angiography patient datasets with a scaled ICRP 110 voxel phantom.
• Conducted dose simulations using ImpactMC software for patient-specific dose calculations.

Main Results:

• Effective dose estimates ranged from 5.0 to 14.6 mSv (80 kV) and 8.9 to 24.7 mSv (100 kV).
• Significant variations in organ and effective doses were observed among patients.
• Patient-specific dose simulations demonstrated considerable differences compared to generic phantom approaches.

Conclusions:

• Patient-specific organ and effective dose estimates are feasible using realistic CT scanner models and Monte Carlo simulations.
• Utilizing actual patient data merged with matched anthropomorphic anatomy is crucial for accurate dose estimation.
• This approach enhances radiation safety and provides a more reliable basis for dose assessment in dynamic CT myocardial perfusion.