A 4D digital phantom for patient-specific simulation of brain CT perfusion protocols

Area of Science:

• Medical Imaging Physics
• Computational Modeling
• Cerebrovascular Imaging

Background:

• Optimizing CT brain perfusion (CTP) protocols is complex due to interactions between acquisition parameters, data calculation, and patient hemodynamics.
• Existing digital phantoms often lack patient-specific hemodynamic data and realistic noise patterns.
• Developing advanced phantoms is crucial to avoid patient radiation exposure and suboptimal parameter choices.

Purpose of the Study:

• To validate a 4D digital phantom approach that incorporates realistic noise and patient-specific hemodynamics for CTP simulation.
• To assess the phantom's ability to realistically simulate mean perfusion values and noise in perfusion data for individual patients.

Main Methods:

• A 4D digital phantom was developed with spatial tissue definition, measured tissue attenuation curves, and measured noise patterns.
• Patient-specific CTP data were retrospectively simulated by assigning measured tissue attenuation curves and noise patterns.
• Twenty patients with suspected ischemic stroke were used to construct patient-specific phantoms; results were compared to actual patient data using ICCs and linear fit.

Main Results:

• Intraclass correlation coefficients (ICCs) for mean perfusion values ranged from 0.92 to 0.99.
• ICCs for the standard deviation in perfusion maps were between 0.86 and 0.93.
• Linear fitting demonstrated strong agreement with slope values ranging from 0.90 to 1.06.

Conclusions:

• A patient-specific 4D digital phantom provides a realistic simulation of mean values and standard deviation in CTP data.
• This phantom enables retrospective studies on how hemodynamics and scan parameters influence CTP values.
• The validated phantom aids in optimizing CTP protocols and understanding perfusion dynamics.