Local attenuation curve optimization framework for high quality perfusion maps in low-dose cerebral perfusion CT

Area of Science:

• Medical Imaging
• Radiology
• Biomedical Engineering

Background:

• Cerebral perfusion x-ray computed tomography (PCT) provides noninvasive hemodynamic brain imaging.
• Conventional PCT involves multiple scans, raising radiation dose concerns.
• Accurate estimation of the arterial input function (AIF) is crucial for high-quality perfusion maps.

Purpose of the Study:

• To develop a PCT reconstruction algorithm for dose reduction while maintaining perfusion map quality.
• To improve the accuracy of AIF estimation by optimizing arterial attenuation curves.
• To enable high-quality perfusion mapping in low-dose scanning protocols.

Main Methods:

• Introduction of the Local Attenuation Curve Optimization (LACO) framework.
• Accurate modeling of attenuation curves within vessel and arterial regions.
• Optimization of attenuation curve shape directly from acquired x-ray projection data.

Main Results:

• LACO was validated using simulations and clinical experiments.
• The algorithm accurately estimates vessel and arterial attenuation curves from limited x-ray projections.
• LACO computes an optimal AIF directly from projection data, outperforming conventional methods reliant on reconstructed images.

Conclusions:

• The LACO algorithm facilitates the estimation of high-quality cerebral perfusion maps.
• This approach enables effective low-dose scanning protocols for PCT.