User-defined vascular input function curves: influence on mean perfusion parameter values and signal-to-noise ratio

Area of Science:

• Radiology
• Medical Imaging
• Neuroimaging

Background:

• Computed tomographic (CT) perfusion imaging is a vital tool for assessing brain perfusion.
• Accurate measurement of perfusion parameters like cerebral blood flow (CBF) and cerebral blood volume (CBV) is essential for diagnosis and treatment planning.
• The quality of CT perfusion maps is influenced by various factors, including the selection of input functions.

Purpose of the Study:

• To investigate the impact of arterial and venous input function properties on perfusion parameter values.
• To evaluate the effect of region of interest (ROI) placement on perfusion parameters and tissue signal-to-noise ratio (SNR).
• To determine how optimal ROI selection can enhance the quality of CT perfusion images.

Main Methods:

• Review of 40 computed tomographic (CT) perfusion studies.
• Analysis of a 10-subject subset to assess the effect of varying venous ROI locations.
• Correlation analysis between peak enhancement in ROIs and perfusion parameters (CBF, CBV) and SNR.

Main Results:

• Venous peak enhancement showed a significant correlation with mean tissue CBF and CBV.
• Both venous and arterial peak enhancement correlated significantly with the SNR of perfusion maps.
• Different ROI locations within the same vein yielded significantly different CBV and CBF values, highlighting sensitivity to placement.

Conclusions:

• Perfusion map parameters are directly influenced by peak enhancement within user-selected ROIs.
• Optimal ROI selection is critical for minimizing variability in perfusion measurements.
• Careful and consistent ROI placement is necessary to improve the quality and reliability of CT perfusion imaging.