Fast 4D flow MRI intracranial segmentation and quantification in tortuous arteries

Area of Science:

• Medical Imaging
• Cardiovascular Imaging
• Neuroimaging

Background:

• 4D flow MRI allows comprehensive blood flow assessment in cerebral arteries.
• Clinical application of 4D flow MRI is hindered by lengthy postprocessing and user-dependent variability.
• Accurate quantification of flow dynamics in complex vascular geometries like the carotid siphon is challenging.

Purpose of the Study:

• To develop, validate, and implement a centerline processing scheme (CPS) for semiautomated segmentation and quantification of carotid siphons using 4D flow MRI.
• To assess the accuracy and efficiency of the CPS compared to manual processing and 2D phase-contrast MRI.
• To investigate flow and area parameters along the internal carotid artery (ICA) siphon in healthy subjects.

Main Methods:

• A novel centerline processing scheme (CPS) was developed for semiautomated segmentation of cerebral artery trees from 4D flow MRI data.
• Flow parameters were quantified at planes perpendicular to the vessel centerline.
• Validation was performed using an intracranial flow phantom and in vivo measurements in the internal carotid artery (ICA) of 10 healthy volunteers, comparing CPS with 2D phase-contrast MRI and manual 4D flow processing.

Main Results:

• Phantom studies showed high correlation for area (R=0.95) and flow (R=0.94) between CPS and 2D measurements.
• In vivo validation demonstrated high correlation for waveforms (R=0.93) with manual processing.
• The CPS reduced processing time by a factor of 4.6 compared to manual methods. Significantly decreased distal ICA siphon area (P=0.0017) was observed, with unchanged flow (P=0.84).

Conclusions:

• The developed CPS enables fast, semiautomated analysis of intracranial 4D flow MRI data.
• The method demonstrates internal consistency through flow conservation along the tortuous ICA siphon.
• This approach addresses key clinical translation hurdles in 4D flow MRI analysis.