Blood flow quantification from 2D phase contrast MRI in renal arteries using an unsupervised data driven approach

Area of Science:

• Medical Imaging
• Cardiovascular Science
• Image Analysis

Background:

• Accurate measurement of renal artery blood flow is crucial for diagnosing renal artery stenosis.
• Current manual segmentation methods are time-consuming and may introduce variability.
• 2D Phase Contrast Cine Magnetic Resonance Imaging (PC Cine MR) offers non-invasive hemodynamic assessment.

Purpose of the Study:

• To develop and validate an automated clustering approach for renal artery segmentation in 2D PC Cine MR images.
• To assess the accuracy of the automated method in measuring arterial blood velocity and flow compared to manual segmentation.
• To evaluate the clinical utility of the method for grading renal artery stenosis and guiding interventions.

Main Methods:

• A novel clustering algorithm was applied to segment the renal artery from 2D PC Cine MR datasets.
• Arterial blood velocity and flow parameters were calculated from the segmented vessel areas.
• Quantitative comparison was performed against manual delineations of vessel areas.

Main Results:

• The automated clustering approach successfully segmented renal arteries across 20 datasets (3 volunteers, 7 patients).
• Velocity profiles derived from automated segmentation showed high correlation (r = 0.977) with manual delineations.
• The method demonstrated reliable extraction of hemodynamic information.

Conclusions:

• Automated renal artery segmentation using the proposed clustering method is accurate and efficient.
• This technique provides a valuable tool for objective assessment of renal artery hemodynamics.
• The findings support the use of this method in clinical practice for stenosis grading and intervention planning.