Radiomics-based aortic flow profile characterization with 4D phase-contrast MRI.

Area of Science:

• Cardiovascular Imaging
• Medical Physics
• Radiology

Background:

• 4D Phase Contrast MRI (4D PC MRI) is standard for aorta imaging.
• Quantitative assessment of complex aortic flow patterns remains challenging.
• Existing single parameters offer limited clinical applicability for intricate flow dynamics.

Purpose of the Study:

• To introduce a radiomics-based concept for quantitative characterization of aortic flow patterns.
• To develop reproducible radiomics features for analyzing complex hemodynamics.
• To assess the potential of radiomics in differentiating sex-, age-, and disease-related flow variations.

Main Methods:

• Derivation of cross-sectional scalar parameter maps from 4D PC MRI data.
• Calculation of radiomics features including throughflow, flow direction, vorticity, and normalized helicity.
• Selection of features based on inter-scanner/inter-observer reproducibility and differentiation performance.

Main Results:

• Identified reproducible radiomics features applicable to aortic flow patterns.
• Demonstrated the suitability of selected features for characterizing different flow profile types.
• Validated feature performance in differentiating sex, age, and disease-related flow properties.

Conclusions:

• Radiomics offers a promising approach for quantitative characterization of complex aortic flow.
• Reproducible radiomics features can aid in objective flow assessment.
• Future applications include clinical studies, disease phenotyping, and enhanced diagnostic capabilities.