Supplementary Material for: Quantification of blood flow complexity from carotid Doppler ultrasonography: perspectives for atherosclerotic risk

Introduction: Turbulence plays a crucial role in atherosclerosis. However, it is not currently quantified in carotid ultrasound studies due to the lack of a validated method. This study aims to develop a robust method for quantifying blood flow complexity in carotid Doppler ultrasound studies using the fractal dimension of the color Doppler signal. Methods: Observational study of adult outpatients with a clinical indication for carotid Doppler ultrasound. Exclusion criteria were technical difficulties in image analysis and refusal to participate. Color Doppler signal from the internal carotid artery was extracted and analyzed. Green pixels, with high Doppler-frequency variance suggestive of turbulent-like blood flow, were identified in hue-saturation-value color space. The Hausdorff fractal dimension was calculated using the box-counting method to quantify flow complexity. On each image, the goodness of fit of the linear regressions was calculated through the coefficient of determination (R2). Results: 54 patients were enrolled between August 2020 and March 2022. Critical parameters for the method were a hue range from 7.2° to 208.8° and a minimum pixel occupancy threshold of 0.0025%. The chosen metric was the amplitude of the temporal oscillation of the Hausdorff dimension. This method successfully differentiated patients with plaques with stenosis under 50% (0.24 [0.21–0.31]) from those with normal Doppler ultrasound studies (0.30 [0.24–0.35]; p = 0.0143) and those with increased intima-media thickness (0.31 [0.24–0.35]; p = 0.0405). Conclusions: This study introduces an innovative method for assessing internal carotid artery blood flow complexity.