Finding the peak velocity in a flow from its Doppler spectrum

Area of Science:

• Medical imaging
• Ultrasound physics
• Hemodynamics

Background:

• Doppler ultrasound generates a power spectrum from blood cell velocity distribution.
• Spectral broadening causes observed peak Doppler frequency to differ from peak velocity.
• Existing methods for estimating maximum velocity are often inaccurate due to noise and heuristic thresholds.

Purpose of the Study:

• To develop a robust method for accurately determining peak velocity from Doppler power spectrum.
• To address the limitations of current approximate methods affected by spectral broadening and noise.

Main Methods:

• Modeling the Doppler power spectrum to identify the frequency corresponding to peak velocity.
• Determining peak velocity from the descending slope of the power spectrum, specifically at the half-power point for parabolic flow.
• Validating the method's robustness against noise and various acquisition parameters.

Main Results:

• A novel method accurately models the Doppler power spectrum to determine peak flow velocity.
• For parabolic flow, peak velocity corresponds to the Doppler frequency at the half-power point on the spectrum's descending slope.
• Experimental validation showed mean bias error <1% for flow rates between 100-300 mL/min.

Conclusions:

• The developed Doppler power spectrum modeling method accurately estimates peak blood flow velocity.
• This technique offers improved accuracy and robustness compared to existing heuristic methods.
• The findings have significant implications for clinical Doppler ultrasound velocity measurements.