Spectral analysis of laser Doppler signals in real time using digital processing

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Devices

Background:

• Laser Doppler flowmetry (LDF) is crucial for non-invasive blood flow measurement.
• Real-time analysis of LDF signals requires efficient digital signal processing.
• Programmability in LDF systems enhances adaptability for various research needs.

Purpose of the Study:

• To develop a versatile, programmable spectrum analyzer for laser Doppler shift signals.
• To evaluate different signal processing algorithms for blood flow analysis.
• To identify the most reliable indicator for blood flow velocity.

Main Methods:

• Development of a digital signal processing (DSP) based spectrum analyzer.
• Real-time generation and display of laser Doppler shift signals and parameters.
• Investigation of algorithms including Fast Fourier Transform (FFT) and Maximum Entropy Spectral Estimation (MESE).
• Validation using a blood-flow phantom with varying hematocrit levels.

Main Results:

• The spectrum analyzer demonstrated real-time processing capabilities.
• An index based on the first moment of the Doppler power spectrum showed high reliability.
• Linearity of the flow indicator extended to 5 mm s-1 at 5% blood hematocrit.
• The system supported both linear (FFT) and non-linear (MESE) spectral analysis techniques.

Conclusions:

• The developed programmable spectrum analyzer is a versatile tool for LDF signal analysis.
• The first moment of the Doppler power spectrum is a robust indicator of blood flow velocity.
• The system's programmability allows for flexible algorithm implementation and system attribute adjustment.
• This technology has potential applications in various fields requiring precise blood flow monitoring.