The novel three-dimensional pulse images analyzed by dynamic L-cube polynomial model.

Area of Science:

• Biomedical Engineering
• Medical Imaging Analysis
• Cardiovascular Diagnostics

Background:

• Static L-cube polynomials previously analyzed static 3D pulse images.
• Hypertension diagnosis requires accurate analysis of dynamic cardiovascular signals.

Purpose of the Study:

• To propose a dynamic L-cube polynomial for analyzing dynamic 3D pulse images (d3DPIs).
• To introduce a weighted least squares (WLS) method for fitting d3DPI amplitude C(t).
• To evaluate the method's efficacy in diagnosing hypertension.

Main Methods:

• Developed a dynamic L-cube polynomial model.
• Applied a weighted least squares (WLS) method to fit C(t) at four physiological key points.
• Compared WLS fitting with two other functions for d3DPIs.
• Analyzed normal and cold-pressor-test (CPT)-induced taut 3DPIs.

Main Results:

• The WLS method achieved the least relative error in fitting C(t) at key points.
• The dynamic L-cube polynomial demonstrated a statistically significant best fit to measured 3DPIs.
• The model accurately captured dynamic time characteristics of pulse images.

Conclusions:

• The dynamic L-cube polynomial, fitted by WLS, accurately reflects cardiovascular dynamics.
• This method offers a reliable tool for hypertension diagnosis using d3DPIs.
• The approach provides significant evidence for hypertension detection.