Multi-domain analysis of microvascular flow motion dynamics in NAFLD

Area of Science:

• Cardiovascular physiology
• Biomedical engineering
• Non-invasive diagnostics

Background:

• Cardiovascular disease (CVD) risk assessment often relies on traditional markers.
• Microvascular dysfunction is an early indicator of CVD risk.
• Assessing microvascular network perfusion complexity offers a novel approach.

Purpose of the Study:

• To evaluate the efficacy of complexity-based methods in analyzing microvascular network perfusion.
• To determine if these methods can discriminate between individuals with varying CVD risk.
• To investigate the relationship between signal complexity and CVD risk factors.

Main Methods:

• Laser Doppler recordings of skin blood flux were analyzed in participants with non-alcoholic fatty liver disease.
• Lempel-Ziv complexity was determined at single and multiple time-scales.
• Power spectral density was evaluated to quantify signal information content.

Main Results:

• Complexity analysis revealed associations with dilatory capacity, respiration, and baseline blood flux.
• The relationship between flow modulators and signal complexity varied with time-scale, improving group discrimination.
• Multiscale Lempel-Ziv complexity achieved a classification accuracy of 86.1%.

Conclusions:

• Traditional time and frequency domain measures are insufficient for group discrimination.
• Decreased blood flux signal complexity correlates with increased CVD risk.
• Complexity-based methods, especially multiscale variants, demonstrate significant discriminatory capabilities for CVD risk assessment.