Neural network based classifier for cardio vascular diseases based on vascular aging

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Cardiovascular diseases (CVDs) are the leading global cause of death, particularly in developing nations.
• Vascular aging is a significant risk factor and precursor to vascular disease.
• Early detection of vascular aging is crucial for CVD prevention strategies.

Purpose of the Study:

• To develop a cost-effective method for classifying individuals as healthy or at risk for cardiovascular disease based on vascular aging markers.
• To assess the efficacy of a multilayer perceptron (MLP) neural network in analyzing common carotid artery ultrasound data for health classification.

Main Methods:

• Utilized ultrasound imaging data from the common carotid artery.
• Extracted parameters including arterial diameter and distension, along with subject age.
• Employed a multilayer perceptron (MLP) neural network with one hidden layer for classification tasks.

Main Results:

• The MLP network successfully classified subjects into 'healthy' or 'needing cardiologist consultation' categories.
• The classification was based on early-stage indicators derived from carotid artery ultrasound analysis.
• The method demonstrated effectiveness across various age groups.

Conclusions:

• Early detection of vascular aging through carotid artery ultrasound analysis is feasible and cost-effective.
• This approach represents a potential milestone in the early diagnosis and management of cardiovascular diseases.
• Integrating AI with medical imaging offers a promising avenue for proactive cardiovascular health assessment.