Artificial neural network based model for cardiovascular risk stratification in hypertension

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Informatics

Background:

• Cardiovascular risk stratification is essential for hypertension management.
• Current clinical methods suffer from low accuracy due to subjective physician input and patient statement uncertainty.
• Objective risk assessment is needed to guide hypertension treatment strategies.

Purpose of the Study:

• To develop an objective, artificial neural network (ANN)-based model for cardiovascular risk stratification in hypertension.
• To overcome the limitations of subjective clinical experience and patient data uncertainty.
• To improve the accuracy of cardiovascular risk evaluation in hypertensive patients.

Main Methods:

• An artificial neural network model utilizing a backpropagation algorithm was developed.
• Clinical investigation data served as the input for the model.
• The model's target output was cardiovascular risk stratification, learned from hypertension treatment guidelines.

Main Results:

• The ANN model demonstrated consistent stratification results with standard hypertension guidelines in 81.61% of cases.
• The study included 348 normotensive and hypertensive subjects.
• The model proved accurate in evaluating cardiovascular risk for hypertension.

Conclusions:

• The developed ANN model provides an objective and accurate method for cardiovascular risk stratification in hypertension.
• This objective approach can enhance the decision-making process for hypertension treatment.
• The model shows significant potential for improving risk evaluation in clinical practice.