Blood Pressure Classification Using the Method of the Modular Neural Networks

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Medicine
• Cardiovascular Health

Background:

• Accurate blood pressure classification is crucial for diagnosing and managing cardiovascular conditions.
• Existing methods may lack the precision needed for nuanced patient assessment.
• Modular Neural Networks (MNNs) offer a flexible architecture for complex data analysis.

Purpose of the Study:

• To develop and evaluate a novel Modular Neural Network (MNN) model for classifying patient blood pressure levels.
• To determine the optimal MNN architecture for accurate systolic, diastolic, and pulse classification.
• To assess the MNN model's utility in aiding cardiologists with diagnosis and treatment planning.

Main Methods:

• A modular neural network (MNN) architecture was designed, comprising three distinct modules.
• Module 1 processed diastolic pressure data; Module 2 analyzed systolic pressure details; Module 3 utilized pulse data.
• Response integration was performed using an averaging method, with Levenberg-Marquardt (trainlm) and scaled conjugate gradient backpropagation (traincsg) training methods tested.

Main Results:

• The MNN model demonstrated excellent performance in classifying blood pressure levels.
• The modular approach allowed for specialized processing of different blood pressure parameters.
• The study successfully identified an effective MNN architecture for this classification task.

Conclusions:

• The proposed MNN model provides a highly accurate method for blood pressure classification.
• This technology can serve as a valuable tool for cardiologists, enhancing diagnostic accuracy and patient care.
• The model facilitates the analysis of blood pressure behavior relative to diagnoses, aiding in heart disease prevention.