A Constructive Fuzzy Representation Model for Heart Data Classification.

Area of Science:

• Cardiology
• Artificial Intelligence
• Biomedical Engineering

Background:

• Telemonitoring offers a promising avenue for early detection of Heart Disease (HD) and Heart Failure (HF).
• Reducing patient mortality, morbidity, and treatment costs are key goals in cardiovascular disease management.
• Existing methods may lack robustness or intuitive feature selection for complex health data.

Purpose of the Study:

• To propose a novel classification model for Heart Disease (HD) detection and Heart Failure (HF) prediction.
• To leverage fuzzy logic for robust data classification and intuitive feature selection in telemonitoring data.
• To evaluate the model's accuracy on real-world and benchmark datasets.

Main Methods:

• Development of a fuzzy logic-based classification model.
• Representation of data using fuzzy phrases constructed from fuzzy words (fuzzy sets).
• Validation using real home telemonitoring data and a public UCI dataset.

Main Results:

• The fuzzy logic model demonstrated robust data classification capabilities.
• The approach provided an intuitive method for feature selection.
• Accuracy was investigated on both real and public datasets, showing promising results.

Conclusions:

• The proposed fuzzy logic model shows potential for accurate Heart Disease (HD) detection and Heart Failure (HF) prediction.
• This approach offers advantages in data robustness and feature interpretability for telemonitoring applications.
• Further validation could support its clinical integration for improved cardiovascular patient care.