Related Experiment Video
Updated: Jan 31, 2026

Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
Automatic heart activity diagnosis based on Gram polynomials and probabilistic neural networks
Francesco Beritelli1, Giacomo Capizzi1, Grazia Lo Sciuto1
11Department of Electrical, Electronic and Informatics Engineering (DIEEI), University of Catania, Catania, Italy.
This study introduces a novel method for diagnosing heart conditions using Gram polynomials and probabilistic neural networks (PNN) to analyze heart sound recordings. The approach achieves high accuracy in classifying normal and abnormal heart sounds from phonocardiogram (PCG) data.
Area of Science:
- Biomedical Engineering
- Cardiology
- Artificial Intelligence in Medicine
Background:
- Accurate heart disease diagnosis is crucial for timely intervention.
- Phonocardiogram (PCG) signals contain valuable information for assessing heart activity.
- Existing diagnostic methods may have limitations in sensitivity or specificity.
Purpose of the Study:
- To develop and evaluate a novel approach for heart disease diagnosis using Gram polynomials and probabilistic neural networks (PNN).
- To enhance the feature extraction process for phonocardiogram (PCG) signals.
- To assess the performance of the proposed system in classifying normal and abnormal heart sounds.
Main Methods:
- Feature extraction from PCG signals using Gram polynomials and Fourier transform.
- Classification of heart sound recordings using Probabilistic Neural Networks (PNN).
- Validation using a public database of over 3000 heart beat sound recordings.
Main Results:
- The proposed system achieved high diagnostic performance.
- Overall sensitivity was 93%, specificity was 91%, and accuracy was 94%.
- The method demonstrated effectiveness in classifying normal versus abnormal heart sounds.
Conclusions:
- Gram polynomials combined with PNN offer an efficient technique for heart disease characterization using PCG signals.
- The novel feature extraction method significantly contributes to diagnostic accuracy.
- The proposed system shows promise for clinical application in heart disease diagnosis.
Related Concept Videos
Real Zeros of Polynomials
Long Division of Polynomials
Introduction to Polynomial Functions
Synthetic Disvision of Polynomials
Automatic Processing and Automatic Social Behavior
Protein Networks
These interactions can be represented through maps depicting protein-protein interaction networks, represented as nodes and edges. Nodes are circles that are representative of a protein,...

