Artificial neural network-based method of screening heart murmurs in children
View abstract on PubMed
Summary
This summary is machine-generated.Artificial neural networks (ANNs) can accurately differentiate between innocent and pathological heart murmurs. This technology shows promise for developing a cost-effective screening device for pediatric heart disease.
Area Of Science
- Pediatric cardiology
- Biomedical engineering
- Artificial intelligence in medicine
Background
- Early detection of pediatric heart disease is crucial.
- Current screening methods for heart murmurs have limitations.
- Artificial neural networks (ANNs) excel at complex pattern recognition.
Purpose Of The Study
- To train an artificial neural network (ANN) to effectively distinguish between innocent and pathological heart murmurs.
- To assess the diagnostic potential of ANNs in pediatric cardiology.
Main Methods
- Heart sounds were recorded from 69 pediatric patients using an electronic stethoscope.
- Digital signal analysis processed the sound samples.
- A custom artificial neural network (ANN) was trained and evaluated.
Main Results
- The ANN achieved 100% sensitivity and specificity in classifying heart sounds on the collected data.
- Optimal settings were determined for the ANN classification system.
- Improved generalization for future unknown cases may require enhanced training data representation.
Conclusions
- ANNs demonstrate significant potential as accurate diagnostic tools for classifying heart sounds.
- This technology holds promise for a high-volume pediatric heart disease screening device.
- Further development could lead to improved diagnostic capabilities in pediatric cardiology.