YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Heart Murmur Detection Using Audio Signals & Deep Learning
Overview
This project uses audio signal processing and an LSTM deep learning model to classify heart sound recordings. It includes a Streamlit interface for uploading audio and viewing prediction results.
Model Details
- Architecture: LSTM
- Framework: TensorFlow / Keras
- Feature extraction: Mel-Frequency Cepstral Coefficients (MFCC)
- Supported input formats: WAV and MP3
- Configured sample rate: 22,050 Hz
- Configured MFCC coefficients: 52
Classes
The project is designed to classify five categories:
- Artifact
- Murmur
- Normal
- Extrahls
- Extrastole
Model File
The trained model is stored in lstm_model.keras.
Intended Use and Limitations
This project is intended for educational and research purposes. Its predictions are not a medical diagnosis and must not replace assessment by a qualified healthcare professional. Model performance depends on the training data and evaluation results.
Dataset
Heartbeat Sound Dataset:
https://www.kaggle.com/datasets/abdallahaboelkhair/heartbeat-sound
Technologies
Python, TensorFlow, Keras, librosa, NumPy, and Streamlit.
- Downloads last month
- 15