A complete End-to-End AI Solution featuring a custom-trained Deep Learning model for real-time ECG abnormality detection.
Heart disease remains a leading cause of health emergencies in the workplace. This project provides an intelligent desktop dashboard that monitors ECG signals and instantly classifies them as Normal or Abnormal.
The core of this project is a custom Deep Learning model, trained on a self-curated dataset to ensure high precision in industrial and corporate environments.
| Feature | Interface / Result | Description |
|---|---|---|
| AI Dashboard |  |
The main desktop interface showing real-time ECG wave analysis and health status. |
| Model Performance |  |
Accuracy and Loss graphs showing the successful training of the custom AI model. |
| Detection Result |  |
Instant visual feedback (Normal/Abnormal) with confidence score. |
Unlike many generic solutions, this project covers the full Machine Learning lifecycle:
- Dataset Curation: I designed and pre-processed the ECG dataset, implementing signal filtering and noise reduction to ensure data quality.
- Model Architecture: A custom-built Convolutional Neural Network (CNN) optimized for temporal signal patterns.
- Training: Leveraged TensorFlow/Keras to train the model, achieving high sensitivity to cardiac arrhythmias.
- Desktop Integration: Wrapped the model into a professional GUI for non-technical workplace users.
- Core: Python 3.x
- Deep Learning: TensorFlow & Keras
- Data Processing: NumPy, Pandas, Scikit-learn
- GUI: [Tkinter]
- Visualization: Matplotlib / Seaborn
- Clone the repository:
git clone https://github.com/labsisouleimen/AI-Heart-Guardian.git
Developed by Souleimen Labsi
