This project is a Deep Learning-based system that analyzes chest X-ray images to detect Pneumonia. It demonstrates how AI can assist healthcare professionals in faster and more accurate diagnosis.
To build a CNN-based model that:
- Classifies X-ray images as Normal or Pneumonia
- Improves diagnostic support using AI
- Provides visual performance evaluation
- Python
- TensorFlow / Keras
- OpenCV
- NumPy
- Matplotlib
- Scikit-learn
- Chest X-ray dataset (Normal vs Pneumonia)
- Loaded using Kaggle API
- Organized into training and testing sets
- Image preprocessing (resizing, normalization)
- CNN model for classification
- Model training and validation
- Accuracy and loss visualization
- Confusion matrix evaluation
- Real-time prediction on new images
- Data Collection
- Image Preprocessing
- Model Building (CNN)
- Training
- Evaluation
- Prediction
- Open the notebook (
medical_image_analysis.ipynb) - Run all cells step-by-step
- Upload Kaggle API key
- Train the model
- Test predictions
- Install dependencies:
pip install -r requirements.txt- Run:
python main.pyAI-Medical-Image-Analysis/ │ ├── medical_image_analysis.ipynb ├── main.py ├── outputs/ ├── images/ ├── README.md ├── requirements.txt
- Deep Learning using CNN
- Medical image classification
- Model evaluation techniques
- Data preprocessing methods
- Real-world AI application
This project reflects real-world applications of AI in:
- Healthcare diagnostics
- Radiology analysis
- Medical imaging systems
Special thanks to my mentor Umesh Yadav for guidance and support.
- Deploy as a web app (Streamlit)
- Improve accuracy using transfer learning
- Add multi-disease detection
Vaishnava Devi
✨ This project showcases practical implementation of AI in healthcare and serves as a strong portfolio project.