🩻 Chest X-Ray Pneumonia Detection AI

A Deep Learning-based web application that detects Pneumonia from Chest X-Ray images using a Convolutional Neural Network (CNN) built with TensorFlow/Keras. The application provides an interactive clinical dashboard developed with Streamlit, displaying prediction probabilities, confidence scores, patient information, and downloadable AI-generated diagnostic reports.

Note: This project is intended for educational and research purposes only and is not a substitute for professional medical diagnosis.

---

🚀 Features

• 🩻 Chest X-Ray Classification

  • Binary classification of Chest X-Ray images
  • Detects:
    • ✅ Normal
    • 🦠 Pneumonia

• 🤖 Deep Learning CNN

  • Built using TensorFlow/Keras
  • Image preprocessing and normalization
  • Binary classification using Sigmoid activation

• 📊 Prediction Dashboard

  • Prediction result
  • Confidence score
  • Normal probability
  • Pneumonia probability
  • Risk assessment

• 👤 Patient Information

  • Patient Name
  • Patient ID
  • Age
  • Gender
  • Hospital
  • Doctor

• 📈 Interactive Probability Chart

  • Built using Plotly
  • Displays prediction probabilities visually

• 📄 PDF Report Generation

  • AI-generated diagnostic report
  • Includes patient information
  • Prediction summary
  • Confidence score
  • Probability values
  • AI recommendation

• 🌙 Modern Streamlit UI

  • Responsive layout
  • Dark theme
  • Interactive dashboard

---

📸 Dashboard

---

🩻 AI Prediction Dashboard

---

📊 Prediction Analytics

---

📄 AI Diagnostic Report

📸 Application Workflow

Chest X-Ray Image
        │
        ▼
 Image Preprocessing
        │
        ▼
 TensorFlow CNN Model
        │
        ▼
Prediction
        │
        ▼
Probability Analysis
        │
        ▼
Risk Assessment
        │
        ▼
PDF Report Generation

---

🏗️ Project Structure

Pneumonia-Detection-AI/
│
├── app.py
├── README.md
├── requirements.txt
├── x_ray.ipynb
├── .gitignore
│
├── assets/
│   ├── dashboard.png
│   ├── prediction.png
│   ├── analytics.png
│   └── report.png
│
├── utils/
│   ├── helpers.py
│   ├── preprocess.py
│   ├── predict.py
│   └── report.py
│
└── LICENSE

If pneumonia_model.keras is not included, download it separately and place it in the project root.

---

🛠️ Tech Stack

Deep Learning

• TensorFlow
• Keras

Computer Vision

• OpenCV
• Pillow (PIL)

Web Framework

• Streamlit

Visualization

• Plotly
• Matplotlib

Data Processing

• NumPy
• Pandas

PDF Generation

• ReportLab

---

🧠 CNN Architecture

Input Image (224 × 224 × 3)

        │
        ▼

Conv2D (32 Filters)

        │
        ▼

MaxPooling2D

        │
        ▼

Conv2D (64 Filters)

        │
        ▼

MaxPooling2D

        │
        ▼

Flatten

        │
        ▼

Dense (128 Units)

        │
        ▼

Dense (1 Unit - Sigmoid)

        │
        ▼

Prediction

---

📊 Model Output

The model predicts:

Prediction	Description
Normal	Chest X-Ray appears normal
Pneumonia	Features associated with pneumonia detected

The application also displays:

• Confidence Score
• Normal Probability
• Pneumonia Probability
• Risk Level
• AI Recommendation

---

📂 Dataset

The model was trained using a labelled Chest X-Ray image dataset containing two classes:

• Normal
• Pneumonia

All images are resized to:

224 × 224 × 3

before inference.

---

⚙️ Installation

1. Clone the Repository

git clone https://github.com/Shaipratt51/Pneumonia-Detection-AI.git

---

2. Navigate to the Project

cd Pneumonia-Detection-AI

---

3. Create a Virtual Environment (Optional)

Windows

python -m venv venv
venv\Scripts\activate

Linux / macOS

python3 -m venv venv
source venv/bin/activate

---

4. Install Dependencies

pip install -r requirements.txt

---

5. Run the Application

streamlit run app.py

The application will open at:

http://localhost:8501

---

📝 How to Use

1. Launch the Streamlit application.
2. Enter patient information.
3. Upload a Chest X-Ray image.
4. Wait for the model prediction.
5. Review:
   • Prediction
   • Confidence
   • Probability chart
   • Risk level
   • AI recommendation
6. Download the generated PDF report.

---

📄 PDF Report Includes

• Patient Details
• Hospital Information
• Doctor Name
• Prediction
• Confidence Score
• Normal Probability
• Pneumonia Probability
• Risk Level
• AI Recommendation

---

📦 Requirements

tensorflow
streamlit
opencv-python
numpy
pandas
matplotlib
plotly
pillow
reportlab

Install using:

pip install -r requirements.txt

---

📈 Future Improvements

• Multi-class lung disease detection
• Tuberculosis detection
• COVID-19 detection
• DICOM image support
• Cloud deployment
• User authentication
• Patient history management
• Model monitoring dashboard

---

📊 Performance

The model uses:

• Convolutional Neural Network (CNN)
• Binary Cross-Entropy Loss
• Adam Optimizer
• Sigmoid Activation

Evaluation metrics include:

• Accuracy
• Prediction Probability
• Confidence Score

---

📥 Model File

The trained model (pneumonia_model.keras) may not be included in this repository because it exceeds GitHub's file size limit.

If unavailable, place the trained model file in the project root:

Pneumonia-Detection-AI/
│
├── pneumonia_model.keras

---

⚠️ Medical Disclaimer

This project is intended solely for educational and research purposes.

The predictions generated by this AI system should NOT be considered a medical diagnosis or used as a substitute for professional medical advice, diagnosis, or treatment.

Always consult a qualified physician or radiologist for clinical decisions.

---