Dental-Disease-Detection-on-X-Ray-Images

📌 Overview

This project presents an advanced dental disease detection on X-Ray images powered by YOLO (You Only Look Once), integrated into a Streamlit-based web interface. The model efficiently processes dental X-ray images to identify and classify common dental conditions, including:

• 🦷 Dental caries (cavities)
• 🦷 Restorative fillings
• 🦷 Impacted teeth
• 🦷 Dental implants

🏗 Model Details

🔬 Architecture and Training Parameters

• Model: YOLO11n (nano version)
• Input Resolution: 640×640 pixels
• Target Classes: 4 (Cavities, Fillings, Impacted Tooth, Implant)
• Hyperparameters:
  • Epochs: 5
  • Batch Size: 16
  • Optimizer: AdamW (lr=0.00125, momentum=0.9)
  • Dataset Composition: 753 training images, 215 validation images

📊 Performance Metrics (mAP50)

Condition	Detection Accuracy (mAP50)
Overall	0.603
Implants	0.916
Fillings	0.827
Impacted Teeth	0.644
Cavities	0.0246

🌟 Web Application Features

🔍 Image Processing and Analysis

• Supports PNG, JPG, and JPEG image formats.
• Real-time inference on dental X-ray images.
• Side-by-side display of original and processed images.

📌 Condition-Specific Analysis

• Dedicated visualization tabs for each detected pathology.
• Confidence scores for model predictions.
• Multi-perspective visualization:
  • 🖼 Annotated full image with detection boxes
  • 🔍 Cropped region of interest (ROI)
  • 🎯 Focused view of marked ROI

📈 Statistical and Graphical Representations

• 📊 Distribution of detection confidence scores.
• 🥧 Class distribution pie charts.
• 📉 Interactive visualizations powered by Plotly.

⚡ Installation & Setup

🖥 System Requirements

• Python 3.9+
• Streamlit
• Ultralytics YOLO
• OpenCV
• Pillow
• Plotly
• NumPy
• Pandas
• Streamlit-lottie

🔧 Installation Steps

1. Clone the repository:

   git clone [repository-url]

2. Install dependencies:

   pip install -r requirements.txt

3. Launch the application:

   streamlit run app.py

🚀 Usage Guide

1. Run the application:

   streamlit run app.py

2. Upload a dental X-ray image via the web interface.
3. Review the analysis, including:
   • 🦷 Bounding-box annotations for detected conditions.
   • 📊 Model confidence scores per detection.
   • 🔎 Detailed breakdown of each pathology.
   • 📈 Statistical summaries and visual analytics.

🎯 Model Training

The model was trained using the Ultralytics YOLO framework with the command:

yolo detect train data=datasets/Dental-X-ray/data.yaml model=yolo11n.pt epochs=5 imgsz=640

⚠️ Limitations

• 🚫 Cavity detection performance requires improvement (mAP50: 0.0246).
• 📷 Currently supports only X-ray image analysis.
• 🕒 Inference time depends on image resolution and system capabilities.

🚀 Future Enhancements

✅ Enhancing cavity detection through additional data augmentation
✅ Expanding compatibility to 3D dental imaging
✅ Integrating AI-driven treatment recommendations
✅ Developing real-time video processing capabilities
✅ Adding multilingual support for broader accessibility

🤝 Contributing

We welcome contributions! Please fork the repository and submit a Pull Request with any improvements or feature additions.

---

📧 Contact

For inquiries or collaborations, contact us via email: samridh@iiotengineers.com