/Teeth-Segmentation-Using-U-Nets

Primary LanguageJupyter NotebookMIT LicenseMIT

Teeth Segmentation using U-Net Variants on TUFTS Dataset

This repository contains code for teeth segmentation using variants of U-Net on the TUFTS dataset using PyTorch. Please note that this repository is still under construction.

Dataset

The TUFTS dataset contains X-ray images of teeth. Each image has corresponding ground truth masks for teeth segmentation.

U-Net Variants

I'am are experimenting with different variants of U-Net for teeth segmentation. These include:

  • Vanilla U-Net
  • Attention U-Net

Usage

To train the models, please follow these steps:

  1. Clone the repository
  2. Install the required packages listed in requirements.txt
  3. Download the TUFTS dataset and place it in the data directory
  4. Run the training script train.py

Results

Here are some example results from my experiments:

Vanilla U-Net

True Label 3

Predicted pred_3

Attention U-Net

True Label 13

Predicted pred_13

License

This project is licensed under the MIT License - see the LICENSE file for details.