Thyroid tumor classification

Introduction

This project is about classifying thyroid tumors based on ultrasound photos. It uses not only CNN with custom architecture, but also some pre-trained models such as VGG16 or MobileNetV2. Project contains models evaluations on test and validation datasets using confusion matrix and epoch plots for CNN.

Instalation

1. Clone the Repository

   • Clone the project repository from GitHub:

     git clone https://git.comtegra.pl/michalstrus/thyroid_tumor_classification.git

2. Create Virtual Environment

   • Go to thyroid_tumor_classification and create virtual environment named 'venv' there

     cd thyroid_tumor_classification
     python3 -m venv venv

3. Activate Virtual Environment
   • Activation on Windows

    venv\Scripts\activate

   • On macOS/Linux

    source venv/bin/activate

   You should now see that your virtual environment is active
4. Install required packages

    pip install -r requirements.txt

Downloading the dataset

1. Zipped dataset is already inside the project and will be unzipped in the notebook code

Running the notebook

1. Run the jupyter notebook

   jupyter notebook

2. Acces the file
   • Open 'actual_Thyroid.ipynb'
3. Execute the cells in notebook and uncomment unzipping part to load the dataset

Additional notes

Ensure you have followed the steps in the given order to avoid any issues.