Thyroid Disease Detection using Machine Learning

This project compares the performance of several machine learning models for detecting thyroid diseases using structured datasets.

Contents

• ann-train.data: First dataset for training
• thyroid_cancer/Thyroid_Diff.csv: Second dataset for training.
• cnn_dataset1.ipynb, cnn_dataset2.ipynb: Convolutional Neural Networks.
• svm_dataset1.ipynb, svm_dataset2.ipynb: Support Vector Machines.
• rf_dataset1.ipynb, rf_dataset2.ipynb: Random Forest classifiers.
• gbm_dataset1.ipynb, gbm_dataset2.ipynb: Gradient Boosting Machines.

Setup

1. Download and unzip the folder.
2. Install the required Python packages:

   pip install requirements.txt
   

3. Open any .ipynb file in Jupyter Notebook to run model training and evaluation.

Models

• CNN (Keras/TensorFlow): Deep learning for pattern recognition.
• SVM (scikit-learn): Margin-based classifier.
• Random Forest (scikit-learn): Ensemble of decision trees.
• GBM (scikit-learn): Boosted decision trees for better accuracy.

Metrics Used

• Classification Report
  • Precision
  • Recall
  • F1-Score
  • Support
  • Accuracy
  • Macro/Weighted Average
• Matthews Correlation Coefficient
• Cohen’s Kappa
• Confusion Matrix
• Average Precision

License

This project is for academic and educational purposes.