ThyroPredict 🧠💉

A web-based application for predicting thyroid cancer risk (Benign vs Malignant) using machine learning.

🔍 Overview

ThyroPredict is an intelligent medical diagnosis app developed using Python, Flask, and Scikit-learn. It uses clinical, demographic, and lifestyle data to predict the likelihood of thyroid cancer. The app provides an easy-to-use interface for users to enter their information and receive an instant prediction.

🛠 Features

• Predicts thyroid cancer diagnosis as Benign or Malignant
• Built using a trained machine learning models
• Clean and responsive frontend with HTML, CSS, and Bootstrap
• Real-time prediction on form submission
• Displays risk category and explanation

📊 Dataset

• Source: https://www.kaggle.com/datasets/mzohaibzeeshan/thyroid-cancer-risk-dataset/data
• Target Variable: 'Diagnosis' (Benign or Malignant)
• Features Include:
  • Age
  • TSH, T3, T4 levels
  • Nodule Size
  • Gender, Country, Ethnicity
  • Family History, Radiation Exposure
  • Smoking, Obesity, Diabetes

🧠 Machine Learning

• Data preprocessing and cleaning using 'Pandas'
• Feature encoding, scaling
• Trained using:
  • ✅ Logistic
  • ✅ SVM
  • ✅ Desicion Tree
  • ✅ Random Forest
  • ✅ XGBoost
  • Evaluated on metrics: Accuracy, Precision, Recall, F1 Score

🌐 Tech Stack

• Backend: Python, Flask
• Frontend: HTML, CSS, Bootstrap
• ML Tools: scikit-learn, pandas
• IDE: VS Code, Jupyter Notebook