---

Disease Prediction Using Machine Learning

Welcome to the Disease Prediction repository! This project involves training five different machine learning models, each designed to predict a specific disease. The models use various machine learning algorithms to achieve high accuracy in predicting conditions like brain tumors, heart health, diabetes, lung health, and chest health.

Project Overview

Objectives

• Brain Tumor Prediction: Using Convolutional Neural Networks (CNNs) to predict the presence of brain tumors from MRI scans.
• Heart Health Prediction: Applying Logistic Regression to assess heart health based on clinical data.
• Diabetic Prediction: Utilizing Random Forest for predicting diabetes based on patient information.
• Lung Health Prediction: Implementing K-Nearest Neighbors (KNN) to detect lung cancer.
• Chest Health Prediction: Employing CNNs to predict chest health issues using X-ray images.

Algorithms and Accuracies

• K-Nearest Neighbors (KNN):

  • Disease: Lung Cancer
  • Accuracy: 100%

• Convolutional Neural Networks (CNN):

  • Disease: Brain Tumor
  • Accuracy: 93%
  • Disease: Chest Health (X-Ray)
  • Accuracy: 95%

• Random Forest:

  • Disease: Diabetic Prediction
  • Accuracy: 96%

• Logistic Regression:

  • Disease: Heart Health Prediction
  • Accuracy: 92%

Installation

Requirements

Ensure you have Python 3.x installed, along with the following packages:

pip install numpy pandas scikit-learn tensorflow keras opencv-python matplotlib seaborn

Repository Setup

Clone the repository and navigate to the project directory:

git clone https://github.com/d-rushma/VITA_SCAN.git
cd VITA_SCAN

Usage

1. Data Preparation

Before training the models, ensure that you have your datasets organized. The datasets should be preprocessed and split into training and testing sets.

2. Model Training

You can train each model separately by running the corresponding scripts:

• KNN for Lung Cancer Prediction:

  python train_knn_lung_cancer.py

• CNN for Brain Tumor Prediction:

  python train_cnn_brain_tumor.py

• CNN for Chest X-Ray Prediction:

  python train_cnn_chest_xray.py

• Random Forest for Diabetic Prediction:

  python train_rf_diabetic.py

• Logistic Regression for Heart Health Prediction:

  python train_lr_heart_health.py

3. Model Evaluation

After training, evaluate the models using the provided evaluation scripts:

• KNN for Lung Cancer Prediction:

  python evaluate_knn_lung_cancer.py

• CNN for Brain Tumor Prediction:

  python evaluate_cnn_brain_tumor.py

• CNN for Chest X-Ray Prediction:

  python evaluate_cnn_chest_xray.py

• Random Forest for Diabetic Prediction:

  python evaluate_rf_diabetic.py

• Logistic Regression for Heart Health Prediction:

  python evaluate_lr_heart_health.py

4. Predictions

Once the models are trained and evaluated, you can use them to make predictions on new data:

from models import load_trained_model

model = load_trained_model('path_to_model.h5')
predictions = model.predict(new_data)

Results

Accuracies:

• KNN for Lung Cancer: 100%
• CNN for Brain Tumor: 93%
• CNN for Chest X-Ray: 95%
• Random Forest for Diabetic Prediction: 96%
• Logistic Regression for Heart Health: 92%

Include visualizations such as confusion matrices, ROC curves, and example predictions in this section.

Contributions

Feel free to fork this repository, submit issues, or make pull requests. Contributions are welcome!

Contact

For any questions or feedback, please reach out at [email protected].

Link for the website: https://mellifluous-hamster-da427b.netlify.app/ Overviwe of the project: https://drive.google.com/file/d/1dzahJkshf2mK2N7tPwVRxJbL2EbdCNki/view?usp=sharing