🖤🏣🌡Disease prediction using mahcine learning

Diseases of often become dangerous and fatal when they go unnoticed.People of then ignore the symptoms which ultimately leads to death.

The main aim of this project is to predict weather a person has a heart/kidney/parkinson's disease or diabetes or not.To do this machine learning models have been applied to various datasets of the respective disease.

In order to make predictions on diabetes various factors like glucose levels,blood pressure,skin thickness etc have been taken into account.The user must enter these values based on which they get to see wheather they actually have diabetes or not .Out of the many features only some of them have been selected based on their contribution to the prediction of the disease.This can be explained using the feature importance graph.The same applies to prediction of heart disease as well .

At this point this system can predict heart diseases with an accuracy of 82 percentage and diabetes with an accuracy of 77 percentage.

Acknowledgements

• Hyperparameter Tuning of Decision Tree Classifier Using GridSearchCV
• How to find the optimal value of K in KNN?
• readme.so
• scikit learn documentation

Datasets

All the datasets that have been used in the project are from kaggle. The heart disease dataset has been deprecated so the link redirects to a github repository where it was saved before it had deprecated.

• Heart disease dataset
• diabetes dataset
• kidney disease dataset

install

• NumPy
• Pandas
• matplotlib
• scikit-learn

Authors

• Vignesh.S
• Jay Kumar Sahoo