Kidney Disease Prediction Using Machine Learning

The healthcare industry is producing massive amounts of data which need to be mine to discover hidden information for effective prediction, exploration, diagnosis and decision making. Machine learning techniques can help and provides medication to handle this circumstances. Moreover, Chronic Kidney Disease prediction is one of the most central problems in medical decision making because it is one of the leading cause of death. So, automated tool for early prediction of this disease will be useful to cure. In this Project, the experiments were conducted for the prediction task of Chronic Kidney Disease using the machine learning algorithm

Dataset:

I downloaded the dataset from the Kaggle named Chronic Kidney Disease. This dataset has 26 attributes. Total 400 instances of the dataset is used for the training to prediction algorithms, out of which 250 has label chronic kidney disease (CKD) and 150 has label non chronic kidney disease (NCKD)

We have used Random Forest Classifier to train and build our Model

Tools / IDE

We used Jupyter NoteBook (Google Colab) for model training. And PyCharm for model deployment on the local system.

Software Requirments

• Python == 3.7.7
• TensorFlow == 2.2.0
• scikit-learn==0.22.2
• NumPy
• Flask

Model Accuracy & Classification Report

Deploy Kidney Disease Prediction Model on Local System

In this project we have deployed a model on the local machine(laptop/PC) using the python flask framework.

Open PyCharm IDE and create a new project then create folders and files according to below hierarchy of the project.

How to use App?

Just run ‘aap.py’ file in Pycharm or we can run it using anaconda prompt. Then we will get localhost address like ‘http://127.0.0.1:5000/‘ enter it in any browser in your system then enjoy it.

Just click on Predict button and you will get Predicted output as is there any chances of getting the disease or No need to fear, You have no dangerous symptoms of the disease