Prediction of Chronic Kidney Disease Progression Using Recurrent Neural Network and Electronic Health Records

Description

This repository includes code for predicting chronic kidney disease (CKD) progression using recurrent neural network (RNN) model and preprocessed data of electronic health records. The following is description of files in the repository.

1. environment.yml: Python packages and their version numbers required to run the scripts for setting up computational environment.
2. Functions.py: All functions required to run the prediction analysis script and the sequential forward variable selection script.
3. Prediction_Analysis.py: Script to run the prediction analysis of CKD progression, including building prediction model and evaluate its prediction performance through cross-validation.
4. Sequential_Forward_Selection.py: Script to run the sequential forward variable selection analysis to identify most predictive variables.