Practice Fusion had made an anonymized electronic health records (EHR) dataset on Kaggle. These ipython Notebook analyzes the data and use them to build machine and Deep learning models to make some predictions.
Here is the folder structure for the attached zip file -
-
Data Analysis of Practice Fusion Data.ipynb - This file has code to analyze and understand data.
-
Feature preparation with Medication Types.ipynb - This file has the code to build the input features for Medicine Types. Features build here are used for Machine Learning Models.
-
Implementation of Machine Learning Models.ipynb - This file has implementation of all machine learning models with Medicine Types as input. Results from multiple machine learning models are analyzed and presented here.
-
Neural_network_implementation.ipynb - This file has implementation of Deep Learning models using Tensorflow.
-
Additional Features to predict acid reflux.ipynb - This file has the code to look at additional features that can be used for Machine Learning models. This notebook analyzes and builds additional features.
-
Merge_all_additional_features.ipynb - This file combines all the features into a single file that can be fed to Machine Learning models.
-
Implementation of Machine Learning Models with additional features.ipynb - This notebook implements the machine learning models using all the data. Also, analyzes and presents the results.
-
Neural_network_implementation with additional features.ipynb - This notebook implementa the Deep Learning models with all features using tensorflow.