MIDS 251 Project: EHR Big Data Pipeline

Image processing tool for Electronic Health Records (EHR) using Spark, HDFS, Tesseract, and ElasticSearch

Introduction

Setup Repo

1. Clone this repo
2. Import and activate the conda environment. All following steps must be done in this environment.

conda env create -f environment.yml
source activate ehr-big-data

Setup MongoDB

Need some content here

How to Use

EAST

python text_detection.py --image examples/images/example_09.jpg --east frozen_east_text_detection.pb

OCR

First, run the following to create a directory of PDF's.

python pdf_generator.py

Second, run the following to detect, decode, and upload predictions into a MongoDB.

python ocr_tesseract.py