QDoc OCR: optical character recognition
Main contact
Project scope
Categories
Healthcare Information technology Software development DatabasesSkills
manual data entry python (programming language) machine learning data capture amazon rekognition amazon web services automation test planning data entry medical prescriptionProblem to Solve
Secure and accurate data entry into the QDoc is essential to the success of this health care service delivery model. Manual data entry is inefficient and prone to error. The OCR capability to scan the Manitoba Health card and automatically transfer the data directly to the patient file will address these challenges. This is the first step towards fully automated data capture throughout the entire QDoc workflow. Future iterations will include data capture from prescription records and medication labels, for example.
The students will:
1. Gather samples of all Canadian Health Cards, and checksums
2. Work with training the machine learning in AWS Rekognition.
3. Ability to program in Python creating Lambda functions that can interface with Rekongnition.
4. Develop a full end-to-end test plan for OCR of Health Cards.
Using industry standards, we plan on implementing OCR to read all Canadian Provincial health cards. Using Amazon Rekognition we will convert text into machine-readable text. It is our plan to leverage machine learning in this project.
About the organization
QDOC designs and builds cutting edge medical software. Our premier product connects patients seeking either immediate or appointment-based encounters with Canadian licensed Physicians.
Building on the rideshare model of matching clients with transportation, QDOC implements custom algorithms, along with secure PHIA/HIPAA compliant technology to create a seamless, streamlined experience for both patient and physician.