[DMP 2024]: HCX - Digitizing Insurance Plans: Encoding health insurance products for machine readability

[SudoSu-bham]

README.md

Description

This project aims to create tools to digitise insurance plans by converting health insurance product documents into machine-readable formats. Our objective is to enhance accessibility, efficiency, and transparency within the insurance industry. By digitising insurance products we aim to facilitate faster claims adjudication, this initiative will empower both insurance providers and consumers.

Goals

• [ ] Create machine readable insurance plans as per the FHIR standards :
• [ ] Data Extraction & parsing : Extract and interpret data from health insurance product documents (primarily pdf or docs), including policy details and coverage information.
• [ ] Data Standardization and coding : Standardise and code extracted data into structured formats compatible with FHIR standards as prescribed by Open HCX platform, ensuring consistency for interoperability across insurance systems.
• [ ] Data Quality Enhancement : Identify and rectify discrepancies in digitised insurance plans, filling missing information and resolving ambiguities for completeness.
• [ ] Testing and validation harness for the coded insurance plans :
• [ ] Develop the testing and insurance plan validation harness for the digitised insurance plans created by other ecosystem participants.

Expected Outcome

Backend required to extract, process, and encode insurance product data.
The tool should generate standardised insurance plans with FHIR codes, enabling machine-readable formats for processes like auto adjudication.
Test & validate insurance plans against the finalised format.

Acceptance Criteria

Any human readable format such as PDF etc. are converted into FHIR Insurance profiles https://ig.hcxprotocol.io/v0.7.1/StructureDefinition-HCXInsurancePlan.html

Implementation Details

1. Project requires a detail understanding of JAVA or Python and APIs
2. Setup HAPI FHIR server in local machine
3. Understand FHIR profiles https://ig.hcxprotocol.io/v0.7.1/StructureDefinition-HCXInsurancePlan.html
4. Use programming concepts to convert available insurance plans in PDF format into FHIR structures

Mockups / Wireframes

• Not applicable (NA)

Product Name

Digitizing Insurance Plans

Project Name

Digitizing Insurance Plans: Encoding health insurance products for machine readability

Organization Name

Swasth Alliance

Domain

Healthcare

Tech Skills Needed

• JAVA
• HAPI FHIR
• PYTHON

Mentor(s)

• @saurabh-singh-17

Complexity

High

Category

• Performance Improvement
• Feature
• PoC

Sub Category

• API
• Accessibility
• Backend
• Research

[SudoSu-bham]

Weekly Goals

Week 1

• [ ] Understanding the HCX protocol

Week 2

• [ ] Researching and testing the FHIR API

Week 3

• [ ] Examining FHIR bundles and analyzing their various components and resources

Week 4

• [ ] Understanding about the insurancePlans profile
• [ ] Testing various PDF libraries for extracting the data from Health Insurance

Week 5

• [ ] Study insurance plan of PMJAY and one private companies
• [ ] Extracting valuable information from the PMJAY pdf

Week 6

• [ ] Making a Demo script for extracting Exclusion data from PDFs

Week 7

• [ ] Script which can extract both Benefits and Exlusion data from the health Insurance PDFs
• [ ] It should work for PMJAY PDfs

[SudoSu-bham]

Weekly Learning & Updates

Week 1

• Learning 1: Gained a comprehensive understanding of the HCX protocol, including its key components and functionalities.
• Learning 2: Acquired knowledge about the Primary flow APIs, focusing on their usage, integration, and common challenges.

Week 2

• Learning 1: Gained insights into the structure and functionality of the FHIR API through research and testing.
• Learning 2: Developed a solid understanding of FHIR implementation for HCX, including best practices and potential challenges.

Week 3

• Learning 1: Explored FHIR bundles in depth, understanding their structure and the types of resources they can contain.
• Learning 2: Analyzed various components of FHIR bundles, gaining insight into how they are used to group and manage related healthcare information.

Week 4

• Understoop about the insurancePlans profile
• Testing various PDF libraries for extracting the data from Health Insurance

Week 5

• Study insurance plan of PMJAY and one private companies.
• found out the basic pattern being used by all the Health Insurance PDFs, there was only a minor changes in PDFs otherwise all of them were created in similar pattern.

Week 6

• Made a Script which can extract Exclusion data from the PDFs.
  • Script was tested on PMJAY plan of Gujarat, J&K and Nagaland.
  • Along with PMJAY plans it can also extract Exclusion data from "Bupa Niva" health Insurance Plan.

Week 7

• Changed the data Extraction script to extract both Exclusion and Benefits data from the Health Insurance PDFs.

[samr874]

Where is the source code for this project?? @SudoSu-bham Tell me where is the source code