Call for Applications: AI Optical Character Recognition (OCR)

Local Health System Sustainability (LHSS) Vietnam

CALL FOR APPLICATIONS

AI Optical Character Recognition (OCR)

I. Background:

Abt Associates is a mission-driven, global leader in research and program implementation in the fields of health, social and environmental policy, and international development. Known for its rigorous approach to solving complex challenges, Abt Associates was ranked as one of the top 20 global research firms in 2012 and also named one of the top 40 international development innovators. The company has multiple offices in the U.S. and program offices in nearly 40 countries.

Through Abt Associates, the 4-year USAID-funded Local Health System Sustainability (LHSS) provides technical assistance in health systems strengthening in Vietnam to sustainably manage holistic human immunodeficiency virus (HIV) and tuberculosis (TB) programs that will drive the achievement of the country’s commitment to end HIV and TB by 2030. The activity will also work with the Government of Vietnam to increase domestic financing for health and strengthen the sustainability of domestic financing mechanisms for Vietnam’s greater financial, administrative, and technical ownership of its HIV and TB response.

The USAID in Vietnam has played a fundamental role in the GVN’s transition to greater financial, administrative, and technical ownership of its HIV and TB response. Efforts are needed to reinforce mechanisms that ensure accountable, transparent, and efficient use of public funds Vietnam’s social health insurance (SHI) scheme is devising ways to control costs for eligible services provided at health facilities, particularly at the hospital level. The MOH and VSS plan to reform the provider payment system including the adoption of a diagnosis related group (DRG) payment method. This requires developing and introducing an appropriate management and monitoring system to guide health facilities with implementation and to allow VSS to monitor its financial claims. Currently, during the SHI claim review process, there are many scanned documents that need to be uploaded to the SHI claim review system for better reviewing and comparison.

Manual review for such documents is very time-consuming and prone to errors, especially when it comes to C79 forms applied to each visit. As the amount of data increases, the manual review will waste significant time and resources.

Extracting scanned documents' content into machine-readable data for programmatic activities (e.g., searchable, comparable, etc.) is imperative. Technically, the best solution could be to use a combination of image processing and Optical Character Recognition (OCR). OCR is a technology to read and interpret text and figures from scanned files. OCR is especially able to handle different kinds of documents such as invoices, passports, name cards, files, etc. Hence, the application of OCR to automatically recognize information in scanned documents will save time, effort, and increase accuracy in health insurance claim review.  

II. Objectives:

  1. Using existing Machine Learning frameworks, implement an AI module for image processing, to support the Optical Character Recognition (OCR) of official (signed and stamped) documents, including health services claims that are scanned and imported to the social health insurance (SHI) claim review system.
  2. Integrate the OCR module into the SHI claim review system, and develop functions that focus on the algorithm for matching the scanned document’s extracted data with its corresponding records already on the system.
  3. Provide training for 63 provincial social security agencies on the proficient use of the new OCR functions.

III.  Methodology:

The software agency will work closely with LHSS and VSS to understand the detailed requirements, build an effective AI model, and implement the module to meet the VSS requirements.

IV.  Major deliverables:

  • An AI-based OCR module with requested features integrated into the online SHI review system.
  • A technical document (blueprint) describing in detail how to build the OCR module—including how existing OCR models/frameworks were incorporated—and how to adjust parameters in case of new SHI forms.
  • Source code for the AI OCR module.

V.     Eligibility and qualification criteria:

  • A well-established Viet Nam-based firm with experience in software development, especially in AI/ML systems, image processing, and OCR.
  • Strong financial and accounting system
  • The contractor is not in the process of litigation or dispute in court.

VI.  Submission timeline and contact point:

Deadline for submitting the organization’s credentials, technical and financial proposals: May 10, 2023.

 

Job Details
Organisation Name: 
Abt Associates
Application Deadline: 
Wed, 2023-05-10