Fraunhofer Institute for Industrial Mathematics ITWMIn the »KIDAGO« project, our Fraunhofer ITWM research team is developing a hybrid system that uses Artificial Intelligence (AI), image processing and OCR (Optical Character Recognition) to digitize handwritten medical documents. The acronym stands for »AI-based Data Analysis for Healthcare Billing Optimization in Sub-Saharan Africa«.
Projekt KIDAGO »KI-basierte Daten-Analyse für Gesundheitsabrechnungs-Optimierung in Subsahara-Afrika«
More than 800 million people in sub-Saharan Africa have no health insurance and cannot afford health care. Despite a network of more than 25,000 public and private insurance and financing options. The project's goal is to improve health care billing in sub-Saharan Africa.
© freepik / World Health Organization
»KIDAGO« – Digitization of Health Data for Innovative Billing Systems in Sub-Saharan Africa
Fraunhofer ITWM and mTOMADY Are Focusing on Artificial Intelligence in Healthcare
In this project, our experts are analyzing data to develop a scalable, automated billing system for healthcare in the sub-Saharan region. In the project, our researchers from the departments »Image Processing« and »Financial Mathematics« are cooperating with experts from mTomady. mTOMADY gGmbH is a social enterprise that facilitates access to healthcare through financial solutions, especially in low-income regions such as Madagascar.
Insurance protects individuals from impoverishment, for example in the event of sudden illness and costly treatment. A functioning system is essential to ensure that services in a healthcare system are billed for transparently and fairly.
Our project partner mTomady gGmbH has already set up a system in Madagascar that is being used successfully. However, data is currently still being entered manually. This means that experts have to go through a laborious process to detect anomalies or irregularities. This makes it difficult and prevents the scaling of the system to other countries in the Sub-Saharan region (SSA).
Our Solution: Ai-Supported Digitization – Efficient Document Capture
In the future, semi-handwritten documents will be digitized using Artificial Intelligence. A multi-step hybrid system combines AI, classic image processing and OCR (Optical Character Recognition). Our experts then analyze the digitized data using statistical methods. In doing so, our team draws on years of experience in digitizing and detecting billing fraud in the healthcare system. The solution should not only work in this pilot project, but should ultimately become a sustainable tool for billing services that can also be transferred to other systems.
We plan to design the developed algorithms for the billing systems in such a way that they can be used in other countries in the sub-Saharan region. This also means that the system will be adapted to the needs of different markets and expanded to include a larger number of healthcare providers and patients to enable widespread use.
Strong Team for Digital Health Solutions
In this project, our researchers are cooperating with experts from mTomady. We are contributing our experience in digitization and automation. The mTomady gGmbH is already using software for automated billing with over 300,000 users. This means that a large amount of billing data is available, which is important for training data-based methods using Machine Learning (ML).
Innovative Approach: Solving the Challenge of Handwritten Data
The digitization of semi-handwritten data has been a major challenge so far because the documents vary greatly and there is a lack of uniform standards. In addition, there is often not enough training data available to improve the systems. We at Fraunhofer ITWM have developed an innovative approach that starts with the recognition of irregularities. In the project, this knowledge is combined with the local experience of mTomady to provide access to systematic health care even in economically weak and technologically underdeveloped regions. In practice, this solution, which is used in government and insurance-based projects, particularly benefits the chronically ill and pregnant women.
Market Opportunities: Licensing and Scaling of Health Software Worldwide
Our algorithms will be integrated into mTomady's existing software, initially at no cost for the pilot case in Madagascar. mTomady is continuously looking for new projects, particularly in other countries in the sub-Saharan region. In the future, our software packages will be licensed in these projects as an integral part of the overall system. The details of the license agreements will be determined after the funding phase.
Our project partner mTomady gGmbH has set up a system in Madagascar that is already being used successfully.
Our Project Partner mTOMADY
- Dr. Samuel Knauss – Co-Founder
- Elsa Rajemison – CEO Madagaskar, Co-Founder