Challenge call for corporate and R&D response to COVID19 health crisis in Serbia



Analysis of Covid-19 tests based on AI

Short description:

Application of artificial intelligence technology for interpreting the Covid-19 test results.


Deep Netts doo

Requested Budget:

30.000 USD

Time frame:

3 months
Problem: Since the beginning of the Covid-19 pandemic the WHO has been urging the states to test as many people as possible. The testing has been shown as crucial to understanding the spread of the pandemic and responding appropriately. In Serbia, at one point, the lack of qualified health workers slowed down the process of analysing test results. The epidemiologists leading the crisis response in the country reported that delays in submitting test results were caused by an insufficient number of health professionals qualified to interpret the results and that it was not possible for the results to be processed by administrative workers without appropriate training.
Solution: To develop a machine learning model and appropriate software based on artificial intelligence technology for interpreting test results for Covid-19. This type of diagnostics, data analysis and automatization of procedures based on the specialist’s expertise is a typical case for implementing adequate machine learning technologies. The software would require that the health institutions make their data available. The business model would be based on the sale of software licenses and the provision of support for updating the machine learning model, i.e. the software knowledge base.
Additional info: This idea presents an innovative ICT solution from a lead expert in the ICT field in Serbia. The team already has a data scientist and software developer, a medical device and measurement expert, and a project coordinator and administrator.