MUSCAT: A nine-member research team from Sultan Qaboos University Hospital (SQUH), Medical Research Centre, Sultan Qaboos University (MRC-SQU) and Taqat LLC, an Omani company, are collaborating to deploy an Artificial Intelligence (AI) application to curb COVID-19 infections.
The research team includes Dr Faisal al Azri, Dr Mahmood al Jufaili, Dr Khalifa al Wahaibi, Dr Abdullah Balkhair, Dr Amal al Shibli and Dr Ibrahim al Zakwani from SQUH; Dr Khalid al Rasadi from MRC-SQU; and Dr Abdullah al Zakwani and Said al Adawi from Taqat LLC.
Currently the team has built the system and is testing it. “We hope that the results will be effective, reliable and ground-breaking. Many governments have decided to combat the virus with technology, namely contact tracing apps. However, these apps have proven either inefficient or ineffective or an intrusion to privacy,” says the research team.
It has been widely explained that the effectiveness of tracing apps is very low at best and mainly due to issues like signal inaccuracy and the virus ability to indirectly cause infection.
Bluetooth Apple and Google have joined up to reduce the impact of these issues. However, privacy issues are currently seen to be a bigger plague than COVID-19 itself. Norwegian tracing app has already been pulled off and many others are to follow suit.
Another problem is surface-to-person accounts for more infections than person to person, which neither the Bluetooth nor GPS can capture.
“We are proposing an approach of using AI-based infection detection coupled with early warnings for the possibility of getting exposed. Since we do not need user information, this is the only concept where user data is not stored at all in a centralised server. The idea is that the user marks a location anonymously,” the team notes.
The location is identified and using reinforced learning AI identifies the possible length the location can keep the risk and thus the mark. Any user who is about to enter the location is given instructions on how to protect themselves and minimise the exposure.
Users get marked as risky after this point and are asked to test themselves using the AI function. If they test positive, they are given an appointment to visit a testing centre.
It is anticipated most people will follow protocols and the advice since it is very convenient, easy to perform and especially most people will not want to expose themselves to unnecessary risk. This should reduce the number of infections.
The research team is of the view that the system is designed to initially cast a wide net and narrow it down using AI based testing system to reduce the cohort by over 80 per cent so as to increase the accuracy of prediction.
“Our concept uses existing published research for initial analysis like the cough and temperature analysis. These have been proven to be very efficient on the specific task (namely cough and temperature) but since these symptoms can be caused by other diseases, they become low in accuracy of the prediction of COVID-19,” the team members note.
They have created their own AI algorithms which they believe if combined together with these methods can increase accuracy to a satisfactory level.
With the completion of AI and software, their current objective is to test the concept with clinical data and to note its efficacy.