A Predictive Risk Model for COVID-19

COVID-19, or Corona virus disease 2019 is a newly discovered variant of the coronavirus family, which severely affects the respiratory system of a human. The severity of this illness, perhaps varies from a mere respiratory difficulty to the death. Often these deaths are occurring due to the pneumonia situation prevailing within the lungs. As per the statistics published by the WHO, the percentage of no of deaths due to the illness is slightly above 7%.

At SLTC Prof. Dush N. K. Jayakody and Anupama Tabrew study on the development of a specific research, a predictive risk model, using binary logistic regression and deep learning techniques, to address the above issues. The model is also capable of generating a predictive score for all suspected cases where they can be categorized according to the severity of the risk by identify the severity of the infection of the person to ensure the testing is carried out only to the required. Also, it will be an intelligent platform by learning as the data available.

The researchers at SLTC believe that this model will be helpful for various stake holders, including government, health authorities take timely decision making which will help to prevent the spread of the COVId-19 virus as well to make maximum use of the medical resources available without a wastage. The SLTC researchers are in collaboration with the Sri Lanka Navy to get the data and trial implementation of this work.

With the support of Oshadi Jayaratne at SLTC, researchers plan to present a tool that can be used by the public to estimate the probability of being infected.