AI-powered algorithm released to detect the third wave in South Africa

Credit: Wit University

An artificial intelligence (AI) -based algorithm designed by the University of the Witwatersland (University of the Witwater) in collaboration with the Gauteng state government iThemba LABS and York University in Canada shows a one-third lower risk. COVID pandemic infectious waves in all South African states.

An AI-powered early detection system predicts future daily confirmed cases, including features such as mobility index, stringency index, and epidemiological parameters, based on historical data from South Africa’s past infection history. Works by

“These parameters are consistent with clinical public health measures that can contain, control, and mitigate the COVID-19 pandemic,” said Dr. James Orbinski, director of the Institute for Global Health, York University. Says.

AI-based algorithms work in parallel, supporting data from existing algorithms based on more classical analysis. Both of these algorithms work independently and are updated daily. The presence of two independent algorithms adds robustness to the predictive power of the algorithms. Data for AI-based analytics is published on a website that is updated daily.

“Current data show that most states in South Africa have a low risk of a third wave of COVID-19 infection, but they are still very vulnerable,” said Witt’s Collider Particle Physics Study. Professor Bruce Melado, the director of the institute, said. University.

It is important that South Africans continue to comply with the South African Government’s COVID regulations and take all necessary precautions to prevent the spread of the pandemic.

The appearance of waves of infection is caused by situations that are difficult to predict and therefore difficult to control. In this complex environment, early detection algorithms can provide early warning to policy makers and the public. The early detection algorithm can issue an alert when the data shows a significant change that coincides with the appearance of a new wave.

Algorithm-based predictions are never 100% accurate, but Mellado is confident that the model will present very good predictions for at least two weeks. Predictions can be made over a long period of time, but these predictions are less accurate.

Models are trained in the mid-term between Wave 1 and Wave 2 in all South African states. The algorithm was tested with data acquired during past peak periods to assess its performance.

“AI technology offers invaluable potential for developing early detection and alert systems that are highly needed for rapid and dynamic decision making under current pandemics of risk and uncertainty. “We do,” said Arias Garry, a professor of disaster and emergency management and associate director of York. Simulation of advanced disaster, emergency, and rapid response at the university (

AI is very effective at navigating complex problems with many parameters and dimensions while learning from the data. The data hides a wealth of information that AI can efficiently extract.

“Our team’s development of a third-wave early detection algorithm demonstrates the power of AI to generate database solutions for very complex problems,” said the director of the Collider Particle Physics Institute. Professor Mellado said.

AI helps identify new COVID-19 hotspots in Howten

Courtesy of Wit University

Quote: The AI-based algorithm released to detect the third wave (April 12, 2021) in South Africa is https: // -Obtained from africa.html on April 12, 2021.

This document is subject to copyright. No part may be reproduced without written permission, except for fair transactions for personal investigation or research purposes. The content is provided for informational purposes only.

AI-powered algorithm released to detect the third wave in South Africa

Source link AI-powered algorithm released to detect the third wave in South Africa

Related Articles

Back to top button