Last updated: 13 March 2023
Published on: 29 July 2019
7 MINS READ
By Arun Poojari
We’re in a time of unprecedented change.
Governments and public officials are working closely with leaders in healthcare, technology, education, and nonprofits to pool together resources to combat the COVID-19 virus.
While it’s still unknown when life will return to normal - and what that ‘normal’ may look like - companies like ours are contributing AI technology to fight COVID-19 and its after-effects.
AI-based tools have the unique ability to uncover and identify hidden patterns buried deep in data to help in the fight against COVID-19.
By tapping into millions of geospatial points of interest, social and demographic data, we can track and predict the spread of the virus, discover areas where ‘hot spots’ show an increase of infected carriers, and recommend the easing of restrictions.
Here are our top 5 use cases for battling COVID-19 with AI technology:
Dynamic, predictive heatmaps
The ability to predict - with high levels of accuracy - the likelihood of an asymptomatic carrier passing a specific spot allows authorities to create counter-measures aimed at preventing the virus from spreading, in both low and high risk areas.
By augmenting movement data with millions of external data points, hundreds of common high-level patterns of movement can be identified prior to diagnosis.
This kind of data can be reflected as a heatmap that visualizes movement in real time, as shown below.
These types of visualizations update and refresh automatically, predicting places where asymptomatic carriers are likely to pass, giving local officials the information they need to make decisions to enact social distancing measures, create education policies, and establish processes for working people.
The governments of Israel, Italy, Germany, and France have already begun using advanced analytics to create solutions to their various COVID-19 challenges. Each country’s health ministry utilizes anonymized individual movement data in the days prior to positive COVID-19 test results.
Easing the strain on healthcare systems
As the COVID-19 virus spreads quickly, hospital administrators and healthcare workers are in danger of being overwhelmed with too many patient, straining each country’s healthcare system.
The power of AI technology lies in finding unique insights into the virus spread, including the types of people most affected, severity of impact, and the ability to predict the likelihood of complications.
By obtaining data on patients who have tested positive and combining it with other data such as occupation, age, medical history, and mobility, we can compare this to the general population and build a predictive model.
This helps hospital administrators and healthcare directors determine whom to test, how to prioritize treatment, and decide the right amount of medical personnel to be deployed in the areas most affected by the virus.
Decision makers can also use these insights to redirect low-risk patients to other hospitals and free up resources for more seriously affected patients.
Law enforcement can increase compliance
While home isolation restrictions have become commonplace, social distancing policies in public areas are being monitored to measure their effectiveness.
Innovative AI systems can quickly identify patterns of where social distancing measures have successfully limited movement and where they have been ineffective.
Most importantly, this insight can be seen within a city block, postal code, or county resolution.
We can pinpoint areas where law enforcement may need to be dispatched in order to increase awareness and adherence to social distancing measures and the wearing of protective gear.
Predicting the impact of weather
The weather may be a key factor of COVID-19 infection rates.
Scientists and researchers believe that as warmer weather develops, the number of virus cases may decrease, while colder weather may cause a rise in infection rates.
By augmenting weather reports with medical, transportation, and demographical data, it’s possible to predict how rising or decreasing temperatures impact the spread of COVID-19. The use of these types of external data sources for collecting and classifying massive amounts of information is therefore critical.
Ultimately, decision makers can use this information to prepare to handle an increase of infections or to free up resources in the case of fewer infections.
Easing and lifting restrictions
In order to determine where, when, and how the easing or lifting of restrictions should take place, policymakers must first understand how people are moving about, where they go, how many others they come into contact with, and how long they remain outside.
The lifting of restrictions must be done responsibly.
AI technology can accomplish this by testing millions of patterns across a wide range of scenarios in minutes, minimizing health and economic risks and maximizing positive impact. These patterns may reveal specific industries which pose less of a risk of the virus spreading and therefore may be prioritized for restriction lifting above others.
A heatmap that visualizes mobile movement combined with medical data can shed light on whether restrictions should be lightened or lifted completely. It also shows how effective social distancing restrictions have been, helping public officials decide when - and how - to act.
Combining geospatial data on a large sample of the population with external data can determine effective courses of action for easing COVID-19 restrictions. This could include data sources such as cell tower communications or ad-based tracking information, with the use of ‘Blindfolded Analytics’ to maintain users’ privacy and anonymity.
Real-time insights can help public officials strategize plans for the gradual release of lockdown processes for their communities. Business leaders can use the same insights to create safe plans for a gradual return to work and the resumption of business operations.
Up to 80% of traditional models and forecasting systems have been built on frameworks that are no longer relevant in a COVID-19 world. Business intelligence must keep up with the rapid pace of change.
That’s where AI-powered systems excel, by rapidly identifying developing trends and patterns, even where limited data exists, and by automatically updating as new data emerges.
Once restrictions are lifted, we'll be able to deploy resources and create effective solutions quickly and accurately, turning the disruption of COVID-19 into new business opportunities.
AI isn’t limited by human bias
One final note about bias.
A human data scientist or analyst may need weeks or months to extract key information from hundreds of thousands of data points.
As humans are limited by bias, they would likely only find answers to questions they think to test, leaving so much unexplored. During a global pandemic, we simply can’t afford to leave any stone unturned.
At SparkBeyond, we believe in using next-gen AI technology to solve today’s real world problems without the limitations of human bias.
Together with government and business leaders, we’re uncovering new solutions to COVID-19 with the help of AI-powered solutions, and helping to create lasting, positive impact in the years to come.
This article was contributed by Arun Poojari General Manager, Asia Pacific & Japan of SparkBeyond, an SG:D Accredited company in the IMDA’s Accreditation@SG Digital community. The opinions and views in this article do not represent any official positions of IMDA.
First launched in July 2014, Accreditation@SGD contributes to an innovative infocomm media ecosystem by accrediting promising Singapore-based tech product companies to establish their credentials, build business traction, and help them to grow and compete in the global market. The evaluation process provides an independent third party evaluation on the SGD-accredited companies’ claimed product core functionalities and ability to deliver.
As of February 2020, over S$430 million worth of project opportunities have been created for accredited companies. Close to 1,000 projects have also been awarded. For more information, please refer to this link: www.imda.gov.sg/accreditation