How could AI help tackle the coronavirus outbreak? Please explain in detail.AI and analytics can help to detect early signals of symptoms that would point at a possible new epidemic. With these sophisticated techniques, early signals can be found often weeks before officials raise the alarm and this can help limit the spread of the virus. They require special analytical techniques that can find rare but meaningful events, such as a spike in school absenteeism in a certain region or state. Each outbreak requires a combination of epidemiological, clinical and AI skillsets to adapt to the infectious agent or virus under study.
To increase accuracy and precision, diverse information sources are combined into analytical data sets, e.g. official incidence records, clinical emergency data, physician’s records, social media, flight records, school absence, and sales data of anti-fever medication. AI can also help to complement the clinical findings, specific adverse events, and model characteristics of a new viral epidemic or pandemic such as 2019-nCOV. These early findings are crucial to ready the health care system and ensure the right capacity to put patients in quarantine or have enough antiviral medicines and materials ready.In a later stage, policy decisions like the need for quarantines can be further evaluated; today it is estimated that only 1 out of every 20 infected patients are being diagnosed for 2019-nCOV so the worst is yet to come.
AI can even help to predict where next outbreaks with new viruses will happen, by scanning for high-risk open-food markets with lots of people around. AI can help in automation tasks for physicians and citizens, e.g. in the use of chatbots to rapidly survey citizens for symptoms. These systems can deal with thousands of patients per hour, unlike call centres, and generate high quality reports. AI also helps in the clinical discovery, trials and manufacturing to ensure safe and efficient antiviral medicines and vaccines. AI especially excels at seeing connections and correlations that humans would not find or observe.Can AI be used in producing a vaccine for the coronavirus? How can AI help in this regard?
AI and analytics is used during every step of the development, manufacturing and commercialisation of vaccines already today. Clinical trial information is analysed using SAS and other analytical technology and show to authorities that the new vaccines are working and safe in a strictly controlled regulatory context. Activity is tested during and after manufacturing of vaccines using AI, and the quality of vaccine batches is monitored with a whole plethora of analytical techniques, such as image analytics and shelf life analysis.In addition, once the vaccine is administered to the population, possible adverse events are collected, analysed and reported using AI to see if it safe. This is called pharmacovigilance and some of the analytical approaches to detect rare adverse events have similarities with epidemic surveillance.
AI is also using to screen scientific literature and other sources of unstructured information such as social media to detect consumption trends or countries or regions that have not yet been vaccinated. There are a lot of other areas where AI is becoming important such as digital or augmented clinical trials that allow capturing patient wearables or medical devices and learning more about the effect of the investigated vaccine or therapy.
Syndromic surveillance uses clinical features that are showing without a formal diagnosis being confirmed in an individual. With text mining and social media analysis – advanced text analytics or natural language processing techniques can detect entities in free-form text, understand its context and use the resulting digital information for further statistical analysis.Surveillance of symptoms or social media analytics alone is not able to reliable detect a new epidemic (like the halted Google Flu Trends project has demonstrated) but it is in the combination of different data sources and the application of sophisticated rare event analysis that data scientists and analysts can start to investigate and look for patterns.
How easy or difficult it is to deploy “predictive analytics” at hospitals and airports, in the UAE as an example?Hospitals and airports already use predictive analytics technologies to better predict when nurses and doctors will be needed, score patients for the risk to develop sepsis, or score travellers for possible security or health issues. When starting to implement this fascinating technology, an analytics culture needs to be established and impactful use cases need to be identified. This requires considerable investment, a true data-driven analytics strategy supported by hospital or airport management over years. These investments need to occur well before a pandemic like 2019-nCOV starts.
Have any of these new technologies been deployed in any country so far? If so, where and how accurate were the results in general?As mentioned, all these techniques have been deployed in hospitals, countries and at government agencies that deploy surveillance techniques at a lot of hospitals and government health systems all over the world. What changes between countries is the maturity of the e-health system, the readiness of the health care system to execute decisions and measure the outcomes, and the ability to gather high quality digital information. Not all regions and countries have centres of excellence where clinical specialists can closely work together with the statisticians and data scientists.
As you know, delivering the right information to the public - thus preventing an “infodemic” - is important in the battle against the virus. How can AI and/or new technologies help in this regard?At SAS, our vision is to transform a world of data into a world of intelligence. Intelligence, or the results generated by AI, should drive decisions. By being transparent about those data-driven decisions, and explaining what these results means, what actions are taken to contain the problem and how the spread of the virus can be controlled – the public can deal with the reality and will understand. Transparency, showing the results of new programs, building dashboards for everyone, is key to build trust.
It is the rumours that are not founded in reality cause irrational behaviour or panic. But not everything can be controlled – and that’s why governments, private companies and the WHO must do more to be ready for the next epidemic. We have learned a great deal since the SARS epidemic of 2003 and the ebola epidemic of 2013-2016 but we need to remain vigilant and invest in both new vaccine technologies (such as the plug-and-play vaccine technologies), good health care networks and a digital health system that can generate warnings and shape policy.