What You Think
How AI is used to curb Covid-19 — Nurafifah Mohammad Suhaimi
Malay Mail

MARCH 19 — While organisations like the World Health Organisation (WHO) and the United Nations are releasing funds to facilitate research on the Covid-19 pandemic, government agencies and private companies around the world are increasingly looking toward AI-based techniques to provide insight on its spread and support treatment for those who have been infected.

Advertising
Advertising

WHO agrees that Artificial Intelligence (AI) and Big Data played a significant role in China’s response to Covid-19. At its core, computer scientists and researchers are actively tackling this outbreak by compiling datasets from the affected patients and building algorithms to "learn” from them.

Also, according to WHO, four critical criteria lead to a successful public health response due to an outbreak of a new infection:

1. Understanding of transmissibility and risk populations;

2. Establishing the natural history of infection, including incubation period and mortality rate;

3. Identifying and characterising the causative organism; and

4. Epidemiological modelling to suggest effective prevention and control measures.

Owing to this, computer scientists believe that these solutions can be applied to each of these criteria to assist frontline health workers as well as supporting epidemiologist and virologist specialists in preventing, diagnosing and treating communicable diseases like Covid-19.

How does AI do it?

Prediction

According to experts, 3 out of 4 newly discovered diseases found in humans are coming from animals. What is more terrifying, research also shows there are around 800,000 unknown animal viruses that could infect humans in many ways.

Fortunately, with the help of AI technology, the hotspots of where these new diseases could arise are possible to forecast. All the past data regarding the already known viruses, animal and human populations as well as cultural or social practices around the world are utilised to forecast the outbreaks.

Hence, in the future, government and health organiations can make use of this data to be proactive in preventing future outbreaks. Or at least, to get them prepared beforehand.

Self-diagnose tool

Recently, the Ministry of Health Malaysia has launched a Virtual Health Advisory Portal Novel Coronavirus (Covid-19) for people to self-diagnose their risk of being infected. This system is not only being used in Malaysia but also in India, the United Kingdom and many more.

Technically, an individual has to answer few questions and AI then uses an algorithm to rapidly assess this information and provide the risk level of the individual — no risk, minimal risk, moderate or high risk.

This tool powered by AI reduces the pressure on healthcare institutions and warning those at high risk of developing the infection. According to experts, this tool helps healthcare institutions to understand the patient’s symptoms and determine the coronavirus risk, thereby educating the public on the virus and collecting data for healthcare officials for early intervention too.

Detection

Alibaba claims its new system can detect coronavirus in CT scans of patients’ chests with 96 per cent accuracy against viral pneumonia cases. And it only takes 20 seconds for the AI to decide — humans generally take about 15 minutes to diagnose the illness as there can be upwards of 300 images to evaluate.

Technically, this is what a computer scientist called anomaly detection. It is a technology that relies on AI to identify abnormal behaviour or unexpected items or events that differs from the norm within the pool of data of viral pneumonia.

Simply put, this abnormal and expected items are used to recognise Covid-19 disease from viral pneumonia more effectively and accurately. Thus, quicker treatment can be done on those infected.

Response

Making an informed decision promptly is crucial once a disease is identified to limit the impact of it. Here, AI can accelerate the time it takes to develop new medications or vaccines for the newly detected disease.

A research institute in Israel is claiming they could have a vaccine within a few weeks and will be available in 90 days. Over the last four years, the institute had developed a vaccine for a bronchial disease affecting poultry.

By chance, they had used a coronavirus as a proof of concept model for their technology. Like SARS, they believe the DNA sequence of their model is like Covid-19 and that increases the likelihood of achieving an effective human vaccine in a very short timeframe, possibly several months.

Thus, this AI-driven drug discovery will transform the health industry and lessen the amount of time and money needed to develop these life-saving vaccines.

Supplies delivery via autonomous vehicle

Apollo, Baidu’s autonomous vehicle platform, partnered with a local self-driving start-up called Neolix is used to deliver supplies and food to the Beijing Haidian Hospital. With daily deliveries, this initiative is helping feed over 100 frontline staff members as they work to treat their patients.

Autonomous vehicles are playing a useful role in providing access to necessary commodities for health-care professionals as it is effectively minimising person-to-person transmission in curbing the spread of the coronavirus.

All in all, at this point, it is very unlikely that the outbreak will be contained in a few locations, according to experts. Nonetheless, people can take steps to reduce the number of infections in your area — by following what the government has ruled out such as restricting movements, cancelation of mass gatherings and practising good hygiene.

Sooner or later, the virus will come to an end and we will be able to live our old, familiar life again.

* Nurafifah Mohammad Suhaimi is Research Assistant at EMIR Research, an independent think tank focused on strategic policy recommendations based on rigorous research.

** This is the personal opinion of the writer or publication and does not necessarily represent the views of Malay Mail.

Related Articles

 

You May Also Like