Global AI initiative to fight COVID-19

Coronavirus (COVID-19) Infected Persons and Deaths Count

The mysterious Coronavirus (COVID-19) is not an unknown subject anymore; with it spreading at an alarming rate, there have been over 27,78,634 infections 1,70,389 deaths that have been caught worldwide. On 28th January, 2020, the World Health Organization (WHO) declared the coronavirus as a  Global emergency after it exceeded the numbers infected during the 2002-2003 outbreak of SARS (Severe Acute Respiratory Syndrome) in China.

While every country is working hard in order to quell the virus through quarantines, lock-downs, travel limits and accelerated research, it seemed to have gone out of control. However, the question of whether technologies like AI can help has arisen. And interestingly enough, it already has.

It is for sure that at the moment, Artificial intelligence is not going to stop the outbreak of the new virus or replace the role of the experts in it. However, according to health data specialists, for the first time in a global outbreak, it has definitely become a useful tool to monitor and respond to the crisis. As history shows, In prior outbreaks, AI offered limited value due to the shortage of data and equipment. However, with the development in technology as well as in social media, a million posts associated with the coronavirus over different media types are allowing algorithms to generate close to real-time information for public health officials to track its spread.

Considering that AI is constantly being used in detection and diagnosis of the disease, below are 10 ways how Artificial intelligence and data science technologies are currently being used to manage and fight COVID-19:

1. To identify, predict and track disease outbreaks

BlueDot, a venture-backed startup has managed to build a sophisticated AI platform that processes billions of pieces of data to identify these outbreaks. This data can include the world’s air travel network, health reports, and many others. With respect to the case of the coronavirus, BlueDot made its first alert on December 31st, 2019. This was much ahead as opposed to US Centers for Disease Control and Prevention, which made its own determination only on January 6th, 2020.

An MIT team has recently trained an ML Model that accurately predicts the spread of COVID-19. The team has trained the model to extrapolate publicly-available data for insights into the disease’s spread — taking into account different factors of social distancing, quarantine and standard epidemiology parameters handled by different governments. According to the results, we are to see a plateau where the amount of new cases begins to level off in the US and Italy in the coming weeks. The data also further indicates that strict public health measures will play a role in the success of halting the spread of infection and prevent it from exploding exponentially.

2. To help in the diagnosis of the virus

Incidentally, BlueDot is not the only organisation to have successfully leveraged AI in helping to curb the coronavirus. Alibaba (a Chinese tech company) said AI can detect Covid-19 in just 20 seconds! The company has said that with the help of AI in CT Scans, it can detect Coronavirus in just 20 seconds. The model has a precision rate of 96% and if doctors were to do the same, it would take up to 20 minutes. The algorithm was built by Alibaba’s research institute named Damo Academy. The AI model was trained with the help of data from more than 5,000 confirmed cases and can identify patients with the deadly virus in CT scans.

The Company has also joined forces with DingTalk to launch a series of AI technologies along with Cloud-based solutions to support organizations around the world in the fight against Covid-19. These new technology offerings have been designed to help researchers and medical practitioners to address and overcome challenges across several major areas associating the virus. 

Another AI-based startup, Qure.ai, is assisting hospitals in installing their AI-based Softwares to analyze X-rays and CTs. While the company has been working with this software for the past 3 years, the COVID-19 installations just got going about two weeks ago. A huge advantage of their software is that it can be deployed remotely within a matter of hours. This means that lockdowns will not impede any new sites from going live. Furthermore, this technology can be used even by mobile units – to go door-to-door in poorer neighbourhoods to conduct on-site tests.  

3. To process healthcare claims

With the virus spreading like a wildfire, not only are the clinical operations of healthcare systems taxed but also the business and administrative divisions of it. A blockchain platform offered by Ant Financial is assisting in speeding up the processing claims by reducing the number of face-to-face interactions between patients and hospital staff.

4. Drones to deliver medical supplies

One of the safest and the fastest methods of delivering medical supplies during an outbreak is through drones. Terra Drone is currently using its unmanned aerial vehicles to deliver and transport medical supplies with little risk between Xinchang County’s disease Control Centre and the People’s Hospital. 

5. Robots to sterilize, deliver food supplies and perform similar tasks

Considering that robots are not susceptible to the virus, they are being deployed to complete tasks like cleaning, sterilizing and delivering medical supplies in order to reduce the amount of human-to-human contact. Pudu Technology, in China, has deployed robots that were primarily used in catering to more than 40 hospitals around the country.

6. To assist in the development of drugs

Within weeks of the outbreak, AI’s predictive capabilities already began proposing existing drugs that might be useful in its cure. Google’s DeepMind division uses its latest AI algorithms to study and understand proteins that make up the virus and are publishing their findings to help in the development of treatments.

ChemRxiv is using AI algorithms to identify progeny drugs similar to the parents, which is already being tested against COVID-19. These algorithms evaluate similarity not only by the molecular make-up of the molecules but also by the context in which each specific functional group is arranged – by a three-dimensional distribution of pharmacophores. Progenies are essentially either already approved drugs or are medications in advanced clinical trials. However, should the currently tested parent medicines fail, the progenies could be re-purposed against the COVID-19.

Makers Lab, the R&D arm of Tech Mahindra Ltd., has leveraged Artificial Intelligence to conduct research and create potential therapeutic drugs for the treatment of Covid-19. The Lab has collaborated with renowned bio-scientists to find plausible therapeutic drugs and to synthesise and test compounds for the same. Considering the high transmission rates of Covid-19, the team is using Molecular docking approaches to make a difference and create disruptive solutions for the virus. According to Nikhil Malhotra, Global Head of Makers Lab, their objective is to prevent the entry of the virus into human host cells since it is harmless outside a host. The technique – molecular docking, allows the search for therapeutically potent drugs and molecules in real-time, to find compounds that are capable of acting as inhibitors against a viral protein computationally.

7. Everyday Wear for Precaution and Protection

An Israeli startup, Sonovia aims at to arming healthcare systems with face masks made from their AI-based anti-pathogen, anti-bacterial fabric that relies on metal-oxide nanoparticles.

With fever being one of the symptoms of the virus, security staff at Hongyuan Park in eastern China’s Hangzhou city have been given AI-enabled glasses that can detect the body temperature of visitors. Developed by an Hangzhou-based startup, Rokid Corp, these glasses work along the concept of non-contact thermal augmented reality. Fitted with a camera and a cable, the regular-looking glasses detect anybody with a fever – immediately sending alerts and making digital records.

8. Identification and risk predictions of the infected

A team of physicians at the Zhongnan Hospital at Wuhan in China are now using a GPU-accelerated AI software model to detect visual signs of the COVID-19. The model relies on NVIDIA GPUs for both training as well as for inference, and it has managed to help overworked staff screen and prioritize patients that are likely to have the virus. Developed by a Beijing-based startup Infervision, the software detects signs of pneumonia caused by the virus. While the software was primarily used in the detection of cancer in lung CTs, the newly developed model looks more closely at coronavirus pneumonia. Trained with over 22,000 notations by human radiologists, the algorithm overlays X-rays with color-coded maps that indicate pneumonia probability. Furthermore, the algorithm can consistently localise areas of pneumonia, despite the fact that the images were taken at several different hospitals – considerably varying in technique, contrast and resolution.  Till now the new system has been deployed in over 34 hospitals in China and has successfully reviewed more than 32,000 cases of the coronavirus or COVID-19.

Also read: Detection of COVID-19 Infected Lungs Using Transfer Learning

Computer scientists at the University of Copenhagen are currently developing AI-based Computer Models that are capable of calculating the risks of an individual patient’s need for a ventilator or intensive care. While the models are not aimed to treat individual patients, their objective is to create a planning tool to make a big difference for hospital staff. This will be useful for hospitals to plan and deploy their resources in the best possible way.

9. Chatbots to share information and spread awareness

AI-based Chatbots are an essential means of communication for people to access free online health consultation services, updates on the latest travel procedures and disruptions, and more.

10. Supercomputers to work on Coronavirus Vaccines

Supercomputers of various major tech companies like Tencent, DiDi, and Huawei are being used by researchers for the development of a cure or a vaccine for the virus. The speed these computers is much faster than the standard computer processing and hence can run faster calculations and provide model solutions for necessary situations.

11. AI tools could tell if someone is infected with the help of voice recording

A team of researchers from MIT and Harvard are working with machine learning models to identify specific characteristics from the voice recordings of the Covid-19 patients and healthy people. They are trying to identify specific vocal signatures that could indicate that an individual is carrying the virus. Similar research is being conducted at Carnegie Mellon University’s Cylab.

If the research is successful, it will be an easy and valuable alternative for Covid-19 testing, especially among individuals who do not have easy access to Covid-19 test.

12. AI Robots for Touchless Sanitation

Companies in South Korea are using AI powered robots to automate sanitation procedures. This method has enabled a safe process owing to its minimal human interaction. These robots are equipped with cameras, temperature mapping systems, and LED screens so that every time a visitor greets it, these robots can examine and sanitize them in a touchless manner. They also dispense hand sanitiser and disinfect the floor.

An increasing number of companies have resorted to this hands-free touchless system to minimise human contact in the midst of the outbreak of the deadly COVID virus. An official from these companies confirmed this affinity towards AIML technology, saying that robots take much less time in checking temperatures and they do that accurately, apart from encouraging social distancing. Hence, this practice has been highly encouraged.

AI’s potential only continues to grow each day. In fact, with AI-empowered algorithms that are genome-based neural networks have already been built for personalized treatment. And this, according to specialists, can prove to be extremely helpful in managing adverse events and symptoms of the coronavirus.

Are you curious to learn more about AI? Explore and build a career in Artificial Intelligence.

10

LEAVE A REPLY

Please enter your comment!
Please enter your name here

7 + seven =