The World Health Organization (WHO) has declared artificial intelligence (AI) will play a key role in fighting the pandemic. This guide explains how healthcare and AI joined forces to manage the first global public health crisis of the 21st century.
In the next five minutes, learn 7 science-backed ways of using artificial intelligence to combat Covid-19:
1. Artificial intelligence in detecting Covid-19
Detecting the disease is the first step towards stopping the pandemic. Here’s how AI helped:
- A Canadian startup BlueDot first notified the world about the outbreak of a respiratory illness in Wuhan, China. BlueDot uses artificial intelligence to sift through over 100,000 news reports in 65 languages, animal disease networks, and airline data to detect outbreaks and anticipate disease dispersion. Afterward, the startup offers the data to hospitals, public health officials, and airlines.
- Chinese tech giant Baidu used computer vision algorithms for Covid-19 detection using artificial intelligence. Cameras equipped with infrared sensors and computer vision predicted people’s temperatures in public areas. The system screens up to 200 people each minute. The system flags anyone with a temperature above 37.3 degrees. Beijing’s Qinghe Railway Station uses this technology.
- Countries including Korea, Israel, and Poland use contact tracing systems including using contact-tracing apps to detect possible infection routes. As per the University of Oxford’s Big Data Institute, the process is essential as contact-tracing apps can stop the pandemic.
2. Artificial intelligence in diagnosing Covid-19
RT-PCR tests can take up to a few days to deliver the report. Here’s how AI helped in Covid-19 diagnosis:
- An AI system diagnosed 17 out of 25 patients positive for Covid-19 even though their CT scans were normal. Researchers in Mount Sinai, New York built the system using an algorithm trained on over 900 CT scans of lungs from Chinese patients. The AI system cross-references the CT scan data with the patient’s age, symptoms, possible contact with the virus, and bloodwork.
- Researchers at MIT developed an AI system that can diagnose Covid-19 from the patient’s cough. The system uses audio inputs from thousands of coughs and self-reported clinical information.
- Alibaba built an AI system that detects the virus in chest CT scans with 96% accuracy. Furthermore, the system can perform the test in 20 seconds.
3. Artificial intelligence in tracking information
Artificial intelligence helped in tracking the right information in the below-mentioned ways:
- The U.S. government and partner organizations granted access to over 29,000 academic research articles for Covid-19 and coronavirus through the Covid-19 Open Research Dataset Challenge.
- An AI startup Closedloop developed an open-source Covid vulnerability index ‘C-19 Index.’ The predictive model identifies people with the highest risks of severe complications from Covid-19. Healthcare management and insurance companies use the model to identify high-risk individuals and call them to share the importance of social distancing and handwashing. Furthermore, the organizations deliver food, toilet paper, and other essentials to the individuals so they won’t have to leave home.
- John Hopkins Bloomberg School of Public Health, Amazon Web Services (AWS), and a group of volunteer professionals led by White House Chief data scientist DJ Patil built a scenario-planning tool. The tool models the potential impact of Covid-19 to answer questions such as ‘How many hospital beds will we need?’ and ‘What should be the duration of a shelter-in-place order?’
4. Artificial intelligence in finding medicines
Discovering new medicines is time-consuming and expensive. Thus, health officials needed existing drugs that can combat Covid-19. Check out the role of AI in finding Covid-19 medicines:
- Researchers at MIT found targeting the protein RIPK1 can help combat Covid-19. The researchers identified 3 approved drugs that act on the same protein.
- Researchers at MIT and IBM developed an AI system to identify anti-Covid properties on existing drugs.
- A UK AI company BenevolentAI identified that Bactirinib can be an ideal drug against Covid-19. The company arrived at the decision by investigating approved drugs that can inhibit the growth of novel coronavirus using the company’s AI drug discovery platform.
5. Artificial intelligence in spreading the correct information
With the pandemic, arrived the ‘infodemic’ — the sea of misinformation and myths. AI has helped to offer the correct information and clear the myths in the below-mentioned ways:
- A French startup Clevy.io launched a chatbot to share government communications about Covid-19. The chatbot has sent 3 million messages to date. The chatbot is powered by real-time information from the World Health Organization and the French government.
- Countries including the U.S., Italy, France, Canada, and Finland have deployed chatbots and virtual assistants to support healthcare organizations.
- A study in France confirmed chatting with chatbots increased people’s positive attitude toward Covid-19 vaccines by 37%. Furthermore, the percentage of people who had a negative attitude toward Covid-19 vaccines lowered by 20%.
6. Artificial intelligence in discovering vaccines
Without vaccines, we’ll never stop the pandemic. However, developing vaccines can take over a billion dollars and up to 12 years. Here’s how AI helped in developing Covid vaccines to speed things up:
- Google-owned AI company DeepMind developed an AI system AlphaFold 2. The system predicts a 3D structure of a protein’s DNA sequence to within an atom’s width of accuracy. The process helped in creating vaccines much faster.
- AI along with the Vaxign machine-learning platform has accelerated the vaccine discovery process.
- AI used natural language-processing models to help drive Covid-19 vaccine development. The model algorithms analyze data including patients’ social interactions and geographic locations to facilitate vaccine development.
7. Artificial intelligence in minimizing human contact
AI-based robots have reduced the contact between patients and healthcare workers. The process lowered the risk of cross-infections. Here’s how:
- Healthcare facilities in China used robots to check people for fever and other Covid-19 symptoms and dispensing hand sanitizers for everyone. Robots served medicine and food to patients. Furthermore, robots disinfected rooms to minimize contact between humans as well.
- Chinese firms used robots and drones for contactless delivery and sanitizing public areas.
- A company called Boston Dynamics built a robot dog Spot. The robot dog helped lower contact between potentially contagious patients and healthcare workers.
- Doctors in Seattle used a robot for treating and communicating with patients remotely.
Using artificial intelligence to combat Covid-19 — what the future holds
Using artificial intelligence to combat Covid-19 has helped in detecting, diagnosing, and fighting the pandemic.
The International Finance Corporation claimed in its latest report artificial intelligence can help rebuild countries after the pandemic passes. Furthermore, the EU has announced additional funding for using artificial intelligence to help combat Covid-19 as well.
However, governments and world leaders must remain cautious against the misuse of data and the breaching of privacy.