Akhtar Nazir & Dr Abrar Ul Haq Wani
The medical-scientific response to Covid-19 is analogous to that towards the previous epidemic of SARS but with a major difference: it involves a new emerging contrivance known as Artificial intelligence (AI), which for current purposes can be defined as Machine Learning (ML), Natural Language Processing (NLP), and Computer Vision applications that use big data-based models for pattern recognition, explanation, and prediction.
This multifunctional AI will be useful to diagnose, predict, and treat Covid-19 infections. It will also lend a hand in managing socio-economic impacts. There has been a hustle, since the outbreak of the pandemic, to use and explore AI and other data analytic tools. Baylor College of Medicine, USA, asserts that AI/ machine intelligence can be used to predict how the Covid-19 outbreak might be affected by seasonality. Historically, respiratory coronavirus infections peak in the winter months and AI might help in predicting how warmer weather may limit its spread. Such a forecast could stabilise financial markets right now.
Effective AI needs high-quality input data because it’s a case of garbage in, garbage out. There are a lot of AI companies that are working to contain the current pandemic. One such company is Infervision, a Beijing-based AI company that uses its algorithm to spot Covid-19 on lung images as different from other respiratory infections. Infervision’s AI applications accelerate the diagnoses and monitoring of Covid-19. As more and more scans are finished, the algorithm learns and improves precision. From a lung CT scan, the AI is designed to swiftly identify lesions of possible Covid-19 pneumonia, to measure their volume, shape, and density, and to compare changes of multiple lung lesions from the image, which together provide a quantitative report to assist clinicians in making a fast judgment. While a manual interpretation of a CT scan can take up to 15 minutes, AI can finish reading the image in 10 seconds. PCR-based diagnosis takes too long (sometimes over a week), but CT imaging with Artificial Intelligence could serve as a surrogate for doctors when fast judgment is needed.
It is still too early to tell if and to what extent AI will have an impact on the Covid-19 outbreak. The numbers of confirmed cases and deaths climb daily and so too does the supply of data. One thing is certain: AI is relevant to this outbreak and in the future, it will become even more so. We can see countries like South Korea, Japan, and many more containing the spread of this virus using big data analysis and AI-enabled warning systems. How is this all done? The logic is simple: the government stores information of all citizens and integrates all government organisations, hospitals, internet service providers, and other services into it. Whenever someone is tested positive for Covid-19, all the people in the vicinity are updated with a person’s travel details and activities for the previous two weeks through mobile notifications. Hospitals receive the same information.
There also has been drive-through coronavirus testing where a person drives his car inside a mobile testing lab, gets his sample collected while sitting inside the car, and then gets the results in minutes. If found infected, the person is immediately isolated. While driving on the road, users are updated with the location of the nearest drive-through lab where they may undergo medical tests. Several research projects are using AI to identify drugs that could now be repurposed to take on coronavirus. By studying the molecular set-up of existing drugs with AI, companies want to identify which ones might disrupt the way Covid-19 works.
The pandemic has resulted in great demand for digital services for healthcare and monitoring, and countries across the world are using solutions for population screening, tracking infection, allocating resources, and managing communication and responses. Larsen & Toubro (L&T), India’s biggest engineering conglomerate, is using new technologies, artificial intelligence, and its digital platforms to support local authorities in 20 cities so that they can monitor and implement measures to help combat the spread of Covid-19. L&T has rolled out smart technology solutions in these cities that help local administration in monitoring crowds, tracking patients, communicating with the public – and even maintaining law and order.
AI is not yet playing a significant role in the fight against Covid-19, at least from the epidemiological, diagnostic and pharmaceutical points of view. Its use is constrained by a lack of data and by too much noisy and outlier data. So, the creation of unbiased time series data for AI training is necessary. Not only for providing training data to get AI models operational, but moreover for more effectively managing the pandemic and reducing its cost in terms of human lives and economic damage. So, we must fight it with every available powerful tool, like artificial intelligence, machine learning, and big data.
Akktar Nazir is Youth Ambassador for YIGF India, Dr Abrar Ul Haq Wani is a researcher at Dept of Medicine, FVSC&AH, SKUAST-K