The public notification of the Pandemic Covid-19 or Coronavirus by the World Health Organization (WHO) and the US Centre for disease control (CDC) were made on January 9, 2020, and January 6, 2020, respectively. However, an AI-based company Blue Dot detected and provided an early warning system for the disease as early as December 31, 2019, to its customers. The technology it follows is predictions based on various inputs ranging from foreign news to flight ticket patterns. A similar model was used by AI-driven Health Map predicting data from local news and social media posts such as We Chat and Weibo. The early prediction model stands out when referring to roles AI will play, especially in case of such large outbreaks. China, which was the epicenter of the Corona virus, deployed AI on large scale restricting the movement, genome sequencing, chatbots for queries, maintenance of, and delivery robots.
The connection of AI to the medical industry has been gaining ground for the past years with varying degrees of success. However, coronavirus has presented a different case where the pandemic has been the gateway to AI adoption in healthcare. The various applications ranging from detection of the symptoms, drug development to managing the disease through various tools that AI provides have proven consequential in the ongoing fight against the virus, including in deep learning. Others are working on drugs that target the virus base on the molecular setup of existing drugs. In demonstrating AI capability, the scientist team has been successful in detecting Covid-19 and differentiating it from other acquired diseases like pneumonia. Though, the usage of the CT scan for detecting Coronavirus has been disputed by other doctors, especially for initial detection.
Amidst the overwhelming burden on the medical staff and authorities, expanding the scope of detection by AI is being assigned innovatively. Researchers launched a voice assistant that can help callers suffering from potential coronavirus symptoms and direct them towards emergency services or their doctors using artificial intelligence. Others have put early detection to screen patients providing X-rays of the lungs who can be further put to the test depending on the results. The application in the case of biomedical research is a specialized subject, and criticism has mounted where companies often have prepared and passed applications without clinical trials. Also, the real-world setting is very different from changing variables of demographic, ambient, temperature, and movement. An AI company, Athena, utilizes its surveillance technology for thermal screening of fever, requiring deep face scans where the probability of accurate detection and bias in data is prominent.
Apart from medical applications, the social challenge has been presented by large scale misinformation over the pandemic. The WHO termed the pandemic as ‘infodemic’ with various news floating about it with conspiracy theories such as the virus originating in Chinese lab to the various remedies available to treat the virus. In Iran, people consumed methanol amid fake news being circulated about it being a cure resulting in numerous deaths. Guidance for reliable news has emerged at the forefront of the challenges facing the epidemic, where tech companies are employing AI. Chatbots are dispensing information about Coronavirus while the world’s largest search engine Google, is privileging information from WHO to remove misinformation from search engines and YouTube. Moreover, the tech companies have partnered to model the Covid-19 economy recovery data based on the consumer patterns based on economic, business, travel, and retail data set combined with behavior and sentiment data. These innovative ways of the application of the AI are especially helpful in the management of the disease.
Several Governments have asked tech companies to assist in the pandemic. The White House asked for the role of tech companies in overcoming the pandemic in the US. One particular usage has been geolocation of major affected areas through mobile data. However, the absence of the regulatory framework to provide data security on its usage is still a sticking point. The example of the Cambridge Analytica, where data of the Facebook users was used for political purposes, is a case in point.
In the review of the various AI-based systems and health applications available, there is limitation depending on the context of datasets (its historicity) and, most importantly, the inputs by practitioners. Researchers point to the context of data, medical applications where doctors are key to understanding it. The convergence of medical and software is a pathway for further expansion, especially considering the scale and scope of a pandemic like Covid-19. Knowledge is key to understanding the virus. At the current stage, AI’s role can be categorized as an assistive technology with evolving conditions pushing for its usage on a large scale in the future.
*** The author is currently a PhD scholar at the Centre for Canadian, US & Latin American Studies, School of International Studies, Jawaharlal Nehru University ***