We may never know the full impact Artificial Intelligence is having on the effort to help control the spread of the COVID-19 virus. However, we can be sure it is playing a significant role. Artificial Intelligence, in the form of chatbots, is being used to automate communication with concerned and at-risk patients. Artificial Intelligence, in the form of natural language processing, helps to quickly disseminate the most useful research data. Artificial Intelligence, in the form of machine learning, is tracking and predicting the spread of the disease.
Using Microsoft’s Healthcare Bot Service in Azure, the United States Centers for Disease Control and Prevention (CDC) was able to create a COVID-19 virus Assessment Bot. This assessment bot is an Azure based public cloud service. The fact that this bot is available through the Azure cloud has allowed hospitals and other organizations to quickly build and implement their own custom chat-bots. These chat-bots became a very significant tool when cases of COVID-19 started appearing in the United States. Hospitals started to become overwhelmed with patients that thought they might have COVID-19. Once implemented, the chatbots were able to have a conversation with a patient and ask them a series of questions to help determine if he or she should seek further medical attention. These chat-bots also help eliminate some of the fear patients may have been facing by showing them their illness may be something other than COVID-19.
Natural language processing (which is a part of artificial intelligence) has also played a significant role fighting the COVID-19 virus through the availability of data and research articles. The “COVID-19 Open Research Data” or “CORD-19” data set was created by a group of tech leaders including Microsoft, Chan Zuckerburg’s charity, and Seattle’s Allen Institute for Artificial Intelligence. This data set is now hosted on the semantic scholar website. The CORD-19 data set takes all articles related to the COVID-19 virus and uses natural language processing and machine learning to present the information to researchers in a more useful manner. The CORD-19 database saves valuable time by cutting out the unnecessary data and articles. This data set is also machine readable, which allows researchers to mine data more efficiently after it has been dissected by the machine learning and natural language processing. This data set is a huge step forward in the collaboration between the artificial intelligence community along with the scientific and medical community.
Tracking where the disease is and where is it predicted to spread was very important early in the pandemic. Dashboards tracking the virus became very popular as the number of cases continued to grow and artificial intelligence, more specifically machine learning, played a big role in providing data for those dashboards. Machine learning came into play because of its ability to handle large amounts of data. With this data, machine learning algorithms were used to help predict where the virus would spread next. Machine learning also came into play to help determine the relative success of quarantine measures. These models were also able to estimate the number of cases per country and when certain countries would plateau based on the effectiveness of their preventive measures. Early on, with the lack of historical and unbiased data, it became difficult to train the models. However, as time passed, and more data became available the models became more accurate and more useful.
As a society we may never know how long the pandemic could have lasted or how dangerous this virus could have been without the aid of Artificial Intelligence. With the use of chat-bots, natural language processing, and machine learning numerous areas of research were easily and quickly able to combine their expertise and efforts to help control the COVID-19 virus.
https://healthitanalytics.com/news/machine-learning-tools-predict-impact-of-quarantine-on-covid-19
https://www.healthcarefinancenews.com/video/using-chatbots-fight-against-covid-19
https://towardsdatascience.com/machine-learning-methods-to-aid-in-coronavirus-response-70df8bfc7861
https://towardsdatascience.com/how-to-fight-the-coronavirus-with-ai-and-data-science-b3b701f8a08a
