Throughout history, humankind has faced numerous challenges due to infectious diseases, and the battle isn’t over yet: Outbreaks continue to emerge or re-emerge at extraordinary speed.
Are we now equipped to keep up? Can data science, technology, and artificial intelligence (AI) help prevent, understand, and mitigate the risk of infectious diseases?
According to the World Health Organization (WHO), different parts of the world have witnessed disease outbreaks, epidemics, and pandemics caused by over 20 infectious agents over the past decade alone, with COVID-19 recently becoming a major global threat. Thankfully, the world today is better equipped to battle this latest menace, from implementing containment strategies to executing mitigation and suppression interventions, all while urgently working to develop an effective vaccine.
Additionally, global monitoring and early warning systems are becoming more prominent, using data science, AI, and machine learning to turn torrents of unstructured data into structured data for timely, actionable insights. Applying this information can help mitigate the spread of a pathogen, as well as monitor health-seeking behaviors and public sentiment during a disease outbreak or pandemic.
See also: Today’s Challenges in Infectious Disease
Many infectious pathogens, particularly viruses, recombine in hosts. A reservoir is an animal, plant, or environment in which a disease can persist for a long period of time. When changes occur in an animal species, such as a bat, and take place over a long period of time, it defies the ability to detect them early before they emerge, putting the burden back on human surveillance systems to identify dangerous pathogens in real time.
An example of an initiative to identify health events related to infectious diseases is the Program for Monitoring Emerging Diseases (ProMED) developed by the International Society for Infectious Diseases (ISID). The largest publicly available system conducting global reporting of infectious diseases outbreaks, ProMED has been the first to report on numerous major and minor disease outbreaks and serves as an essential source of information for clinicians and laboratorians around the globe. A number of private companies and research collaborations have also emerged in this space, employing sophisticated technology in order to combine public health, advanced data analytics, and medical expertise in order to follow and forecast infectious disease risk.
More and better data has been, and continues to be, the basis for improvements in epidemiological and mortality models, as well as economic projections. In many cases the volume of data requires the kind of analysis that only AI can provide. A few days before the WHO issued a warning about the novel coronavirus, AI technology had already noticed the threat posed by the virus and also identified the cities highly connected to Wuhan, China, the geographic origin of the virus. Using data from hundreds of thousands of sources and natural-language processing (NLP) to interpret it, such automated surveillance systems are able to detect the presence of infectious diseases. Their predictive algorithms can also provide insight into how diseases are likely to spread by analyzing travel data.
See also: Epidemiological models explained
Social media is another major data source being used to monitor and forecast public behavior while measuring disease activity during epidemics and pandemics. The digital epidemiology field continues to grow. Researchers track and assess digital data sources such as social media posts using specific search terms, keywords, and illness surveillance data to create accurate estimation models. However, it is important to strictly adhere to all privacy regulations and to remember that social media and data may create unnecessary fear and hype and spread significant misinformation.
Beyond surveillance and modeling, AI can help inform epidemiologists, healthcare officials, and governments about the appearance and behavior of diseases. Such real-time platforms can not only forecast and track outbreaks but also help with coordinated resource planning during a crisis. By running simulation tests to screen various chemical compounds for potential drug development, AI is being used to accelerate drug treatments and even vaccine discovery, expediting the usual trial process for this kind of testing. In addition, AI is helping to identify new drug targets for therapeutics with the aim to improve treatment efficacy.
Medical use of AI is nothing new: It has been used to accelerate genome sequencing, drive faster diagnosis, and carry out radiological imaging analysis, among other applications. Although it doesn’t eliminate the need for human expertise, AI has also facilitated greater access to scientific research and publications worldwide, helping to optimize individual patient care. Such international collaboration and data sharing have been encouraging throughout the COVID-19 pandemic.
The use of big data, innovations in technology, and AI all offer new hope in pre-empting, preventing, and combating the threats of infectious disease outbreaks, as well as facilitating our understanding of public behavior during health crises.
While the world works to stop the spread of SARS-CoV-2, the progress we make today should help better prepare us for the infectious disease threats of tomorrow.
