Imagine a world where COVID-19 was not a pandemic, where it had been detected in the early stages before wide spreading. Then the epidemic would have stopped in its tracks by 2019 itself. Even before COVID-19, severe infectious diseases such as smallpox, malaria, tuberculosis and influenza have caused the death of millions across the globe. COVID-19 has claimed at least six million lives and is expected to cost $12.5 trillion globally by 2024. But how can these pandemics be detected early? 

Filtering and analysis of open-source data by AI & ML tools could stop future epidemics from becoming global pandemics. Traditional public health surveillance relies heavily on statistical techniques, which have several limitations. However, recent years have witnessed tremendous growth of AI-enabled methods, including but not limited to Deep Learning models complimenting statistical approaches. 

AI models for detecting global pandemics 

 Mentioned following are some of the AI-grounded applications developed in different parts of the world that can raise a red flag against emerging pandemics: 

 BlueDot: Developed in Canada, BlueDot helps anticipate outbreaks, mitigate risk and build resilience. They were among the first to identify undiagnosed pneumonia in Wuhan in 2019. It is a proprietary software-as-service designed to track, locate and conceptualize infectious disease spread. It alerts anomalous disease outbreaks that its AI has discovered and the risk it may pose to health care, government, business, and public health clients. It uses NLP and ML to collect data from hundreds of thousands of sources. 

Know more about BlueDot here.

 EPIWATCH: EPIWATCH, developed by Medical Researchers and epidemiologists at the University of New South Wales Sydney, functions using Artificial Intelligence to mine vast open-source data and can identify epidemics in the early days. This model can scan global reports and social media signs for a new illness or unfamiliar disease symptoms in the community. EPIWATCH is an open-source outbreak observatory for early outbreak warning and rapid risk analysis. It provides real-time decision support tools with researched and validated data. It can identify potential outbreak risks earlier than detection based on traditional reporting. The model applies algorithms and Machine Learning to detect unusual spikes in clinical syndromes or diseases. 

Know more about EPIWATCH here.

HealthMap: HealthMap consists of researchers and software developers at Boston’s Children’s Hospital. It is a freely available website and mobile app that delivers real-time intelligence on a broad range of infectious diseases to a diverse audience, including libraries, local health departments, governments and international travelers. The system organizes, integrates, filters, and visualizes online information about emerging diseases in nine languages through constant monitoring.  

Know more about HealthMap here.

ProMED Mail: Nonprofit organization- the International Society for Infectious Diseases, developed this moderated site. Built for detecting infectious diseases, it receives alerts from health professionals about unusual, severe outbreaks and illnesses. PromMed- The Program for Monitoring Emerging Diseases is a moderated system and can function in multiple languages.  

Know more about ProMED here.

Time is a significant component in preventing a pandemic. Even detecting it a few days early can cause a massive difference in the long-term outcome. In a post-pandemic world, there are a lot of ongoing discussions on developing public health big data for surveillance. AI applications in the medical domain will enable enhanced potential in the health surveillance system. 

Sources:

1. Paper published in National Library of Medicine
2. The Week