Algorithms’ ability to analyze and interpret large amounts of information in a relatively short time, too big and too complex for humans, may well be the blessing that public health agencies need to improve their response to outbreaks in the age of globalization.
Nevertheless, there are opportunities, volumes of data, human expertise, ethical issues, and planning to be considered when introducing algorithms. Whether analyzing reports, contributing to public health intelligence, or responding to epidemics, algorithms’ involvement in all three areas has the potential to make a substantial contribution. It is important to look not only at their benefits but also at their costs and limitations.
Opportunities
There are opportunities for algorithms in public health surveillance:
- Early outbreak detection : Algorithms can be used to detect outbreaks of specific diseases. For example, analyzing various sources of information, including wastewater, electronic health records (EHRs), and social media, can help identify patterns or trends that may indicate the impending rise in incidence.
- Contributing to public health intelligence (PHI): Algorithms can be used to collate and analyze various types of information reported in multiple sources, such as epidemiological, laboratory, environmental, and open sources, mentioned above, to build PHI about outbreaks.
- Enhancing efficiencies: Algorithms can free up time for surveillance activities by eliminating the need for people to perform surveillance-related tasks. Time spent analyzing information can be redirected to epidemiologists and surveillance officers to allow them to do other things.
- Preparing for and responding to outbreaks: An algorithm could be used to model health scenarios and identify the actions to take in response to each scenario so that public health officials are always prepared. In addition, algorithms could be used to estimate the number of personnel and facilities needed to provide care in the event of an outbreak.
Challenges and limitations
- Nevertheless, algorithms have some limitations and challenges that need to be overcome:
- Their effectiveness is determined by the quality of the data with which they are trained: if the data is erroneous, the results of their work will also be erroneous.
- Biases in algorithms: If an algorithm is trained on data that is not diverse enough, the results it produces will be biased, which could compromise surveillance activities, such as addressing health disparities.
- Limited transparency and explainability: The inner workings of algorithms can sometimes be inaccessible to most people, which can make the algorithms unavailable to ordinary users. This characteristic could also make algorithms challenging to use or interpret because it may not always be clear whether the results are correct.
- Inappropriate use of data and privacy: To be accurate, algorithms need to be trained on vast amounts of data, some of which may contain personal information, which compromises privacy.
- Limited expertise: Algorithms require specialized knowledge to function; however, the number of experts in this area may be limited. In addition, algorithms may be too complex for public health professionals to understand how they work.
- Financial and infrastructural constraints: The process of training, implementing, and maintaining algorithms can be very expensive and time-consuming, especially for low- and middle-income countries.
AI Governance:
Algorithms require ethical and transparent oversight to maximize health gains while minimizing risks to individuals and populations.
Limited role for human expertise: Although most of the above challenges are significant, it is important to note that the benefits of algorithms outweigh the costs. Algorithms are likely to be a critical part of modern public health surveillance. It is important to keep in mind that human expertise will be required despite the capabilities of algorithms. Professions such as epidemiology, surveillance officers, or microbiology will continue to be relevant in the future, as surveillance officers will be needed to operate algorithms. Moreover, human expertise may still be required to investigate potential health threats and respond to them appropriately. Finally, people may be needed to ensure that algorithms function ethically.
Algorithms can revolutionize public health surveillance by enabling the early detection of outbreaks, contributing to public health intelligence, and increasing efficiencies. Nevertheless, it is critical to remember that algorithms should serve as a tool for surveillance rather than the other way around. For algorithms to realize their full potential in public health surveillance, effective oversight and governance frameworks, as well as multidisciplinary collaboration, must be in place. Public health surveillance, which is becoming more data-driven by the day, will benefit greatly from novel technologies as well as competent professionals who can work together to ensure healthy communities.
This is the conversation we are bringing to the GPHCB Global Webinar:
AI for Public Health Surveillance: The Future of Outbreak Detection and Health Intelligence
Join public health professionals, students, researchers, institutions and emerging leaders as we explore how AI and human expertise can work together to detect earlier, decide better and protect communities more effectively.
August 27–29, 2026
Register now: www.gphcb.org/webinar
Reference
1.World Health Organization. Public Health Intelligence .
2.World Health Organization. Epidemic Intelligence from Open Sources (EOS) database.
3.World Health Organization. Ethics and governance of artificial intelligence for health . Geneva: World Health Organization; 2021.
4.World Health Organization. International Health Regulations (2005 ). 3 rd ed. Geneva: World Health Organization;
