MakeUsWell

All of Us

How AI Can Help Detect and Prevent the Transmission of Zoonotic Diseases to Humans

by Mike Critelli, 


Introduction

Zoonotic diseases are infections transmitted from animals to humans. These diseases pose a significant global health threat, as evidenced by past outbreaks such as the Ebola virus, SARS, and bird flu outbreaks. With rapid globalization and increased human-animal interactions affecting disease transmission patterns, innovative solutions to detect and prevent zoonotic diseases are urgently needed. 

Artificial Intelligence (AI) has emerged as a powerful tool that can help monitor, predict, and mitigate the spread of zoonotic infections through advanced data analytics, machine learning, and automation. AI can be leveraged in the early detection, monitoring, and prevention of zoonotic disease transmission.  

AI in Early Detection of Zoonotic Diseases

One critical aspect of controlling zoonotic diseases is early detection. AI plays a pivotal role in analyzing vast amounts of data to identify patterns and predict potential outbreaks before they escalate. Several approaches illustrate how AI contributes to early detection:

Wearable Technology for Health Monitoring

AI-enabled wearable devices monitor the health conditions of livestock and wild animals, detecting abnormal behaviors or symptoms indicative of disease. By analyzing this data, AI can issue early warnings to farmers and health authorities, allowing for timely interventions.

Wastewater monitoring

Artificial intelligence, combined with sensors that monitor wastewater, can significantly improve our ability to detect the presence of pathogens in either animal or human populations. AI is superior because it can detect anomalies at much smaller concentrations, it can do continuous real-time monitoring, and it can detect pathogens more accurately and cost-efficiently than can humans. Individual testing can be more targeted, which reduces the cost of designing, producing, distributing, and selling test kits and analyzing results from a much larger, more diffuse population.

Predictive Analytics and Surveillance Systems

In addition to wearables and wastewater, AI-driven predictive analytics models analyze data from more remote data sources, such as satellite imagery, drones, and social media, to detect anomalies that may indicate the emergence of zoonotic diseases. AI-powered natural language processing tools can also scan news articles, medical journals, and online reports to detect early warning signs of zoonotic disease outbreaks. By leveraging data, AI can identify correlations between environmental changes, animal migration patterns, and disease outbreaks.

Machine Learning in Pathogen Identification

Machine learning algorithms help researchers identify new pathogens by analyzing genetic sequences from various animal species. By comparing these sequences with known pathogens, AI can predict potential spillover events and provide insights into disease evolution.

The Broader Implications of Zoonotic Disease Transmission

There is no question that, because of the much closer geographic proximity between animals and humans, the odds of zoonotic diseases have significantly increased. We may never know the root cause of the SARS-Cov2 pandemic because there are two competing theories, the “wet market” origin theory and the “Wuhan lab leak” theory. The US CIA and Department of Energy now believe that the “Wuhan lab leak” theory is a credible explanation, but we may never know for sure. 

But even if the “wet market” origin theory does not explain the origins of the SARS Cov-2 viruses, wet markets are a high-risk source of zoonotic diseases. We tend to think of “wet markets” as existing only in China or even less developed countries, but wet markets and many other sources of close animal-human interaction are more prevalent than ever.

In areas of increasing population, Americans build homes in areas previously occupied by animals. They also venture into forested areas, nature preserves and wild, open fields in the ordinary course of recreational activities. In doing so, they are multiplying the opportunities for disease transmission.  

Americans own more dogs, which can transmit pathogens present in nature as either viruses, bacteria, or fungi. According to the American Veterinary Medical Association, the American dog population has increased from 52.9 million in 1996 to 89.7 million in 2024, which increases the number of disease carriers in a community.  

This 2023 Report of a joint study by Harvard Law School and New York University finds that the US has a far higher risk profile for the origination and transmission of zoonotic diseases than many of us would expect, because of many other factors as well. New York City alone has 84 such “wet markets.”

https://animal.law.harvard.edu/wp-content/uploads/Animal-Markets-and-Zoonotic-Disease-in-the-United-States.pdf

The Report contains these unsettling findings:

1. “There is no single, unified federal or state authority responsible for the prevention, detection, and regulation of zoonotic disease.” We have a patchwork of laws, regulations, and practices that imperfectly address a potential existential threat of a major pandemic.

2. “More emerging infectious diseases originated in the United States than in any other country in the world during the second half of the 20th century…And it was the United States that was the likely source of the deadliest disease outbreak of recent record…The 1918 Influenza pandemic…”

3. This Report examines 36 separate consumer markets that result in the importation or production of animals of all kinds, including exotic pets, livestock (according to this report, there are 30 livestock animals for every person in the United States) and domesticated animals.

4. “…Despite the serious risks and magnitude of disease exposure, the United States is not prepared to address these threats, many of which are chronically overlooked and under-regulated.”

We lack the human resources to monitor all these risks, but the tools of artificial intelligence can significantly close the gap between the inadequate monitoring occurring today and the more comprehensive monitoring needed. Wearables on animals and other tools for monitoring micro-environments are great use cases for artificial intelligence. AI agents can take the captured data and present a risk profiling in near real time.

What happens after that depends on how the US chooses to assign responsibility for follow-up actions. Public health responsibility is split between the federal government and the states. Given the strong national interest in preventing the transmission of the H5N1 virus from birds to humans, there is a strong case to be made for the federal government to work with the states to create AI-controlled monitoring systems that will detect the presence of pathogens before they spread widely into the human population.

Conclusion

This is a complex problem with multiple points of failure built into it. The United States organized its governance over its 236-year history to delegate most health management to the states. Building a steady-state AI-driven pathogen monitoring system into our public health infrastructure without either federal or state government declaration of emergency powers is a far better way to avoid amplifying the distrust that sprang up as a result of federal and state government responses to the SARS-Cov2 crisis.