Artificial Intelligence for Early Disease Detection: Bridging Human, Animal, and Environmental Health

Artificial Intelligence (AI) refers to computer systems that can learn from data, recognize patterns, and make predictions with minimal human guidance. Once seen as futuristic, AI is now part of our daily lives—shaping how we search for information, shop online, and even drive cars. In medicine, it is rapidly transforming how we detect and treat disease. By analyzing large and complex datasets, AI can uncover hidden signals and risk factors that traditional methods might miss.

Why is this important? Because early detection often makes the difference between life and death, or between a simple treatment and a lifelong burden of disease. In human health, for instance, AI tools are being used to predict and diagnose conditions such as cancer, preterm birth, and inflammatory bowel disease and many more diseases. By identifying risks earlier, doctors can intervene sooner, reduce complications, and give patients better outcomes.

The same technology is also opening new opportunities in veterinary and agricultural science. AI systems can detect mastitis in dairy cattle, identify cancer biomarkers in dogs, monitor herd health to improve animal welfare and farm productivity, and even recognize early signs of plant diseases in crops such as wheat or potatoes. Detecting these problems sooner helps farmers reduce pesticide use, protect yields, and contribute to more sustainable food production. These advances not only protect animals and plants but also strengthen food security and reduce environmental impact. This reflects the “One Health” perspective—the understanding that the health of humans, animals, and the environment are closely connected.

In my lecture, I will demonstrate how AI methods and statistical modeling can be trained and tested to predict important health outcomes, reduce costs, and support sustainable practices. One recent example comes from veterinary oncology (Sharif, H., Arabi Belaghi, R., Jagarlamudi, K. K., Saellström, S., Wang, L., Rönnberg, H., & Eriksson, S. (2025). A novel cross-validated machine learning based Alertix-Cancer Risk Index for early detection of canine malignancies. Frontiers in Veterinary Science, 12, 1570106.), where the Alertix-Cancer Risk Index (Alertix-CRI) has shown remarkable accuracy in distinguishing dogs with malignant tumors from healthy ones through a simple, non-invasive blood test. 

At the same time, it is important to recognize that AI is not a magic solution. There are real challenges: ensuring that models are transparent and fair, addressing ethical concerns about data use, and making sure that new tools are affordable and accessible. These questions are central if AI is to be trusted and widely adopted in clinical, veterinary, and agricultural practice.

Through examples from my own research and international collaborations, I will highlight some of opportunities and the challenges of using AI for early disease detection. My broader vision is to integrate advanced data-driven approaches into a sustainable framework for health—one that benefits not just patients and animals, but also plants, society, and the environment.