# Integrating explainable AI and One Health: a new frontier in combating infectious diseases

## TL;DR

This paper explores how combining explainable AI with a One Health approach can improve the fight against infectious diseases by making predictions more interpretable and actionable.

## Contribution

The novelty lies in proposing XAI as a new framework within One Health to enhance infectious disease intelligence and decision-making.

## Key findings

- XAI can improve interpretability of AI models used in infectious disease forecasting and surveillance.
- Integrating XAI with One Health can help identify outbreak drivers and optimize resource allocation.
- Challenges include data harmonization, governance, and equitable distribution of benefits.

## Abstract

Infectious diseases (IDs) remain a major threat to global health and societal stability. Because most emerging IDs in humans are zoonotic in origin and shaped by environmental contexts, effective prevention and control call for a One Health approach. Machine learning is widely used for ID modelling and forecasting but often lacks interpretability to explain predictions or guide public health action. Explainable AI (XAI) makes complex models interpretable, enabling attribution of predictions and identification of key outbreak drivers. In this Personal View, we argue that embedding XAI within a One Health framework offers a new organising principle for ID intelligence. We highlight emerging applications in surveillance and forecasting, zoonotic spillover, antimicrobial resistance monitoring and optimisation of resource allocation. We also outline key challenges, including data harmonisation, governance, privacy protection and equitable distribution of risks and benefits. Advancing XAI-enabled One Health systems will require collaboration across sectors and methodological innovation.