Machine learning for environmental toxicology: a call for integration and innovation

Recent advances in computing power have enabled the application of machine learning (ML) across all areas of science. A step change from a data-rich landscape to one where new hypotheses, relationships, and knowledge is emerging as a result. While ML is related to artificial intelligence (AI), they are not the same. ML is a branch of AI involving the application of statistical algorithms to enable a system to learn. Learning can involve data interpretation, identification of patterns and decision making. However, application and acceptance of ML within environmental toxicology, and more specifically for our viewpoint, environmental risk assessment (ERA), remains low. ML is an example of a disruptive research technology, which is urgently needed to cope with the complexity and scale of work required.