Methods of automated analysis of behavioral tests using machine learning and artificial intelligence

Abstract

Behavioral tests in laboratory animals are a vital tool for investigating neurological responses and psychoemotional states. Traditional methods of data processing often involve extensive manual effort, are prone to subjectivity, and exhibit low reproducibility. This review explores contemporary approaches to the automated analysis of behavioral tests in laboratory animals, focusing on the application of machine learning and artificial intelligence. It discusses the benefits and limitations of various techniques, including animal tracking, as well as supervised and unsupervised behavior classification. The review highlights the need for user-friendly, standardized solutions to make automated analysis more accessible to a broader scientific community.