Abstract:

The mental health management of laboratory animals is a critical factor in ensuring the reliability of scientific research data. However, due to the concealed nature of mental state alterations and the limitations of current assessment methods, it is often overlooked by researchers. Therefore, there is a need to explore objective, simple and practical methods for evaluating mental health. Stress, as a primary factor altering the mental state of animals, can influence experimental results across multiple research fields through the neuro-endocrine-immune network. This paper first elucidates the necessity of mental health management in laboratory animals from three perspectives: the factors contributing to stress, the neural mechanisms of stress, and the research areas affected by stress. It highlights that excluding mentally unhealthy animals before experiments can enhance the efficiency and reproducibility of biomedical studies. Second, this paper briefly summarizes methods for assessing the health of laboratory animals, pointing out that approaches such as behavioral studies and metabolomics have limitations in evaluating stress responses. Existing methods struggle to meet the demand for simple, objective, and non-invasive assessments of chronic stress in animal management. Therefore, this paper focuses on hair as a novel indicator for assessing stress. It systematically elaborates on its theoretical basis and evaluation method advancements from four aspects: hair traits, corticosterone, proteins, and artificial intelligence (AI). It innovatively proposes and demonstrates a technical pathway combining hair trait analysis with AI-based image analysis, offering a new solution for non-invasive, objective stress assessment and providing theoretical support for improving the health management system of laboratory animals.

Key words: Laboratory animals, Mental health management, Stress, Hair, Artificial intelligence