IOIBD: the International Organization for the Study of Inflammatory Bowel Disease.

In 2019, the World Health Organization reported that one out of every eight individuals worldwide lives with one or more mental disorders (). The COVID-19 pandemic resulted in a significant increase in depression and anxiety cases within a year (). Chronic stress is the principal environmental factor that triggers a decline in mood, leading to depression and anxiety (). Rodent models for stress responses have been extensively used since the early 1980s to simulate these conditions, including the Chronic Unpredictable Mild Stress (CUMS), Electric Shock Stress (ES), and Chronic Restraint Stress (CRS) models (; , ). In CUMS and its derived models, mice or rats are subjected to constant but unpredictable mild stressors, resulting in the development of depression-like or anxiety-like behaviors, mimicking the core symptoms of clinical depression and anxiety, such as anhedonia and acquired helplessness (). CUMS has demonstrated its excellent reliability and validity in the realm of drug discovery (). However, the biological mechanisms underlying depression and anxiety disorders remain poorly understood, and there are no specific biomarkers. As a result, assessment primarily relies on behavioral studies in rodent models. Rodents’ response to stress is typically observed through sequential behaviors, including anhedonia, abnormal weight, and poor coat condition. In order to quantitatively assess the severity of mental disorders and distinguish between anxiety and depression-like behaviors in experimental animals, conventional neurobehavioral studies such as the open field test (OFT), elevated plus maze (EPM) and light/dark box (LDB) test for anxiety (; ). The sucrose preference (SP) test, forced swim test (FST) and tail suspension test (TST) for depression are commonly employed (; ). Nevertheless, it is important to acknowledge that each of these established methodologies has its own set of limitations. For instance, Cheryl D. Conrad has pointed out that the assessment of anxiety through the OFT does not necessarily correlate with anxiety assessed using the EPM (). Additionally, the TST and FST assays only capture a single facet of depression (), while the EPM may induce anxiety in animals during testing, thereby complicating the evaluation process (). Furthermore, the very procedures involved in these experiments impose additional stress on the animals (), which can interfere with the interpretation of data and restrict the translational relevance of the studies. The rise of the artificial intelligence era has significantly increased the popularity of behavioral research, particularly in the analysis of rodent behavior using mathematical models and artificial intelligence algorithms. For example, in 2021, Professor Wang Liping’s team proposed the use of 3D behavior analysis to examine the detailed activities of rodents (). integrated point tracking with posture dynamics through mathematical fitting to analyze rodent behavior. These deep learning methods represent a promising direction for future development; however, they currently face challenges such as high computational costs and limited throughput. We sought to distinguish between depressed and anxious mice in a simple and efficient manner, enabling high-throughput drug screening.