Developing methods for living systematic review and meta-analyses of complex ethological rodent behaviours in pain research

Pain-associated diseases are the leading cause of disease burden globally, where the emotional state and the daily living activities of an individual can be affected. Despite its impact and the high prevalence, pain is often difficult to treat and therefore there is a great clinical demand for developing effective analgesics to reduce suffering and restore function in patients with either acute or chronic pain. Rodents such as rats and mice are the most commonly used laboratory animals for experimental pain models. However, pain diagnosis in rodents is challenging as these prey species tend to hide signs of pain, suffering or disability in order to reduce the risk of predation. Moreover, the traditional pain endpoints are mostly reflex-withdrawal based behaviours, which are measured by using stimulus-evoked methods, such as von Frey and Hargreaves' tests. These involuntary responses caused by spinal reflexes do not resemble with complex pain-like behaviours, where processing of the nociceptive signal at the cerebral cortex is needed. Furthermore, reflex hypersensitivity (i.e. sensory gain) is very rare clinically, as patients often experience numbness and dysesthesia (i.e. sensory loss) and spontaneous pain, therefore the translational validity of these measure outcomes are questionable. Burrowing and thigmotaxis are ethological behaviours that are highly conserved in most rodents. Since these evolutional-conserved behaviours are not crucial for survival in laboratory setting, the complexity of these voluntary behaviours have been proposed to be comparable to the "activities of daily living" in humans, therefore indicating a higher degree of external validity compared to stimulus-evoked tests. The aims of the projects involve: i) conducting systematic reviews of burrowing and thigmotaxis as pain measure outcomes, ii) explore methods for turning them into "living systematic reviews" using developed tools, iii) explore machine learning of extracted evidence for decision making and methods for analysing thigmotaxis behavioural videos of individual animals.