Rodent Behavior Recognition
Presenters: Elsbeth van Dam, Oscar Fernandez, Malte Lorbach, Noldus Information Technology BV, Wageningen, Netherlands
Schedule: Thursday 28th August 10:00-10:20. Kleine Veerzaal.
In November 2013, Noldus IT released the new software module RBRM for the Automatic Recognition of Rat Behavior. With this module EthoVisionXT can recognize ten behaviors: eating, drinking, jumping, twitching, resting, grooming, walking, sniffing, and rearing (supported such as against the wall, and unsupported). Validation studies have shown that the Rat Behavior Recognition Module is as accurate as human observation – both can reach an accuracy of about 70%, but the software is more consistent and tireless (van Dam et al., 2013). Human observation is subjective and a human will never make the exact same observation twice. Rat Behavior Recognition is consistent: it scores the same way, every time, all the time. The system needs no on-site training; the only inputs needed are the sizes of the cage and the animal. Furthermore, RBR uses an overhead camera view, which is very practical in lab situations and facilitates high-throughput testing more easily than a side-view setup.
Research at rodent behavior recognition proceeds by extending the system in several directions:
- to different species: mice, who are quicker and more deformable than rats
- to social behaviors in groups or pairs, both by analyzing proximity and direction from location parameters, and by recognizing behaviors like playing, social grooming, social aggression etc.
- by using depth sensors based time of flight or structured light.
- recognition of other specific individual rat behaviors for custom made solutions
During the demo we will present prototypes and research results.
Acknowledgements
This research is partially financed by a grant from Agentschap NL (NeuroBasic-PharmaPhenomics) and the European Commisssion (Marie Curie EID project PhenoRat).
References
Dam, E. van; Harst, J.E. van der; Braak, C.J.F. ter; Tegelenbosch, R.A.J.; Spruijt, B.M.; Noldus, L.P.J.J. (2013). An automated system for the recognition of various specific rat behaviours. Journal of Neuroscience Methods, 218 (2), 214-224.
