Answers are grounded in this paper and cite their sources.
Background: The continuous and automated monitoring of animal activity with sensors can help to refine and improve the quality of animal experiments. The aim of this study was to develop a system for real-time automated multi-species monitoring of individual animal activity in research trials ("RAMSMART"), to test this system in terms of battery life and data completeness, and to demonstrate its added value. Methods: We used lightweight commercially available accelerometers with edge computing and uploaded self-written firmware to these accelerometers to calculate the vectorial dynamic body acceleration (VeDBA). The accelerometer is attached, for example, to an animal's ear tag. It then collects raw acceleration data (at a user-defined sampling frequency), calculates the mean VeDBA across a certain period (based on a user-defined updating interval) and creates a data packet with the VeDBA, which is then broadcasted (at a user-defined broadcasting frequency) through Bluetooth Low Energy to a nearby receiving device, e.g., a laptop. On this receiving device, data are stored in a SQLite database and regularly copied to a shared drive to enable (near) real-time visualization of results from outside animal facilities. Results: Battery life depended on the sampling frequency, the updating interval and the broadcasting frequency. For example, when the sampling frequency was kept constant at 12.5 Hz, and the broadcasting frequency at 4 Hz, an updating interval of 1 second or 1 minute resulted in a battery life of 15.9 days (± 1.0 days) or 47.7 days (± 1.9 days), respectively. To avoid missing data, a sufficiently high broadcasting frequency was needed and the distance between sensor and receiver should be limited (preferably to < 20 m), or a multiple receiver set-up could be used. We showcased RAMSMART in three infectious disease trials in pigs, sheep and calves, connecting the VeDBA with other clinical symptoms, and demonstrating the added value of this system for early detection of deviations in activity. Conclusion: RAMSMART provides a low-invasive, affordable, re-usable, generalizable and plug-and-play approach for routine and long-term real-time automated activity monitoring based on acceleration data of (particularly livestock) animals in research trials. The source code and documentation of RAMSMART are openly available.
No takes yet. Share an insight, caveat, or question.
Doekes et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: