A Web-based Software System for Behavior Analysis of Laboratory Animals

Abstract

The analysis of locomotion in laboratory animals plays a crucial role in many scientific research areas. In fact, important information on animals’ behavior and their reaction to a particular stimulus is deduced from a careful analysis of their movements. The techniques commonly adopted to support such analysis have many limitations, which make the related systems particularly ineffective. On the one hand, the human observation and annotation process is strongly observer-dependent and expensive in terms of time and efforts. On the other hand, the use of more sophisticated systems based on video recordings and recognition algorithms is very expensive and complex. In order to face this challenge, this paper presents a tracking solution based on passive Radio Frequency Identification (RFID) technology in Ultra High Frequency (UHF) band, allowing the tracking of laboratory animals with a high accuracy. The overall solution consists of a hybrid system including hardware and software components. In particular, in this paper, the attention is focused on the software component as the hardware has already been described in previous works. The software component is a Web-oriented solution that offers a complete 2D and 3D information tool including reports, dashboards, and tracking graphs. The proposed solution was widely tested using twelve laboratory mice and compared with an automated video-tracking software (i.e., EthoVision) in order to demonstrate its effectiveness and reliability. The obtained results have demonstrated that the proposed solution is able to correctly detect and reconstruct the events occurring in the animals’ cage, and to offer a complete and user-friendly tool to support researchers in behavioral analysis of small laboratory animals.

Keywords