An Arena-Aware Motion-Based Deep Learning Framework for Automated Rodent Behavior Analysis in the Open Field Test

DOI

10.2991/978-94-6239-693-7_82How to use a DOI?

Keywords

Open Field Test; Rodent Behavior Analysis; Deep Learning; Motion-Based Features; LSTM; YOLOv8; Computer Vision

Abstract

Automated analysis of rodent behavior in the Open Field Test (OFT) plays a central role in objective and scalable behavioral assessment within neuroscience research. This paper presents a fully automated, video-based framework that performs arena-aware localization, motion feature extraction, and temporal behavior classification directly from raw top-view video recordings. A YOLOv8-based object detection model localizes both the experimental arena and the animal in each frame, enabling arena-relative spatial normalization of motion trajectories. Compact three-dimensional motion feature vectors comprising normalized position and instantaneous speed are grouped into fixed-length temporal sequences and modeled using a Long Short-Term Memory (LSTM) network, yielding frame-aligned behavioral state predictions with associated confidence scores. The proposed pipeline generates ethograms, behavior transition matrices, bout duration statistics, and annotated videos without any reliance on pose estimation or manual annotations. Experimental results demonstrate temporally coherent behavior predictions and stable confidence distributions concentrated above 0.80, confirming the robustness and reproducibility of the approach for automated behavioral analysis across six ethological categories.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Ronak Ashvinbhai Sheladiya
AU  - Vishvajit Bakarola
PY  - 2026
DA  - 2026/06/16
TI  - An Arena-Aware Motion-Based Deep Learning Framework for Automated Rodent Behavior Analysis in the Open Field Test
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 836
EP  - 846
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_82
DO  - 10.2991/978-94-6239-693-7_82
ID  - Sheladiya2026
ER  -