A new study from the Kamikouchi Group has been published in Science Advances. In this study, we developed an AI‑based behavioral analysis tool called “YORU.”
To understand the mechanisms underlying animal behavior, it is essential to individually analyze and quantify a wide range of behavioral patterns. YORU applies an object‑detection algorithm, a type of machine‑learning method, enabling accurate and rapid analysis of behaviors that were difficult to quantify with conventional tools, such as mating and grooming, which involve complex movements. It also allows precise analysis of social behaviors in which multiple individuals interact simultaneously. Furthermore, by combining real‑time behavioral analysis with a projection‑optical system, we successfully targeted individuals exhibiting specific behaviors and manipulated their neural activity.
Because YORU is designed to perform AI‑based behavioral analysis without requiring any programming, it can be seamlessly integrated into existing experimental setups via microcontrollers such as Arduino. We have released YORU as open‑source software under the AGPL‑3.0 license, allowing users to freely modify and utilize it. This tool enables efficient investigation of the principles of animal behavior, which have traditionally been difficult to analyze or quantify scientifically.
The outcomes of this study are expected to contribute not only to neuroscience and ethology but also to a wide range of research fields related to “behavior,” including ecology and sports science.
Congratulations!
Press Release : https://www.nagoya-u.ac.jp/researchinfo/result/2026/02/aiyoru.html
Science Advances : https://www.science.org/doi/10.1126/sciadv.adw2109
