Development of EthoGrid, a YOLO-Based Software Package for Automatic Annotation of Behavior in Betta Fish, and Its Application in Testing the Potential Anxiolytic Function of Essential Oils

Quantification of rapid, complex social behaviors remains a primary challenge in aquatic toxicology, where manual observation often fails to capture high-intensity, nuanced interactions. We developed EthoGrid, a specialized YOLOv11-based software package engineered for the automated quantification of mouth-locking, mirror-biting, and other aggression-related behaviors in Betta splendens. The platform achieved high detection fidelity, with mean average precision at an IoU threshold of 0.50 (mAP50) of 0.805 for the results from the fighting test and 0.977 for results from the mirror-biting test. To demonstrate the tool’s sensitivity, we evaluated the stress-mitigating and anxiolytic effects of lemongrass (Cymbopogon citratus) and melaleuca (Melaleuca alternifolia) essential oils. EthoGrid accurately quantified a significant reduction in mirror-biting time percentage and an increase in wandering and freezing duration in fish treated with 100 ppm lemongrass, indicating a potent sedative effect. Meanwhile, in the fighting tests, behavioral differences across treatments were less pronounced and did not reach statistical significance, reflecting the robust territorial motivation elicited during live encounters. However, Melaleuca treatment produced subtle behavioral modulations in the mirror-biting test. Taken together, by providing a high-throughput pipeline for the specific detection of physical jaw-locking and striking, EthoGrid substantially minimizes subjective inter-observer variability, eliminates manual scoring fatigue, and offers a robust, objective platform for screening the anxiolytic and toxicological properties of compounds in aquatic models.

Authors

Institutions

Publication Details

Journal
AI Biology
Published
2026-11-13
DOI
https://doi.org/10.3390/aibiol1010003
Primary Topic
Water Quality Monitoring Technologies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Development of EthoGrid, a YOLO-Based Software Package for Automatic Annotation of Behavior in Betta Fish, and Its Application in Testing the Potential Anxiolytic Function of Essential Oils

Chung‐Der Hsiao, Kevin Adi Kurnia, Tzong-Rong Ger, Gilbert Audira et al.
2 links
AI Biology
Water Quality Monitoring Technologies
article

Development of EthoGrid, a YOLO-Based Software Package for Automatic Annotation of Behavior in Betta Fish, and Its Application in Testing the Potential Anxiolytic Function of Essential Oils

Chung‐Der Hsiao, Kevin Adi Kurnia, Tzong-Rong Ger, Gilbert Audira, Minh Quan Tran, Mahmood Yousaf
article en

Abstract

Quantification of rapid, complex social behaviors remains a primary challenge in aquatic toxicology, where manual observation often fails to capture high-intensity, nuanced interactions. We developed EthoGrid, a specialized YOLOv11-based software package engineered for the automated quantification of mouth-locking, mirror-biting, and other aggression-related behaviors in Betta splendens. The platform achieved high detection fidelity, with mean average precision at an IoU threshold of 0.50 (mAP50) of 0.805 for the results from the fighting test and 0.977 for results from the mirror-biting test. To demonstrate the tool’s sensitivity, we evaluated the stress-mitigating and anxiolytic effects of lemongrass (Cymbopogon citratus) and melaleuca (Melaleuca alternifolia) essential oils. EthoGrid accurately quantified a significant reduction in mirror-biting time percentage and an increase in wandering and freezing duration in fish treated with 100 ppm lemongrass, indicating a potent sedative effect. Meanwhile, in the fighting tests, behavioral differences across treatments were less pronounced and did not reach statistical significance, reflecting the robust territorial motivation elicited during live encounters. However, Melaleuca treatment produced subtle behavioral modulations in the mirror-biting test. Taken together, by providing a high-throughput pipeline for the specific detection of physical jaw-locking and striking, EthoGrid substantially minimizes subjective inter-observer variability, eliminates manual scoring fatigue, and offers a robust, objective platform for screening the anxiolytic and toxicological properties of compounds in aquatic models.

AI BiologyVol. 1(1)
National Yang Ming Chiao Tung University (TW), Chung Yuan Christian University (TW)
Water Quality Monitoring Technologies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon