Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 14, 2026

Behavioral And Physiological Analysis In A Zebrafish Model Of Epilepsy

Behavioral And Physiological Analysis In A Zebrafish Model Of Epilepsy

Published on: October 19, 2021

Comparing human annotation and machine learning models for optimizing zebrafish behavioral classification in seizure

Barbara D Fontana¹, Camilla W Pretzel², Mariana L Müller²

¹Laboratory of Experimental Neuropsychobiology, Department of Biochemistry and Molecular Biology, Federal University of Santa Maria, Santa Maria, RS, Brazil; Graduate Program in Biological Sciences, Toxicological Biochemistry, Federal University of Santa Maria, Santa Maria, RS, Brazil.

Journal of Neuroscience Methods

|October 23, 2025

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Therapeutic role of plant bioactive compounds in atrazine-mediated neurotoxicity.

Advances in neurotoxicology·2026

Same author

Behavioral test batteries in zebrafish: methodological variability, experimental design, and implications for neurobehavioral research.

Reviews in the neurosciences·2026

Same author

Opioidergic modulation of stress-induced hyperalgesia in adult zebrafish.

Journal of neural transmission (Vienna, Austria : 1996)·2026

Same author

Para-chlorophenylalanine induces multidimensional affective-like behaviors in adult zebrafish.

Pharmacology, biochemistry, and behavior·2026

Same author

The anoxia escape test as a novel and sensitive protocol to assess despair-like behavior in zebrafish.

Journal of neuroscience methods·2026

Same author

Genetic disruption of leucine rich repeat transmembrane protein 4 like 1 induces a pro-social behavioural phenotype in zebrafish.

Neurobiology of disease·2025

Supervised machine learning (ML) algorithms accurately classify zebrafish behaviors, outperforming human annotators in consistency. However, temporal aggregation methods can obscure critical short-lived behaviors vital for detailed phenotyping.

Area of Science:

Behavioral neuroscience
Machine learning applications
Animal behavior analysis

Background:

Accurate behavioral annotation is crucial but challenging in neuroscience.
Manual scoring is time-consuming, variable, and prone to missing transient behaviors.
Supervised machine learning (ML) offers automated, consistent, and less biased behavior classification.

Purpose of the Study:

To benchmark multiple supervised ML algorithms against expert human annotations for zebrafish seizure-like behaviors.
To evaluate the impact of frame-level analysis, post-processing filters, and block-level temporal aggregation on behavioral classification accuracy.
To compare the reproducibility and scalability of ML-based behavioral phenotyping with traditional manual scoring.

Main Methods:

Keywords:

Human bias Manual labeling Phenotyping Replicability Supervised learning

More Related Videos

Forebrain Electrophysiological Recording in Larval Zebrafish

Forebrain Electrophysiological Recording in Larval Zebrafish

Published on: January 24, 2013

Automated High-throughput Behavioral Analyses in Zebrafish Larvae

Automated High-throughput Behavioral Analyses in Zebrafish Larvae

Published on: July 4, 2013

Related Experiment Videos

Last Updated: Jan 14, 2026

Behavioral And Physiological Analysis In A Zebrafish Model Of Epilepsy

Behavioral And Physiological Analysis In A Zebrafish Model Of Epilepsy

Published on: October 19, 2021

Forebrain Electrophysiological Recording in Larval Zebrafish

Forebrain Electrophysiological Recording in Larval Zebrafish

Published on: January 24, 2013

Automated High-throughput Behavioral Analyses in Zebrafish Larvae

Automated High-throughput Behavioral Analyses in Zebrafish Larvae

Published on: July 4, 2013

Benchmarked five ML algorithms: Random Forest, XGBoost, Support Vector Machine, k-Nearest Neighbors, and Multilayer Perceptron (MLP).

Utilized over 43,000 manually annotated video frames from adult zebrafish by twelve expert raters.

Applied frame-level analysis, behavior-informed filters, and block-level temporal aggregation for evaluation.

Main Results:

ML algorithms, particularly Random Forest, XGBoost, and MLP, achieved high accuracy in classifying zebrafish behaviors.
Annotation variability was highest for ambiguous behaviors, highlighting human annotator challenges.
Post-processing filters improved classification by reducing false positives, while temporal smoothing enhanced overall accuracy but masked short-lived behaviors.

Conclusions:

Supervised ML provides a powerful tool for automated behavioral analysis in zebrafish, offering advantages over manual scoring.
The choice of temporal resolution and classification strategy significantly impacts the reproducibility and interpretability of behavioral phenotyping.
This study provides insights into optimizing ML models for robust and scalable behavioral annotation in neuroscience research.