Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

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

Ovalbumin-peanut protein isolate emulsion gels for enhanced co-delivery of curcumin and quercetin.

International journal of biological macromolecules·2026
Same author

Genome-wide analysis of the ALDH superfamily in <i>Castanea mollissima</i> highlights roles in abiotic stress responses.

Frontiers in plant science·2026
Same author

Bile acid chemosensation in mammals supports species and gut microbiome evaluation.

bioRxiv : the preprint server for biology·2026
Same author

Ginsenoside Rh2 attenuates pyroptosis by inhibiting the PAFR/NF-κB signaling pathway in an osteoarthritis cell model.

Protoplasma·2026
Same author

NH<sub>4</sub> <sup>+</sup>-mediated interfacial chemistry for collaborative dual-pathway high-mass-loading energy storage.

Chemical science·2026
Same author

HCV infection induces dysregulation of glucose-stimulated insulin secretion via the TLR3/TRIF/NF-κB-iNOS-NO axis: Implications for prediabetes in HCV patients.

Cellular and molecular life sciences : CMLS·2026

Related Experiment Video

Updated: Aug 16, 2025

Experimental Assessment of Mouse Sociability Using an Automated Image Processing Approach
08:24

Experimental Assessment of Mouse Sociability Using an Automated Image Processing Approach

Published on: May 15, 2016

8.6K

A Layered, Hybrid Machine Learning Analytic Workflow for Mouse Risk Assessment Behavior.

Jinxin Wang1, Paniz Karbasi2, Liqiang Wang2

  • 1Department of Neuroscience, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642 jinxin_wang@urmc.rochester.edu julian_meeks@urmc.rochester.edu.

Eneuro
|December 23, 2022
PubMed
Summary
This summary is machine-generated.

A novel hybrid machine learning (ML) model improves animal behavior quantification. This layered approach combines supervised and unsupervised ML, enhancing interpretability and reliability for neuroscience research.

Keywords:
hidden Markov modelmachine learningquantification of behaviorrandom forestrisk assessment behavior

More Related Videos

A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents
08:38

A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents

Published on: November 21, 2019

7.7K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K

Related Experiment Videos

Last Updated: Aug 16, 2025

Experimental Assessment of Mouse Sociability Using an Automated Image Processing Approach
08:24

Experimental Assessment of Mouse Sociability Using an Automated Image Processing Approach

Published on: May 15, 2016

8.6K
A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents
08:38

A System for Tracking the Dynamics of Social Preference Behavior in Small Rodents

Published on: November 21, 2019

7.7K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K

Area of Science:

  • Neuroscience
  • Machine Learning
  • Animal Behavior

Background:

  • Accurate animal behavior quantification is crucial for understanding the brain.
  • Existing supervised (e.g., random forest) and unsupervised (e.g., hidden Markov model) ML models have limitations in temporal information and interpretability.
  • Hybrid ML models combining multiple algorithms offer a promising solution.

Purpose of the Study:

  • To develop and validate a layered, hybrid machine learning workflow for enhanced animal behavior quantification.
  • To improve the interpretability and reliability of ML-based behavioral analysis.
  • To investigate mouse risk assessment behavior using the developed hybrid model.

Main Methods:

  • Utilized DeepLabCut for estimating mouse body part positions.
  • Developed a layered hybrid model integrating random forest (RF) and hidden Markov model (HMM) components, including reHMM and reHMM+ architectures.
  • Combined positional features and model predictions as inputs for subsequent HMM layers.

Main Results:

  • The reHMM layer demonstrated improved interpretability compared to initial HMM outputs.
  • The reHMM+ layered hybrid model successfully identified distinctive, interpretable temporal behavioral patterns.
  • Unique risk assessment behaviors in mice exposed to trimethylthiazoline and snake feces odor were identified and separated from neutral stimuli.

Conclusions:

  • The proposed layered, hybrid ML workflow offers a balanced approach to enhance the depth and reliability of ML classifiers.
  • This method is effective for chemosensory and other complex behavioral contexts.
  • The findings advance the field of automated behavioral analysis in neuroscience.