LabGym

LabGym quantifies animal behavior by representing motion as pattern images and animation and analyzing them with customizable deep neural networks to identify behaviors and compute quantitative measurements.


Key Features:

  • Pattern Image and Animation Representation: Represents animal motion using pattern images to capture motion patterns and animation to preserve spatiotemporal details.
  • Customizable Deep Neural Networks: Applies customizable deep neural networks that jointly assess pattern images and animation for behavior identification.
  • Quantitative Measurements: Calculates quantitative measurements for each identified behavior.
  • Multi-Animal Experiment Applicability: Processes experiments involving multiple animals simultaneously.
  • Visualization Capabilities: Produces visualizations of behavioral datasets.

Scientific Applications:

  • Behavioral Research: Captures subtle changes in behavior across diverse animal species for detailed analysis.
  • Holistic Assessment: Integrates pattern images and animation to enable comprehensive behavioral assessment beyond pose-based methods.

Methodology:

Represents animal motion as pattern images and animation and analyzes these representations with customizable deep neural networks to identify behaviors and compute quantitative measurements.

Topics

Details

Programming Languages:
Python
Added:
2/20/2024
Last Updated:
11/24/2024

Operations

Publications

Hu Y, Ferrario CR, Maitland AD, Ionides RB, Ghimire A, Watson B, Iwasaki K, White H, Xi Y, Zhou J, Ye B. LabGym: Quantification of user-defined animal behaviors using learning-based holistic assessment. Cell Reports Methods. 2023;3(3):100415. doi:10.1016/j.crmeth.2023.100415. PMID:37056376. PMCID:PMC10088092.

PMID: 37056376
Funding: - National Institutes of Health: P40OD018537, R01DA044204, R01DK106188, R01DK115526, R01DK130246, R01EB028159, R01NS104299

Documentation