Stride-level analysis of mouse open field behavior using deep-learning-based pose estimation

Sheppard, Keith; Gardin, Justin; Sabnis, Gautam; Peer, Asaf; Darrell, Megan; Deats, Sean P.; Geuther, Brian Q.; Lutz, Cathleen; Kumar, Vivek

doi:10.1016/j.celrep.2021.110231

Cited by 40 publications

(49 citation statements)

References 95 publications

(133 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the slight bimodality of our data, we tested for the presence of Simpson’s Paradox [34] and did not see any evidence (Figure S4). The open field video was processed by a tracking network and a pose estimation network, to produce a track, an ellipse-fit, and a 12-point pose of the mouse for each frame [23, 25]. These frame-by-frame measurements were used to calculate a variety of per-video features, including traditional open field measures of anxiety and hyperactivity [23], grooming [24], gait and posture measures [25], and engineered features.…”

Section: Resultsmentioning

confidence: 99%

“…The frame-by-frame segmentation, ellipse fit, and 12-point pose coordinates were used to extract pervideo features [23][24][25]. All extracted features with explanation and source of the measurements can be found in Table S2.…”

Section: Feature Extractionmentioning

confidence: 99%

“…The 12-point 2D pose estimation was produced using a deep convolutional neural network trained as detailed in [25]. The points captured are nose, left ear, right ear, base of neck, left forepaw, right forepaw, mid spine, left rear paw, right rear paw, base of tail, mid tail and tip of tail.…”

Section: Pose Estimation and Gaitmentioning

confidence: 99%

“…The gait metrics were produced as detailed in Shephard et al (2020) [25]. Briefly, the stride cycles were defined by starting and ending with the left hind paw strike, tracked by the pose estimation.…”

Section: Pose Estimation and Gaitmentioning

confidence: 99%

“…We along with a number of other groups have applied these new methods to animal behavior analysis. Our group has developed methods for image segmentation and tracking in complex environments [23], action detection [24] and pose-based gait and whole body coordination measurements in the open field [25]. These and other highly sensitive methods have advanced animal behavior extraction [15,[26][27][28].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A machine vision based frailty index for mice

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Chronological aging is uniform, but biological aging is heterogeneous. Clinically, this heterogeneity manifests itself in health status and mortality, and it distinguishes healthy from unhealthy aging. Clinical frailty indexes (FIs) serve as an important tool in gerontology to capture health status. FIs have been adapted for use in mice and are an effective predictor of mortality risk. To accelerate our understanding of biological aging, high-throughput approaches to pre-clinical studies are necessary. Currently, however, mouse frailty indexing is manual and relies on trained scorers, which imposes limits on scalability and reliability. Here, we introduce a machine learning based visual frailty index (vFI) for mice that operates on video data from an open field assay. We generate a large mouse FI dataset of both male and female mice. From video data on these same mice, we use neural networks to extract morphometric, gait, and other behavioral features that correlate with manual FI score and age. We use these features to train a regression model that accurately predicts frailty within 0.04 ± 0.002 (3.7% ± 0.01%) of the pre-normalized FI score in terms of mean absolute error. We show that features of biological aging are encoded in open-field video data and can be used to construct a vFI that can complement or replace current manual FI methods. We use the vFI data to examine sex-specific aspects of aging in mice. This vFI provides increased accuracy, reproducibility, and scalability, that will enable large scale mechanistic and interventional studies of aging in mice.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Feature Extractionmentioning

confidence: 99%

Section: Pose Estimation and Gaitmentioning

confidence: 99%

Section: Pose Estimation and Gaitmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A machine vision based frailty index for mice

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Of Mice and Pose: 2D Mouse Pose Estimation from Unlabelled Data and Synthetic Prior

Sosa,

Perry,

Alty

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Mapping the Behavioral Signatures of Shank3b Mice in Both Sexes

Liu,

Ye,

et al. 2024

Neurosci. Bull.

View full text Add to dashboard Cite

Autism spectrum disorders (ASD) are characterized by social and repetitive abnormalities. Although the ASD mouse model with Shank3b mutations is widely used in ASD research, the behavioral phenotype of this model has not been fully elucidated. Here, a 3D-motion capture system and linear discriminant analysis were used to comprehensively record and analyze the behavioral patterns of male and female Shank3b mutant mice. It was found that both sexes replicated the core and accompanied symptoms of ASD, with significant sex differences. Further, Shank3b heterozygous knockout mice exhibited distinct autistic behaviors, that were significantly different from those those observed in the wild type and homozygous knockout groups. Our findings provide evidence for the inclusion of both sexes and experimental approaches to efficiently characterize heterozygous transgenic models, which are more clinically relevant in autistic studies.

show abstract

Stride-level analysis of mouse open field behavior using deep-learning-based pose estimation

Cited by 40 publications

References 95 publications

A machine vision based frailty index for mice

A machine vision based frailty index for mice

Of Mice and Pose: 2D Mouse Pose Estimation from Unlabelled Data and Synthetic Prior

Mapping the Behavioral Signatures of Shank3b Mice in Both Sexes

Contact Info

Product

Resources

About