Dataset Information

Continuous Whole-Body 3D Kinematic Recordings across the Rodent Behavioral Repertoire.

ABSTRACT: In mammalian animal models, high-resolution kinematic tracking is restricted to brief sessions in constrained environments, limiting our ability to probe naturalistic behaviors and their neural underpinnings. To address this, we developed CAPTURE (Continuous Appendicular and Postural Tracking Using Retroreflector Embedding), a behavioral monitoring system that combines motion capture and deep learning to continuously track the 3D kinematics of a rat's head, trunk, and limbs for week-long timescales in freely behaving animals. CAPTURE realizes 10- to 100-fold gains in precision and robustness compared with existing convolutional network approaches to behavioral tracking. We demonstrate CAPTURE's ability to comprehensively profile the kinematics and sequential organization of natural rodent behavior, its variation across individuals, and its perturbation by drugs and disease, including identifying perseverative grooming states in a rat model of fragile X syndrome. CAPTURE significantly expands the range of behaviors and contexts that can be quantitatively investigated, opening the door to a new understanding of natural behavior and its neural basis.

SUBMITTER: Marshall JD

PROVIDER: S-EPMC7864892 | biostudies-literature | 2021 Feb

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Continuous Whole-Body 3D Kinematic Recordings across the Rodent Behavioral Repertoire.

Marshall Jesse D JD Aldarondo Diego E DE Dunn Timothy W TW Wang William L WL Berman Gordon J GJ Ölveczky Bence P BP

Neuron 20201218 3

In mammalian animal models, high-resolution kinematic tracking is restricted to brief sessions in constrained environments, limiting our ability to probe naturalistic behaviors and their neural underpinnings. To address this, we developed CAPTURE (Continuous Appendicular and Postural Tracking Using Retroreflector Embedding), a behavioral monitoring system that combines motion capture and deep learning to continuously track the 3D kinematics of a rat's head, trunk, and limbs for week-long timesca ...[more]

PMID: 33340448

Similar Datasets

Project description:The primary aim of the study was to assess intraday and interday reliability of surface electromyography (sEMG) reflex activity of the pelvic floor muscles during synchronous whole-body vibration (S-WBV) of two intensities (30Hz/2mm; 40Hz/4mm) using band-stop filter and high-pass filter signal processing. The secondary aim of the study was to assess intraday and interday (test-retest) reliability of sEMG obtained from maximal voluntary contraction (MVC) test. We evaluated the intraday reliability of sEMG recordings obtained during sessions 1 and 2 performed on the same day. The sessions consisting of maximal voluntary pelvic floor muscle contraction and synchronous vibration sets with 1-hour rest in-between sessions 1 and 2 in healthy nulliparous women. The next intraday reliability was evaluated between the results of sessions 3 and 4 performed on the same day but followed at an interval of 4 weeks. to include the entire menstrual cycle. The interday reliability was determined based on the results of sessions 1 and 3 using the intraclass correlation coefficient (ICC 3,3). The intraday ICCs for band-stop filtered mean and median sEMG frequency and mean normalized sEMGRMS amplitude of the 30Hz/2mm (ICC = 0.89-0.99) and 40Hz/4mm vibration (ICC = 0.95-0.99) indicated substantial reproducibility. The intraday reliability of high-pass filter at 100-450Hz for these parameters was also substantial (30Hz/2mm ICC of 0.92 to 0.98; 40Hz/4mm ICC of 0.88 to 0.98). The interday reliability (session 1 vs. session 3) of the mean normalized sEMGRMS amplitude for band-stop filtered means of 40 Hz/4mm and 30Hz/2mm vibration recordings was substantial (ICC = 0.82 and 0.93). However, ICCs of the mean and median frequency were indicative of fair reliability (ICC of 0.43 to 0.59). The interday reliability of mean normalized sEMGRMS amplitude for high-pass filter at 100-450Hz was substantial (30Hz/2mm ICC of 0.90; 40Hz/4mm ICC of 0.73) for the 30Hz/2mm S-WBV and moderate (ICC = 0.73) for the 40/4mm S-WBV. The ICCs for mean and median sEMG frequency ICCs indicated slight to fair reproducibility (ICC of 0.16 to 0.56). The intraday reliability of the strongest MVC contraction and average MVC turned out substantial (ICC = 0.91-0.98). The interday reliability coefficients of the strongest MVC contraction and average MVCs were 0.91 and 0.82, respectively. Concluded, the intraday reliability proved satisfactory for all variables; however, the interday comparison showed sufficient ICC levels only for the mean amplitude. We therefore recommend this parameter should be used when analyzing PFM sEMG recorded during vibration. ICCs of the mean and median frequency for both signal processing methods were indicative of insufficient reliability and did not reach the threshold for usefulness. Our study showed similar reliability of PFM sEMG during S-WBV in case of the two filtering methods used.

Project description:Total and regional body composition are strongly correlated with metabolic syndrome and have been estimated non-invasively from 3D optical scans using linear parameterizations of body shape and linear regression models. Prior works produced accurate and precise predictions on many, but not all, body composition targets relative to the reference dual X-Ray absorptiometry (DXA) measurement. Here, we report the effects of replacing linear models with nonlinear parameterization and regression models on the precision and accuracy of body composition estimation in a novel application of deep 3D convolutional graph networks to human body composition modeling. We assembled an ensemble dataset of 4286 topologically standardized 3D optical scans from four different human body shape databases, DFAUST, CAESAR, Shape Up! Adults, and Shape Up! Kids and trained a parameterized shape model using a graph convolutional 3D autoencoder (3DAE) in lieu of linear PCA. We trained a nonlinear Gaussian process regression (GPR) on the 3DAE parameter space to predict body composition via correlations to paired DXA reference measurements from the Shape Up! scan subset. We tested our model on a set of 424 randomly withheld test meshes and compared the effects of nonlinear computation against prior linear models. Nonlinear GPR produced up to 20% reduction in prediction error and up to 30% increase in precision over linear regression for both sexes in 10 tested body composition variables. Deep shape features produced 6-8% reduction in prediction error over linear PCA features for males only and a 4-14% reduction in precision error for both sexes. Our best performing nonlinear model predicting body composition from deep features outperformed prior work using linear methods on all tested body composition prediction metrics in both precision and accuracy. All coefficients of determination (R2) for all predicted variables were above 0.86. We show that GPR is a more precise and accurate method for modeling body composition mappings from body shape features than linear regression. Deep 3D features learned by a graph convolutional autoencoder only improved male body composition accuracy but improved precision in both sexes. Our work achieved lower estimation RMSEs than all previous work on 10 metrics of body composition.

Project description:The study of violin recordings as evidence of interpretation and performance approach has been quite extensive throughout recent decades. Findings such as the limited use of vibrato in early recordings, the relative stabilization of tempo and rhythm characteristic of mid-20th century violin playing, or the effect of 'historically informed performance' (HIP) on 'mainstream' (MS) violinists performing early music repertoire, are fundamental in the identification of prevailing norms of practice and changes of style that have occurred over time. However, compared to the considerable amount of research conducted on 20th century playing, studies focusing on 21st century violin performances are still quite limited, and are mostly based on recordings of J. S. Bach's sonatas and partitas for solo violin. This paper examines contemporary violin recordings made in recent decades (1999-2017). It aims to explore the most current trends in early 21st century violin performance practices and identify the impact of HIP principles on the new generation of players. Unlike most previous research, it focuses on recordings made to 19th century repertoire, thus enlarging the spectrum to include performance analysis of a relatively untouched field and enabling examination of the degree and manner of incorporating 19th century performance attributes (e.g., utilization of rhythmic unevenness, portamento or harmonics) into contemporary praxis. Results suggest an extensive blend of stylistic approaches among currently active violinists, putting into question the relevance of the traditional distinction made between HIP and MS performance styles when it comes to current performance vogue. Incorporating 19th century performance devices while playing 19th century music was traced in varying degrees. Findings are suggested as being part of a general, quasi-postmodern quest for pluralism and elimination of hierarchical classifications, representing an era of 'over-choice' environment.

Dataset Information

Continuous Whole-Body 3D Kinematic Recordings across the Rodent Behavioral Repertoire.

Publications

Continuous Whole-Body 3D Kinematic Recordings across the Rodent Behavioral Repertoire.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets