An integrated and scalable rodent cage system enabling continuous computer vision-based behavioral analysis and AI-enhanced digital biomarker development DOI Creative Commons

T. L. Robertson,

M. Ellis,

Natalie Bratcher-Petersen

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 26, 2024

1 Abstract Home cage monitoring enables continuous observation of animals in familiar environments. It has large utility preclinical testing, mechanistic studies, animal husbandry, and the general practice Replacement, Reduction, Refinement (3R) principles. Despite its acknowledged utility, broad adoption home not been broadly adopted. This is mainly due to complexity tasks that must be solved have a successful system includes hardware sensor development, data management, machine vision expertise, behavioral support, user training. Here, we describe Digital In Vivo System (DIV Sys), modern end-to-end for video-based rodent monitoring. The DIV Sys consists cloud-based study design, monitoring, display, visualization app App), local acquisition cages (DAX), learning model tracking mice (mHydraNet) optimized speed accuracy, display app, an advanced behavior quantification workbench Data). platform seamlessly manages terabytes video cloud built around enterprise-level security standards. Collaborative tools enable teams across geographical locations work together. As demonstration used analyze over century mouse videos multiple geographic locations. We also characterized 8 strains carried out customized analysis. Together, present scalable research community.

Language: Английский

Visual detection of seizures in mice using supervised machine learning DOI Creative Commons
Gautam Sabnis, Leinani Hession, J. Matthew Mahoney

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: May 31, 2024

1 Abstract Seizures are caused by abnormally synchronous brain activity that can result in changes muscle tone, such as twitching, stiffness, limpness, or rhythmic jerking. These behavioral manifestations clear on visual inspection and the most widely used seizure scoring systems preclinical models, Racine scale rodents, use these patterns semiquantitative intensity scores. However, is time-consuming, low-throughput, partially subjective, there a need for rigorously quantitative approaches scalable. In this study, we supervised machine learning to develop automated classifiers predict severity directly from noninvasive video data. Using PTZ-induced model mice, trained video-only ictal events, combined events an univariate recording session, well time-varying Our results show, first time, overall be quantified overhead of mice standard open field using approaches. enable high-throughput, noninvasive, standardized downstream applications neurogenetics therapeutic discovery.

Language: Английский

Citations

2
The fortunes and misfortunes of social life across the life course: A new era of research from field, laboratory and comparative studies DOI
Alessandro Bartolomucci, Jenny Tung, Kathleen Mullan Harris

et al.

Neuroscience & Biobehavioral Reviews, Journal Year: 2024, Volume and Issue: 162, P. 105655 - 105655

Published: April 5, 2024

Language: Английский

Citations

0
An integrated and scalable rodent cage system enabling continuous computer vision-based behavioral analysis and AI-enhanced digital biomarker development DOI Creative Commons

T. L. Robertson,

M. Ellis,

Natalie Bratcher-Petersen

et al.

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 26, 2024

1 Abstract Home cage monitoring enables continuous observation of animals in familiar environments. It has large utility preclinical testing, mechanistic studies, animal husbandry, and the general practice Replacement, Reduction, Refinement (3R) principles. Despite its acknowledged utility, broad adoption home not been broadly adopted. This is mainly due to complexity tasks that must be solved have a successful system includes hardware sensor development, data management, machine vision expertise, behavioral support, user training. Here, we describe Digital In Vivo System (DIV Sys), modern end-to-end for video-based rodent monitoring. The DIV Sys consists cloud-based study design, monitoring, display, visualization app App), local acquisition cages (DAX), learning model tracking mice (mHydraNet) optimized speed accuracy, display app, an advanced behavior quantification workbench Data). platform seamlessly manages terabytes video cloud built around enterprise-level security standards. Collaborative tools enable teams across geographical locations work together. As demonstration used analyze over century mouse videos multiple geographic locations. We also characterized 8 strains carried out customized analysis. Together, present scalable research community.

Language: Английский

Citations

0