Machine learning and artificial intelligence in neuroscience: A primer for researchers DOI Creative Commons

Fakhirah Badrulhisham,

Esther Pogatzki‐Zahn, Daniel Segelcke

et al.

Brain Behavior and Immunity, Journal Year: 2023, Volume and Issue: 115, P. 470 - 479

Published: Nov. 14, 2023

Artificial intelligence (AI) is often used to describe the automation of complex tasks that we would attribute to. Machine learning (ML) commonly understood as a set methods develop an AI. Both have seen recent boom in usage, both scientific and commercial fields. For community, ML can solve bottle necks created by complex, multi-dimensional data generated, for example, functional brain imaging or *omics approaches. here identify patterns could not been found using traditional statistic However, comes with serious limitations need be kept mind: their tendency optimise solutions input means it crucial importance externally validate any findings before considering them more than hypothesis. Their black-box nature implies decisions usually cannot understood, which renders use medical decision making problematic lead ethical issues. Here, present introduction curious field ML/AI. We explain principles well methodological advancements discuss risks what see future directions field. Finally, show practical examples neuroscience illustrate ML.

Language: Английский

SLEAP: A deep learning system for multi-animal pose tracking DOI Creative Commons
Talmo Pereira, Nathaniel Tabris, Arie Matsliah

et al.

Nature Methods, Journal Year: 2022, Volume and Issue: 19(4), P. 486 - 495

Published: April 1, 2022

The desire to understand how the brain generates and patterns behavior has driven rapid methodological innovation in tools quantify natural animal behavior. While advances deep learning computer vision have enabled markerless pose estimation individual animals, extending these multiple animals presents unique challenges for studies of social behaviors or their environments. Here we present Social LEAP Estimates Animal Poses (SLEAP), a machine system multi-animal tracking. This enables versatile workflows data labeling, model training inference on previously unseen data. SLEAP features an accessible graphical user interface, standardized model, reproducible configuration system, over 30 architectures, two approaches part grouping identity We applied seven datasets across flies, bees, mice gerbils systematically evaluate each approach architecture, compare it with other existing approaches. achieves greater accuracy speeds more than 800 frames per second, latencies less 3.5 ms at full 1,024 × image resolution. makes usable real-time applications, which demonstrate by controlling one basis tracking detection interactions another animal.

Language: Английский

Citations

447
Quantifying behavior to understand the brain DOI
Talmo Pereira, Joshua W. Shaevitz, Mala Murthy

et al.

Nature Neuroscience, Journal Year: 2020, Volume and Issue: 23(12), P. 1537 - 1549

Published: Nov. 9, 2020

Language: Английский

Citations

256
Deep learning-based behavioral analysis reaches human accuracy and is capable of outperforming commercial solutions DOI Creative Commons
Oliver Sturman, Lukas von Ziegler,

Christa Schläppi

et al.

Neuropsychopharmacology, Journal Year: 2020, Volume and Issue: 45(11), P. 1942 - 1952

Published: July 25, 2020

Abstract To study brain function, preclinical research heavily relies on animal monitoring and the subsequent analyses of behavior. Commercial platforms have enabled semi high-throughput behavioral by automating tracking, yet they poorly recognize ethologically relevant behaviors lack flexibility to be employed in variable testing environments. Critical advances based deep-learning machine vision over last couple years now enable markerless tracking individual body parts freely moving rodents with high precision. Here, we compare performance commercially available (EthoVision XT14, Noldus; TSE Multi-Conditioning System, Systems) cross-verified human annotation. We provide a set videos—carefully annotated several raters—of three widely used tests (open field test, elevated plus maze, forced swim test). Using these data, then deployed pose estimation software DeepLabCut extract skeletal mouse representations. simple post-analyses, were able track animals their representation range classic at similar or greater accuracy than commercial systems. developed supervised learning classifiers that integrate manual annotations. This new combined approach allows us score humans, current gold standard, while outperforming solutions. Finally, show resulting eliminates variation both within between annotators. In summary, our helps improve quality systems fraction cost.

Language: Английский

Citations

171
DeepEthogram, a machine learning pipeline for supervised behavior classification from raw pixels DOI Creative Commons
James P. Bohnslav, Nivanthika K. Wimalasena, Kelsey J. Clausing

et al.

eLife, Journal Year: 2021, Volume and Issue: 10

Published: Sept. 2, 2021

Videos of animal behavior are used to quantify researcher-defined behaviors interest study neural function, gene mutations, and pharmacological therapies. Behaviors often scored manually, which is time-consuming, limited few behaviors, variable across researchers. We created DeepEthogram: software that uses supervised machine learning convert raw video pixels into an ethogram, the present in each frame. DeepEthogram designed be general-purpose applicable species, video-recording hardware. It convolutional networks compute motion, extract features from motion images, classify behaviors. classified with above 90% accuracy on single frames videos mice flies, matching expert-level human performance. accurately predicts rare requires little training data, generalizes subjects. A graphical interface allows beginning-to-end analysis without end-user programming. DeepEthogram's rapid, automatic, reproducible labeling may accelerate enhance analysis. Code available at: https://github.com/jbohnslav/deepethogram.

Language: Английский

Citations

131
Innovations and advances in modelling and measuring pain in animals DOI
Katelyn E. Sadler, Jeffrey S. Mogil, Cheryl L. Stucky

et al.

Nature reviews. Neuroscience, Journal Year: 2021, Volume and Issue: 23(2), P. 70 - 85

Published: Nov. 26, 2021

Language: Английский

Citations

118
Neural control of affiliative touch in prosocial interaction DOI
Ye Wu,

James Dang,

Lyle Kingsbury

et al.

Nature, Journal Year: 2021, Volume and Issue: 599(7884), P. 262 - 267

Published: Oct. 13, 2021

Language: Английский

Citations

104
Identifying behavioral structure from deep variational embeddings of animal motion DOI Creative Commons
Kevin Luxem, Petra Mocellin,

Falko Fuhrmann

et al.

Communications Biology, Journal Year: 2022, Volume and Issue: 5(1)

Published: Nov. 18, 2022

Abstract Quantification and detection of the hierarchical organization behavior is a major challenge in neuroscience. Recent advances markerless pose estimation enable visualization high-dimensional spatiotemporal behavioral dynamics animal motion. However, robust reliable technical approaches are needed to uncover underlying structure these data segment into discrete hierarchically organized motifs. Here, we present an unsupervised probabilistic deep learning framework that identifies from variational embeddings motion (VAME). By using mouse model beta amyloidosis as use case, show VAME not only motifs, but also captures representation motif’s usage. The approach allows for grouping motifs communities differences community-specific motif usage individual cohorts were undetectable by human visual observation. Thus, segmentation applicable wide range experimental setups, models conditions without requiring supervised or a-priori interference.

Language: Английский

Citations

101
Deep-learning-based identification, tracking, pose estimation and behaviour classification of interacting primates and mice in complex environments DOI
Markus Marks,

Jin Qiuhan,

Oliver Sturman

et al.

Nature Machine Intelligence, Journal Year: 2022, Volume and Issue: 4(4), P. 331 - 340

Published: April 21, 2022

Language: Английский

Citations

83
Open-source tools for behavioral video analysis: Setup, methods, and best practices DOI Creative Commons
Kevin Luxem, Jennifer J. Sun,

Sean P Bradley

et al.

eLife, Journal Year: 2023, Volume and Issue: 12

Published: March 22, 2023

Recently developed methods for video analysis, especially models pose estimation and behavior classification, are transforming behavioral quantification to be more precise, scalable, reproducible in fields such as neuroscience ethology. These tools overcome long-standing limitations of manual scoring frames traditional ‘center mass’ tracking algorithms enable analysis at scale. The expansion open-source acquisition has led new experimental approaches understand behavior. Here, we review currently available discuss how set up these labs recording. We also best practices developing using methods, including community-wide standards critical needs the open sharing datasets code, widespread comparisons better documentation users. encourage broader adoption continued development tools, which have tremendous potential accelerating scientific progress understanding brain

Language: Английский

Citations

51
A-SOiD, an active-learning platform for expert-guided, data-efficient discovery of behavior DOI
Jens F. Tillmann, Alexander Hsu, Martin K. Schwarz

et al.

Nature Methods, Journal Year: 2024, Volume and Issue: 21(4), P. 703 - 711

Published: Feb. 21, 2024

Language: Английский

Citations

16