Next Article in Journal
Impact of Milk Thistle (Silybum marianum [L.] Gaertn.) Seeds in Broiler Chicken Diets on Rearing Results, Carcass Composition, and Meat Quality
Next Article in Special Issue
Objective Video-Based Assessment of ADHD-Like Canine Behavior Using Machine Learning
Previous Article in Journal
Induction of Serum Amyloid A3 in Mouse Mammary Epithelial Cells Stimulated with Lipopolysaccharide and Lipoteichoic Acid
Previous Article in Special Issue
Exploring How White-Faced Sakis Control Digital Visual Enrichment Systems
Article

Deep Learning Classification of Canine Behavior Using a Single Collar-Mounted Accelerometer: Real-World Validation

1
Pet Insight Project, Kinship, San Francisco, CA 94103, USA
2
WALTHAM Petcare Science Institute, Melton Mowbray, Leicestershire LE14 4RT, UK
*
Author to whom correspondence should be addressed.
Academic Editor: Lynette A. Hart
Animals 2021, 11(6), 1549; https://doi.org/10.3390/ani11061549
Received: 8 April 2021 / Revised: 14 May 2021 / Accepted: 19 May 2021 / Published: 25 May 2021
(This article belongs to the Special Issue Animal-Centered Computing)
Collar-mounted activity monitors using battery-powered accelerometers can continuously and accurately analyze specific canine behaviors and activity levels. These include normal behaviors and those that are indicators of disease conditions such as scratching, inappetence, excessive weight, or osteoarthritis. Algorithms used to analyze activity data are validated by video recordings of specific canine behaviors, which were used to label accelerometer data. The study described here was noteworthy for the large volume of data collected from more than 2500 dogs in clinical and real-world home settings. The accelerometer data were analyzed by a machine learning methodology, whereby algorithms were continually updated as additional data were acquired. The study determined that algorithms from the accelerometer data detected eating and drinking behaviors with a high degree of accuracy. Accurate detection of other behaviors such as licking, petting, rubbing, scratching, and sniffing was also demonstrated. The study confirmed that activity monitors using validated algorithms can accurately detect important health-related canine behaviors via a collar-mounted accelerometer. The validated algorithms have widespread practical benefits when used in commercially available canine activity monitors.
Collar-mounted canine activity monitors can use accelerometer data to estimate dog activity levels, step counts, and distance traveled. With recent advances in machine learning and embedded computing, much more nuanced and accurate behavior classification has become possible, giving these affordable consumer devices the potential to improve the efficiency and effectiveness of pet healthcare. Here, we describe a novel deep learning algorithm that classifies dog behavior at sub-second resolution using commercial pet activity monitors. We built machine learning training databases from more than 5000 videos of more than 2500 dogs and ran the algorithms in production on more than 11 million days of device data. We then surveyed project participants representing 10,550 dogs, which provided 163,110 event responses to validate real-world detection of eating and drinking behavior. The resultant algorithm displayed a sensitivity and specificity for detecting drinking behavior (0.949 and 0.999, respectively) and eating behavior (0.988, 0.983). We also demonstrated detection of licking (0.772, 0.990), petting (0.305, 0.991), rubbing (0.729, 0.996), scratching (0.870, 0.997), and sniffing (0.610, 0.968). We show that the devices’ position on the collar had no measurable impact on performance. In production, users reported a true positive rate of 95.3% for eating (among 1514 users), and of 94.9% for drinking (among 1491 users). The study demonstrates the accurate detection of important health-related canine behaviors using a collar-mounted accelerometer. We trained and validated our algorithms on a large and realistic training dataset, and we assessed and confirmed accuracy in production via user validation. View Full-Text
Keywords: canine; accelerometer; deep learning; behavior; activity monitor canine; accelerometer; deep learning; behavior; activity monitor
Show Figures

Figure 1

  • Supplementary File 1:

    ZIP-Document (ZIP, 295 KiB)

  • Externally hosted supplementary file 1
    Doi: 10.5281/zenodo.4836665
    Link: https://zenodo.org/record/4836665#.YLOHjaFRVPY
    Description: Supplementary Materials Figure S1: Guide to interpreting supplementary videos. The videos listed below follow the template in this frame capture. The source video file is played in an inset (a), and the synchronized accelerometer data is scrolled through the plot in (b). The head motion estimate produced by the algorithm (accounting for device position and orientation on the collar) is visualized in (c). The algorithm estimates the probability that various behaviors (d) and postures (e) are occurring, based solely on the accelerometer data (the video data is not seen or otherwise used by the algorithm). A small visual display of the most likely predicted class is superimposed on the inset (f). Table S1: Index of supplementary videos Reason Included Video Examples of eating- and drinking-related behaviors, including licking the empty bowl. S1, S2, S3 Example of general behavior, including playing. S4 Examples of dermatology-related behaviors: scratching, self-licking, rubbing, and shaking. S5 S6 Classification of eating behavior while eating from a slow-feed bowl, which presents some challenges to the algorithms. S7
MDPI and ACS Style

Chambers, R.D.; Yoder, N.C.; Carson, A.B.; Junge, C.; Allen, D.E.; Prescott, L.M.; Bradley, S.; Wymore, G.; Lloyd, K.; Lyle, S. Deep Learning Classification of Canine Behavior Using a Single Collar-Mounted Accelerometer: Real-World Validation. Animals 2021, 11, 1549. https://doi.org/10.3390/ani11061549

AMA Style

Chambers RD, Yoder NC, Carson AB, Junge C, Allen DE, Prescott LM, Bradley S, Wymore G, Lloyd K, Lyle S. Deep Learning Classification of Canine Behavior Using a Single Collar-Mounted Accelerometer: Real-World Validation. Animals. 2021; 11(6):1549. https://doi.org/10.3390/ani11061549

Chicago/Turabian Style

Chambers, Robert D., Nathanael C. Yoder, Aletha B. Carson, Christian Junge, David E. Allen, Laura M. Prescott, Sophie Bradley, Garrett Wymore, Kevin Lloyd, and Scott Lyle. 2021. "Deep Learning Classification of Canine Behavior Using a Single Collar-Mounted Accelerometer: Real-World Validation" Animals 11, no. 6: 1549. https://doi.org/10.3390/ani11061549

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop