Classifying behaviors from animal-borne cameras using machine learning: automated identification of breathing events in sea turtles.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος:	Classifying behaviors from animal-borne cameras using machine learning: automated identification of breathing events in sea turtles.
Συγγραφείς:	Robinson NJ; Institut de Ciències del Mar, Spanish National Research Council - Consejo Superior de Investigaciones Científicas, 08003Barcelona, Spain.; Cape Eleuthera Institute, Cape Eleuthera Island School, Eleuthera, Bahamas.; Fundación Oceanogràfic de la Comunitat Valenciana, Ciudad de las Artes y las Ciencias, 46013 Valencia, Spain., Mazumdar P; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA., Allan BF; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA., Aguzzi J; Institut de Ciències del Mar, Spanish National Research Council - Consejo Superior de Investigaciones Científicas, 08003Barcelona, Spain., Chen B; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA., Hsieh BY; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA., Millet L; Instituto Cavanilles de Biodiversidad y Biología Evolutiva, University of Valencia, Apdo, 22085, 46071 Valencia, Spain., Tomás J; Instituto Cavanilles de Biodiversidad y Biología Evolutiva, University of Valencia, Apdo, 22085, 46071 Valencia, Spain., Terstriep J; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA., Soliman A; National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Champaign, IL 61801, USA.
Πηγή:	The Journal of experimental biology [J Exp Biol] 2026 Jun 01; Vol. 229 (11). Date of Electronic Publication: 2026 Jun 10.
Τύπος έκδοσης:	Journal Article
Γλώσσα:	English
Στοιχεία περιοδικού:	Publisher: Company Of Biologists Limited Country of Publication: England NLM ID: 0243705 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1477-9145 (Electronic) Linking ISSN: 00220949 NLM ISO Abbreviation: J Exp Biol Subsets: MEDLINE
Imprint Name(s):	Publication: London : Company Of Biologists Limited Original Publication: London, Cambridge Univ. Press.
Ιατρικοί όροι (MeSH):	Turtles/physiology , Video Recording/methods , Image Processing, Computer-Assisted/methods , Machine Learning , Respiration* , Behavior, Animal*, Animals ; Classification Algorithms
Περίληψη:	Animal-borne cameras are increasingly used to study animal behavior. Here, we assessed the utility of three machine-learning models - Resnet-50 (3 epochs), Resnet-50 (10 epochs) and Vision Transformer (ViT) (3 epochs) - for identifying breathing behavior from animal-borne camera footage from green turtles (Chelonia mydas). The ViT model had mean Accuracy (97.2%), Precision (63.3%) and F1 (72.7%) scores that outperformed the Resnet-50 models, while all models had a Recall of >99.9%. Thus, the ViT model correctly identified almost all breathing frames although false positives (apnea frames labeled as breathing) were relatively common and led to an over-estimation of breathing rates. We conclude that ViT models are a promising solution for behavioral classification of animal-borne camera footage and even if not yet capable of the fully automated calculation of breathing events in sea turtles, they can still massively reduce the quantity of footage that needs to be manually checked and labeled. (© 2026. Published by The Company of Biologists.)
Competing Interests:	Competing interests The authors declare no competing or financial interests.
Grant Information:	#RYC2021-034381-I Ramon y Cajal postdoctoral program; Cape Eleuthera Institute; Western Connecticut State University; National Oceanic and Atmospheric Administration; NA16NMF0080003 National Marine Fisheries Service; National Center for Supercomputing Applications; University of Illinois; RYC2021-034381-I Ramon y Cajal postdoctoral program; 101112883-GAP-101112883 EU DIGI4ECO; Consejo Superior de Investigaciones Científicas (CSIC)
Contributed Indexing:	Keywords: Chelonia mydas; Artificial intelligence; Behavioral classification; Diving; Image processing; Machine learning
Entry Date(s):	Date Created: 20260518 Date Completed: 20260612 Latest Revision: 20260726
Update Code:	20260726
PubMed Central ID:	PMC13286345
DOI:	10.1242/jeb.251688
PMID:	42144959
Βάση Δεδομένων:	MEDLINE

Περιγραφή
ISSN:	1477-9145
DOI:	10.1242/jeb.251688