Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

Jennifer Magana; Dinu Gavojdian; Yakir Menahem; Teddy Lazebnik; Anna Zamansky; Amber Adams-Progar

doi:10.3389/fvets.2023.1295430

Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

Jennifer Magana, Dinu Gavojdian, Yakir Menahem, Teddy Lazebnik, Anna Zamansky, Amber Adams-Progar

Research output: Contribution to journal › Article › peer-review

3 Scopus citations

Abstract

The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

Original language	English
Article number	1295430
Journal	Frontiers in Veterinary Science
Volume	10
DOIs	https://doi.org/10.3389/fvets.2023.1295430
State	Published - 2023
Externally published	Yes

Bibliographical note

Publisher Copyright:
Copyright © 2023 Magana, Gavojdian, Menahem, Lazebnik, Zamansky and Adams-Progar.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This study was supported by a grant from the Ministry of Research, Innovation, and Digitization, CNCS – UEFISCDI, project number PN-III-P1-1.1-TE-2021-0027, within PNCDI III. During the trial’s implementation, DG received support to join Washington State University through a Romanian – U.S. Fulbright Visiting Scholar grant. We thank the Washington State University Knott Dairy Center staff for providing assistance during data collection.

Funders	Funder number
U.S. Fulbright
Corporation for National and Community Service
Washington State University
Unitatea Executiva pentru Finantarea Invatamantului Superior, a Cercetarii, Dezvoltarii si Inovarii	PN-III-P1-1.1-TE-2021-0027
Ontario Ministry of Research and Innovation

Keywords

animal behavior
dairy cattle
digital dermatitis
machine learning
sensor data

Access to Document

10.3389/fvets.2023.1295430

Cite this

@article{309530da2aad4fb3bd0e34d9f3022d89,

title = "Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data",

abstract = "The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.",

keywords = "animal behavior, dairy cattle, digital dermatitis, machine learning, sensor data",

author = "Jennifer Magana and Dinu Gavojdian and Yakir Menahem and Teddy Lazebnik and Anna Zamansky and Amber Adams-Progar",

note = "Publisher Copyright: Copyright {\textcopyright} 2023 Magana, Gavojdian, Menahem, Lazebnik, Zamansky and Adams-Progar.",

year = "2023",

doi = "10.3389/fvets.2023.1295430",

language = "אנגלית",

volume = "10",

journal = "Frontiers in Veterinary Science",

issn = "2297-1769",

publisher = "Frontiers Media SA",

}

TY - JOUR

T1 - Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

AU - Magana, Jennifer

AU - Gavojdian, Dinu

AU - Menahem, Yakir

AU - Lazebnik, Teddy

AU - Zamansky, Anna

AU - Adams-Progar, Amber

PY - 2023

Y1 - 2023

N2 - The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

AB - The present study aimed to employ machine learning algorithms based on sensor behavior data for (1) early-onset detection of digital dermatitis (DD) and (2) DD prediction in dairy cows. Our machine learning model, which was based on the Tree-Based Pipeline Optimization Tool (TPOT) automatic machine learning method, for DD detection on day 0 of the appearance of the clinical signs has reached an accuracy of 79% on the test set, while the model for the prediction of DD 2 days prior to the appearance of the first clinical signs, which was a combination of K-means and TPOT, has reached an accuracy of 64%. The proposed machine learning models have the potential to help achieve a real-time automated tool for monitoring and diagnosing DD in lactating dairy cows based on sensor data in conventional dairy barn environments. Our results suggest that alterations in behavioral patterns can be used as inputs in an early warning system for herd management in order to detect variances in the health and wellbeing of individual cows.

KW - animal behavior

KW - dairy cattle

KW - digital dermatitis

KW - machine learning

KW - sensor data

UR - http://www.scopus.com/inward/record.url?scp=85179691402&partnerID=8YFLogxK

U2 - 10.3389/fvets.2023.1295430

DO - 10.3389/fvets.2023.1295430

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

C2 - 38105776

AN - SCOPUS:85179691402

SN - 2297-1769

VL - 10

JO - Frontiers in Veterinary Science

JF - Frontiers in Veterinary Science

M1 - 1295430

ER -

Machine learning approaches to predict and detect early-onset of digital dermatitis in dairy cows using sensor data

Abstract

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this