Grant: Discovery Grants Program - Individual
University of Manitoba
$40,000.00
Advancing animal welfare assessment through the development of behavioural metrics of livestock affective states using wearable sensors and computer vision
EF 2024-25
Résumé
Animal welfare is essential for achieving economic viability and meeting societal acceptance of livestock production in Canada and globally. Fulfilling the biological and psychological needs of animals not only optimizes their productive performance but also addresses society's ethical expectations for animal care. Existing animal welfare assessment methodologies are not only costly and time-consuming as they require on-farm visits but are also skewed towards the biological functioning component of animal welfare due to the existence of objectively quantifiable measures. Affective states, which refer to animals' emotions and are composed of valence (i.e., unpleasant vs pleasant) and arousal (i.e., low vs high intensity), are often overlooked. My research program aims to bridge this gap by harnessing the combined potential of data from wearable sensors coupled with machine learning analytics. The long-term vision of my program is to derive reliable behavioural metrics to assess the affective states of dairy cows. The central hypothesis is that data from currently available wearable sensors can objectively measure these states, which can be linked to specific, observable behaviours. In turn, these behaviours can be automatically detected and analyzed using machine learning models for computer vision. My short-term objectives for the next five years are to 1) develop computer vision models for automatic classification of social (i.e., interactions among cows) and individual (i.e., no interaction among cows) cow behaviours; 2) to assess the relationship between agnostic and affiliative social behaviours and cows' affective states; and 3) to study the link between individual cow behaviours and their affective states. The HQPs in this program will develop machine learning skills, meeting the livestock industry's increasing demand for computational animal scientists. Anticipated outcomes encompass pioneering behavioural metrics for assessing dairy cow affective states and developing computer vision models to monitor those behaviours, empowering farmers, policymakers, and stakeholders with actionable insights to enhance the welfare of dairy cows. Moreover, the findings will foster a paradigm shift from human subjective interpretations to a more objective, data-driven elucidation of animal affective states.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Manitoba, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
RGPIN-2024-04885
Name nom
Machado Dallago, Gabriel
Department département
Animal Science
Organization id
19
Institution établissement
University of Manitoba
Province en
Manitoba
Province fr
Manitoba
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
40000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1502
Committee name en
Biological Systems and Functions
Committee name fr
Systèmes et fonctions biologiques
Area of application code
101
Area of application group en
Agriculture and primary food production
Area of application group fr
Agriculture et produits alimentaires primaires
Area of application en
Animal production and animal primary products
Area of application fr
Production animale et produits animaux primaires
Research subject code
5510
Research subject group en
Animal biology
Research subject group fr
Biologie animale
Research subject en
Animal production and breeding
Research subject fr
Production et reproduction animales
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Advancing animal welfare assessment through the development of behavioural metrics of livestock affective states using wearable sensors and computer vision
Keywords
animal welfare; animal affective state; animal behaviour; precision livestock farming; digital livestock technologies; automated monitoring technology; machine learning; deep learning models; livestock production; livestock social acceptability
Application summary
Animal welfare is essential for achieving economic viability and meeting societal acceptance of livestock production in Canada and globally. Fulfilling the biological and psychological needs of animals not only optimizes their productive performance but also addresses society's ethical expectations for animal care. Existing animal welfare assessment methodologies are not only costly and time-consuming as they require on-farm visits but are also skewed towards the biological functioning component of animal welfare due to the existence of objectively quantifiable measures. Affective states, which refer to animals' emotions and are composed of valence (i.e., unpleasant vs pleasant) and arousal (i.e., low vs high intensity), are often overlooked. My research program aims to bridge this gap by harnessing the combined potential of data from wearable sensors coupled with machine learning analytics. The long-term vision of my program is to derive reliable behavioural metrics to assess the affective states of dairy cows. The central hypothesis is that data from currently available wearable sensors can objectively measure these states, which can be linked to specific, observable behaviours. In turn, these behaviours can be automatically detected and analyzed using machine learning models for computer vision. My short-term objectives for the next five years are to 1) develop computer vision models for automatic classification of social (i.e., interactions among cows) and individual (i.e., no interaction among cows) cow behaviours; 2) to assess the relationship between agnostic and affiliative social behaviours and cows' affective states; and 3) to study the link between individual cow behaviours and their affective states. The HQPs in this program will develop machine learning skills, meeting the livestock industry's increasing demand for computational animal scientists. Anticipated outcomes encompass pioneering behavioural metrics for assessing dairy cow affective states and developing computer vision models to monitor those behaviours, empowering farmers, policymakers, and stakeholders with actionable insights to enhance the welfare of dairy cows. Moreover, the findings will foster a paradigm shift from human subjective interpretations to a more objective, data-driven elucidation of animal affective states.