Vasseur, Elsa; Diallo, Abdoulaye Baniré
Responsible Development and Deployment of AI and IoT in the Dairy Industry, bridging gaps between research and implementation: The WELL-E Initiative [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: conférencier invité, Presentation
@misc{vasseur_responsible_2025,
title = {Responsible Development and Deployment of AI and IoT in the Dairy Industry, bridging gaps between research and implementation: The WELL-E Initiative},
author = {Elsa Vasseur and Abdoulaye Baniré Diallo},
url = {https://eaap2025.org/},
year = {2025},
date = {2025-08-01},
urldate = {2025-08-01},
address = {Innbruck, Austria},
abstract = {Responsible AI that properly addresses real-world stakeholder needs is at the heart of the R+I Chair in Animal Welfare and Artificial Intelligence. WELL-E approach relies on the use of IoT, computer vision, and machine learning to improve our ability to detect and monitor changes in animal welfare earlier than possible with visual methods, and to generate predictions to aid on-farm decision-making, ensuring that resources and efforts are focused on the animals most likely to succeed in the long term. Grounded in industry partnerships, members of the dairy community sit on both the scientific and management committees, where they contribute to research orientations, leading to the co-creation of research projects and initiatives. This collaborative approach not only ensures the applicability and relevance of all research carried out, but allows for continuous feedback between the research and industry environments, creating a true digital living lab that can grow with the Chair. Launched in 2023, our team has been conducting pilot research on two research farms and working to build functional and resilient data collection infrastructure for implementation on commercial dairy farms throughout 2025. Through a federated learning approach, this deployment ensures robust continuous monitoring across a network of farms, while respecting data confidentiality and cybersecurity. We have developed a framework for the study of animal behaviours and emotions, presenting a paradigm shift for both annotation and data analysis based on continuouscand heterogeneous data sources. As we grow our network of partner farms, we will continue focussed studies on research farms, allowing for continuous exchange between the two contexts. This collaborative approach promotes the responsible integration of new technologies to the industry and empowers endusers to be at the forefront of these new developments, ensuring their sustainability and reinforcing the importance of stakeholder participation in innovative scientific research.},
keywords = {conférencier invité, Presentation},
pubstate = {published},
tppubtype = {misc}
}
Responsible AI that properly addresses real-world stakeholder needs is at the heart of the R+I Chair in Animal Welfare and Artificial Intelligence. WELL-E approach relies on the use of IoT, computer vision, and machine learning to improve our ability to detect and monitor changes in animal welfare earlier than possible with visual methods, and to generate predictions to aid on-farm decision-making, ensuring that resources and efforts are focused on the animals most likely to succeed in the long term. Grounded in industry partnerships, members of the dairy community sit on both the scientific and management committees, where they contribute to research orientations, leading to the co-creation of research projects and initiatives. This collaborative approach not only ensures the applicability and relevance of all research carried out, but allows for continuous feedback between the research and industry environments, creating a true digital living lab that can grow with the Chair. Launched in 2023, our team has been conducting pilot research on two research farms and working to build functional and resilient data collection infrastructure for implementation on commercial dairy farms throughout 2025. Through a federated learning approach, this deployment ensures robust continuous monitoring across a network of farms, while respecting data confidentiality and cybersecurity. We have developed a framework for the study of animal behaviours and emotions, presenting a paradigm shift for both annotation and data analysis based on continuouscand heterogeneous data sources. As we grow our network of partner farms, we will continue focussed studies on research farms, allowing for continuous exchange between the two contexts. This collaborative approach promotes the responsible integration of new technologies to the industry and empowers endusers to be at the forefront of these new developments, ensuring their sustainability and reinforcing the importance of stakeholder participation in innovative scientific research.
Egger-Danner, C.; Klaas, I.; Brito, L.; Bewley, J.; Cabrera, V.; Gengler, N.; Haskel, M.; Heringstad, B.; Hostens, M.; Iwersen, M.; Schodl, A.; Stock, K.; Stygar, A.; Vasseur, E.
ICAR/IDF initiative: Guidelines and reference standards for using rumination sensor data in animal health and welfare recording [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{egger-danner_icaridf_2025,
title = {ICAR/IDF initiative: Guidelines and reference standards for using rumination sensor data in animal health and welfare recording},
author = {C. Egger-Danner and I. Klaas and L. Brito and J. Bewley and V. Cabrera and N. Gengler and M. Haskel and B. Heringstad and M. Hostens and M. Iwersen and A. Schodl and K. Stock and A. Stygar and E. Vasseur},
url = {https://eaap2025.org/},
year = {2025},
date = {2025-08-01},
urldate = {2025-08-01},
address = {Innbruck, Austria},
abstract = {On-farm sensor data offers new opportunities for animal health and welfare assessment when used along the dairy value chain, thus improving transparency and sustainability in the dairy sector. The aim of the joint ICAR IDF initiative is to develop guidelines to facilitate use of sensor data for improving animal health and welfare. The joint initiative consists of different stakeholder groups, scientists, breeding organizations and manufacturers and has the goal to develop: Definitions and terminology for health conditions and behaviors measured with sensor systemsStandards to facilitate data exchange along the dairy value chain following acknowledged ICAR and IDF principles Guidelines for best practices on data collection, handling and analysis for use in genetics, health and welfare assessmentRecommendations and protocols for testing sensor performanceFirst results from the use case sensor-based rumination data will be presented, recommendations on use of rumination data, key performance indicators and references standards for animal health and welfare assessment given. The close collaboration between the relevant stakeholders in the joint ICAR/IDF initiative enables development of guidelines considering industry needs while promoting use of sensor data along the dairy value chain.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
On-farm sensor data offers new opportunities for animal health and welfare assessment when used along the dairy value chain, thus improving transparency and sustainability in the dairy sector. The aim of the joint ICAR IDF initiative is to develop guidelines to facilitate use of sensor data for improving animal health and welfare. The joint initiative consists of different stakeholder groups, scientists, breeding organizations and manufacturers and has the goal to develop: Definitions and terminology for health conditions and behaviors measured with sensor systemsStandards to facilitate data exchange along the dairy value chain following acknowledged ICAR and IDF principles Guidelines for best practices on data collection, handling and analysis for use in genetics, health and welfare assessmentRecommendations and protocols for testing sensor performanceFirst results from the use case sensor-based rumination data will be presented, recommendations on use of rumination data, key performance indicators and references standards for animal health and welfare assessment given. The close collaboration between the relevant stakeholders in the joint ICAR/IDF initiative enables development of guidelines considering industry needs while promoting use of sensor data along the dairy value chain.
Vasseur, Elsa; Dallago, Gabiel M.; Diallo, Abdoulaye Baniré
Perspectives from Canada on Dairy Cows’ Longevity [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{vasseur_perspectives_2025,
title = {Perspectives from Canada on Dairy Cows’ Longevity},
author = {Elsa Vasseur and Gabiel M. Dallago and Abdoulaye Baniré Diallo},
url = {chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://eaap2025.org/Final_Programme_Innsbruck_2025.pdf?v=5},
year = {2025},
date = {2025-08-01},
urldate = {2025-08-01},
address = {Innbruck, Austria},
abstract = {Our focus is ensuring the sustainable life of dairy cattle, or, how to keep healthy, comfortable animals longer in the herd, encompassing research on animal welfare for better management and facilities, together with our understanding of cow longevity. Our approach is to develop methods to remotely assess changes in cow welfare, and to evaluate how health issues and other risk factors are linked to cow longevity. We aimed to exploit the potential of the considerable dataset regularly collected on Canadian dairies via the Canadian Dairy Network (Lactanet). We have assessed the impact of health events such as lameness and mastitis, finding that primiparous cows afflicted with these diseases have reduced milk yields and gross profit, and higher risk of culling than healthy cows. We also unveiled valuable insights on early-life variables that are linked with future productivity and longevity, such as weight at birth and at weaning, weaning age, and serum IgG levels; these represent early indicators that can be used to select replacement animals based on their potential, and reduce economic and environmental losses incurred by keeping animals that will likely never be profitable. Our work has also led to innovations in the remote detection of welfare. On a herd basis, the analysis of DHI indicators routinely collected on dairies was used to develop a composite herd welfare index. On an individual cow basis, we have tested the use of spectral data from milk samples as a source of information on cow-specific changes in welfare status; this approach was successful in linking milk components with improvements in specific outcomes of cow welfare. Overall, these innovations show that the potential to conduct remote assessments is present and could thus be extended to more specific topics (e.g., lameness in individual cows) and other aspects (e.g., economic and environmental performance of cows and farms).},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Our focus is ensuring the sustainable life of dairy cattle, or, how to keep healthy, comfortable animals longer in the herd, encompassing research on animal welfare for better management and facilities, together with our understanding of cow longevity. Our approach is to develop methods to remotely assess changes in cow welfare, and to evaluate how health issues and other risk factors are linked to cow longevity. We aimed to exploit the potential of the considerable dataset regularly collected on Canadian dairies via the Canadian Dairy Network (Lactanet). We have assessed the impact of health events such as lameness and mastitis, finding that primiparous cows afflicted with these diseases have reduced milk yields and gross profit, and higher risk of culling than healthy cows. We also unveiled valuable insights on early-life variables that are linked with future productivity and longevity, such as weight at birth and at weaning, weaning age, and serum IgG levels; these represent early indicators that can be used to select replacement animals based on their potential, and reduce economic and environmental losses incurred by keeping animals that will likely never be profitable. Our work has also led to innovations in the remote detection of welfare. On a herd basis, the analysis of DHI indicators routinely collected on dairies was used to develop a composite herd welfare index. On an individual cow basis, we have tested the use of spectral data from milk samples as a source of information on cow-specific changes in welfare status; this approach was successful in linking milk components with improvements in specific outcomes of cow welfare. Overall, these innovations show that the potential to conduct remote assessments is present and could thus be extended to more specific topics (e.g., lameness in individual cows) and other aspects (e.g., economic and environmental performance of cows and farms).
Benserir, Nadjib; Etiabi, Yaya; Sabir, Essaid; Amhoud, Elmehdi; Elbiaze, Halima; Diallo, Abdoulaye Baniré
IRSA Over Spreading Factors for Spatio-Temporal SIC in Scalable LoRaWAN IoT Networks [External] Miscellaneous
2025.
Links | BibTeX | Tags: Presentation
@misc{benserir_irsa_2025,
title = {IRSA Over Spreading Factors for Spatio-Temporal SIC in Scalable LoRaWAN IoT Networks},
author = {Nadjib Benserir and Yaya Etiabi and Essaid Sabir and Elmehdi Amhoud and Halima Elbiaze and Abdoulaye Baniré Diallo},
url = {https://ieee-iscc.computer.org/2025},
year = {2025},
date = {2025-07-01},
urldate = {2025-07-01},
address = {Bologna, Italy},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Vasseur, Elsa
The WELL-E Initiative: Inclusive Innovation to deliver data-driven solutions with and for the Canadian Dairy Industry [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{vasseur_well-e_2025,
title = {The WELL-E Initiative: Inclusive Innovation to deliver data-driven solutions with and for the Canadian Dairy Industry},
author = {Elsa Vasseur},
url = {https://www.esee2025.utad.pt/program/},
year = {2025},
date = {2025-07-01},
urldate = {2025-07-01},
address = {Vila Real, Portugal},
abstract = {Purpose textbar Evolving theoretical frameworks of responsible and inclusive innovation argue systems change must properly address real-world stakeholder needs and create positive impacts for society and environment. These principles lie at the heart and mission of the Research and Innovation Chair in Animal Welfare and Artificial Intelligence (WELL-E). Our team works to integrate intentionally both stakeholder and domain expert knowledge with cutting-edge AI methods and tools for the improvement of animal (and human) welfare. WELL-E is a Digital Living Laboratory, focused on the needs of animals and end-users, and is currently running at two research facilities: a university training environment and a vocational training environment for incarcerated persons, creating a community of practice allowing co-creation of technology, knowledge, and information with farmers, educators, researchers, and extensionists.
Design/Methodology/Approach textbar Beginning with the university training environment, we have been working directly with farm staff and management to co-develop and pilot new practices for animal housing and management. Specifically through living lab experiments from 2019-2024, we aimed to understand how end users transition herds from a movement-restricted system to having regular outdoor access in a practical way while enhancing animal welfare and ethics. Our next major project focusses on working directly with the vocational training environment staff on the construction of a new dairy as a research location to test cutting-edge technologies and practices for the deployment of responsible AI tools on farms, embracing F.A.I.R. principles and empowering end users to be at the forefront of these innovations.
Findingstextbar While our team’s research on developing new knowledge has led to major changes in practice, we have since been working directly with farm staff and management at the research training farm on how best to introduce increased movement opportunities (group size on exits, handler methods, etc.), ensuring its swift adoption across Canada. Vocational farm staff have been working directly with WELL-E to build training curriculum on stockmanship and clerical work (data entry, use of sensors), as well as to pilot new tools and technologies.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Purpose textbar Evolving theoretical frameworks of responsible and inclusive innovation argue systems change must properly address real-world stakeholder needs and create positive impacts for society and environment. These principles lie at the heart and mission of the Research and Innovation Chair in Animal Welfare and Artificial Intelligence (WELL-E). Our team works to integrate intentionally both stakeholder and domain expert knowledge with cutting-edge AI methods and tools for the improvement of animal (and human) welfare. WELL-E is a Digital Living Laboratory, focused on the needs of animals and end-users, and is currently running at two research facilities: a university training environment and a vocational training environment for incarcerated persons, creating a community of practice allowing co-creation of technology, knowledge, and information with farmers, educators, researchers, and extensionists.
Design/Methodology/Approach textbar Beginning with the university training environment, we have been working directly with farm staff and management to co-develop and pilot new practices for animal housing and management. Specifically through living lab experiments from 2019-2024, we aimed to understand how end users transition herds from a movement-restricted system to having regular outdoor access in a practical way while enhancing animal welfare and ethics. Our next major project focusses on working directly with the vocational training environment staff on the construction of a new dairy as a research location to test cutting-edge technologies and practices for the deployment of responsible AI tools on farms, embracing F.A.I.R. principles and empowering end users to be at the forefront of these innovations.
Findingstextbar While our team’s research on developing new knowledge has led to major changes in practice, we have since been working directly with farm staff and management at the research training farm on how best to introduce increased movement opportunities (group size on exits, handler methods, etc.), ensuring its swift adoption across Canada. Vocational farm staff have been working directly with WELL-E to build training curriculum on stockmanship and clerical work (data entry, use of sensors), as well as to pilot new tools and technologies.
Design/Methodology/Approach textbar Beginning with the university training environment, we have been working directly with farm staff and management to co-develop and pilot new practices for animal housing and management. Specifically through living lab experiments from 2019-2024, we aimed to understand how end users transition herds from a movement-restricted system to having regular outdoor access in a practical way while enhancing animal welfare and ethics. Our next major project focusses on working directly with the vocational training environment staff on the construction of a new dairy as a research location to test cutting-edge technologies and practices for the deployment of responsible AI tools on farms, embracing F.A.I.R. principles and empowering end users to be at the forefront of these innovations.
Findingstextbar While our team’s research on developing new knowledge has led to major changes in practice, we have since been working directly with farm staff and management at the research training farm on how best to introduce increased movement opportunities (group size on exits, handler methods, etc.), ensuring its swift adoption across Canada. Vocational farm staff have been working directly with WELL-E to build training curriculum on stockmanship and clerical work (data entry, use of sensors), as well as to pilot new tools and technologies.
Roche, Steven; Vasseur, Elsa
Advancing Dairy Extension: A Digital Living Laboratory Approach to Knowledge Mobilization in the Canadian Dairy Sector [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{roche_advancing_2025,
title = {Advancing Dairy Extension: A Digital Living Laboratory Approach to Knowledge Mobilization in the Canadian Dairy Sector},
author = {Steven Roche and Elsa Vasseur},
url = {https://www.esee2025.utad.pt/program/},
year = {2025},
date = {2025-07-01},
urldate = {2025-07-01},
address = {Vila Real, Portugal},
abstract = {The WELL-E initiative is a national research program that applies a Digital Living Laboratory (DLL) model to close the gap between scientific innovation and on-farm application in the Canadian dairy sector. With a focus on animal welfare, AI integration, and digital innovation, WELL-E aims to generate actionable knowledge, enable practical adoption, and build system-wide capacity. Early work has centered on developing a robust knowledge translation and transfer (KTT) strategy grounded in literature, systems mapping, and engagement with advisory networks.
WELL-E employs a dual push-pull KTT model, blending structured dissemination (e.g., training modules, videos) with participatory strategies that ensure tools and messages are co-developed with stakeholders. Findings emphasize the importance of trusted messengers, multi-modal delivery, and the integration of pedagogical principles in digital resource design. WELL-E’s mapping of Canada’s dairy KTT ecosystem—hosted on Kumu—has identified fragmentation and opportunities for improved collaboration.
This work offers both practical and theoretical contributions. It provides a scalable model for digital extension grounded in systems thinking, while reinforcing that digital tools alone are insufficient—effective KTT requires trust, contextual fit, and continuous evaluation. WELL-E’s early lessons highlight a path forward for AI-enabled extension services that are credible, collaborative, and impactful.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
The WELL-E initiative is a national research program that applies a Digital Living Laboratory (DLL) model to close the gap between scientific innovation and on-farm application in the Canadian dairy sector. With a focus on animal welfare, AI integration, and digital innovation, WELL-E aims to generate actionable knowledge, enable practical adoption, and build system-wide capacity. Early work has centered on developing a robust knowledge translation and transfer (KTT) strategy grounded in literature, systems mapping, and engagement with advisory networks.
WELL-E employs a dual push-pull KTT model, blending structured dissemination (e.g., training modules, videos) with participatory strategies that ensure tools and messages are co-developed with stakeholders. Findings emphasize the importance of trusted messengers, multi-modal delivery, and the integration of pedagogical principles in digital resource design. WELL-E’s mapping of Canada’s dairy KTT ecosystem—hosted on Kumu—has identified fragmentation and opportunities for improved collaboration.
This work offers both practical and theoretical contributions. It provides a scalable model for digital extension grounded in systems thinking, while reinforcing that digital tools alone are insufficient—effective KTT requires trust, contextual fit, and continuous evaluation. WELL-E’s early lessons highlight a path forward for AI-enabled extension services that are credible, collaborative, and impactful.
WELL-E employs a dual push-pull KTT model, blending structured dissemination (e.g., training modules, videos) with participatory strategies that ensure tools and messages are co-developed with stakeholders. Findings emphasize the importance of trusted messengers, multi-modal delivery, and the integration of pedagogical principles in digital resource design. WELL-E’s mapping of Canada’s dairy KTT ecosystem—hosted on Kumu—has identified fragmentation and opportunities for improved collaboration.
This work offers both practical and theoretical contributions. It provides a scalable model for digital extension grounded in systems thinking, while reinforcing that digital tools alone are insufficient—effective KTT requires trust, contextual fit, and continuous evaluation. WELL-E’s early lessons highlight a path forward for AI-enabled extension services that are credible, collaborative, and impactful.
Hambly, Helen; Vasseur, Elsa
Learning About Crossing the Lines: The WELL-E Approach to Capacity Development [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{hambly_learning_2025,
title = {Learning About Crossing the Lines: The WELL-E Approach to Capacity Development},
author = {Helen Hambly and Elsa Vasseur},
url = {https://www.esee2025.utad.pt/program/},
year = {2025},
date = {2025-07-01},
urldate = {2025-07-01},
address = {Vila Real, Portugal},
abstract = {Short abstract (200 words): WELL-E, Research and Innovation Chair on Animal Welfare and Artificial Intelligence, is a Digital Living Laboratory, focused on the needs of animals and end-users with deep industry partnerships, as well as multi-disciplinary scientific committees, in which end-users and the research team interact intensively in the co-creation of projects and delivery of high-impact results. WELL-E strategically pursues capacity development (CapDev) as an all-encompassing approach to building capacity, enabling capacities to change ways of knowing and doing, thereby responding and contributing to systems change. WELL-E’s pursuit of CapDev to enable individual students to build direct links with diverse scientists, organizations, and networks across systems of higher education, research partnerships and industry is an important model for future Living Labs. WELL-E’s contribution to transforming agri-food systems and extension services through enhancing capacity crosses many boundaries, not only within science disciplines but twinning science-based industries like dairy and information technology that have not always had sufficient learning opportunities together.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Short abstract (200 words): WELL-E, Research and Innovation Chair on Animal Welfare and Artificial Intelligence, is a Digital Living Laboratory, focused on the needs of animals and end-users with deep industry partnerships, as well as multi-disciplinary scientific committees, in which end-users and the research team interact intensively in the co-creation of projects and delivery of high-impact results. WELL-E strategically pursues capacity development (CapDev) as an all-encompassing approach to building capacity, enabling capacities to change ways of knowing and doing, thereby responding and contributing to systems change. WELL-E’s pursuit of CapDev to enable individual students to build direct links with diverse scientists, organizations, and networks across systems of higher education, research partnerships and industry is an important model for future Living Labs. WELL-E’s contribution to transforming agri-food systems and extension services through enhancing capacity crosses many boundaries, not only within science disciplines but twinning science-based industries like dairy and information technology that have not always had sufficient learning opportunities together.
Muszik, Jasmine; Aigueperse, Nadége; Cellier, Marjorie; Diallo, Abdoulay Baniré; Vasseur, Elsa
The Relationship Between Motivation, Anticipation, and Frustration in Animals: A Scoping Review [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{muszik_relationship_2025,
title = {The Relationship Between Motivation, Anticipation, and Frustration in Animals: A Scoping Review},
author = {Jasmine Muszik and Nadége Aigueperse and Marjorie Cellier and Abdoulay Baniré Diallo and Elsa Vasseur},
url = {https://www.ufaw.org.uk/ufaw-events/ufaw-international-animal-welfare-conference-2025},
year = {2025},
date = {2025-06-01},
urldate = {2025-06-01},
address = {Online},
abstract = {Motivation, anticipation, and frustration, are important concepts for understanding animal emotion. Denying or delaying an outcome that an animal is motivated for, expects, and performs specific behavior in preparation of, can cause frustration, impacting the emotional welfare of the animal. However, diversity in the definitions and methodology related to these concepts can lead to confusing or conflicting results, and may complicate replication of previous work. A scoping review (conducted according to PRISMA guidelines) was completed where a multi-step screening process identified 112 final primary research papers. The objectives of the review were to determine the ways in which the three concepts have been studied in previous literature, the relationship between the main concepts, and the gaps in knowledge regarding these concepts. Advantages and drawbacks of tests and variables used in the selected papers were explored, along with the repeatability, the applicability within and between individuals or species, and the interpretation of results. There were no universally accepted definitions of motivation, anticipation, or frustration, and the methods used to find or show evidence of the concepts varied, even within a species using the same test. However, a clear connection between concepts was confirmed, where it was suggested that authors develop a framework to measure the concepts together to understand the full scope of how different decisions or practices may impact animal emotion.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Motivation, anticipation, and frustration, are important concepts for understanding animal emotion. Denying or delaying an outcome that an animal is motivated for, expects, and performs specific behavior in preparation of, can cause frustration, impacting the emotional welfare of the animal. However, diversity in the definitions and methodology related to these concepts can lead to confusing or conflicting results, and may complicate replication of previous work. A scoping review (conducted according to PRISMA guidelines) was completed where a multi-step screening process identified 112 final primary research papers. The objectives of the review were to determine the ways in which the three concepts have been studied in previous literature, the relationship between the main concepts, and the gaps in knowledge regarding these concepts. Advantages and drawbacks of tests and variables used in the selected papers were explored, along with the repeatability, the applicability within and between individuals or species, and the interpretation of results. There were no universally accepted definitions of motivation, anticipation, or frustration, and the methods used to find or show evidence of the concepts varied, even within a species using the same test. However, a clear connection between concepts was confirmed, where it was suggested that authors develop a framework to measure the concepts together to understand the full scope of how different decisions or practices may impact animal emotion.
Arpin, Catherine; Vliet, Rachel; Cellier, Marjorie; Aigueperse, Nadège; Robichaud, Marianne Villettaz; Diallo, Abdoulaye Baniré; Vasseur, Elsa
Udderly important behaviours: The study of the behavioural needs of adult dairy cows across common intensive housing systems [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{arpin_udderly_2025,
title = {Udderly important behaviours: The study of the behavioural needs of adult dairy cows across common intensive housing systems},
author = {Catherine Arpin and Rachel Vliet and Marjorie Cellier and Nadège Aigueperse and Marianne Villettaz Robichaud and Abdoulaye Baniré Diallo and Elsa Vasseur},
url = {https://www.ufaw.org.uk/ufaw-events/ufaw-international-animal-welfare-conference-2025},
year = {2025},
date = {2025-06-01},
urldate = {2025-06-01},
address = {Online},
abstract = {Behavioural needs represent behaviours that animals perform no matter the environment or their physiological state. If these needs are not met, the welfare of the animals has the potential to be compromised. Characterizing the behavioural needs of farm animals is pivotal in understanding how to satisfy them by for example modifying their captive environment. However, behavioural needs are not well understood, and both their definition and application are inconsistent across the literature, particularly in the case of dairy cows. The objectives of this review are to identify papers that study behavioural needs of dairy cows, identify which needs were studied, how they were studied, and how that changed over time. A scoping review (conducted according to PRISMA guidelines) was completed, where 11,512 articles were reduced to 144 through a multi-step screening process. The included papers were published between 1946 and 2024 mostly in Canada (15%), in the United States (14%) and in the UK (11%). Of the remaining articles, only 15% acknowledged the existence of behavioural needs, and there was a lack of consistency in the terms used surrounding this concept. Resting and feeding behaviours were the most studied (in 122 and 86 papers, respectively), as opposed to grooming and exploratory behaviours that were only studied in 16 and 9 studies, respectively. By examining the timeline of when different behavioural categories were studied, it was revealed that the study of behavioural needs in adult dairy cows has evolved greatly in time and was guided by the priorities and the tools available at the time. This review summarises where the concept of behavioural needs fits within the study of dairy cows and highlights the gaps in the literature observed in the past and the present. It is imperative that these concepts be better understood and defined in the literature for the study of dairy cow welfare to move forward.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Behavioural needs represent behaviours that animals perform no matter the environment or their physiological state. If these needs are not met, the welfare of the animals has the potential to be compromised. Characterizing the behavioural needs of farm animals is pivotal in understanding how to satisfy them by for example modifying their captive environment. However, behavioural needs are not well understood, and both their definition and application are inconsistent across the literature, particularly in the case of dairy cows. The objectives of this review are to identify papers that study behavioural needs of dairy cows, identify which needs were studied, how they were studied, and how that changed over time. A scoping review (conducted according to PRISMA guidelines) was completed, where 11,512 articles were reduced to 144 through a multi-step screening process. The included papers were published between 1946 and 2024 mostly in Canada (15%), in the United States (14%) and in the UK (11%). Of the remaining articles, only 15% acknowledged the existence of behavioural needs, and there was a lack of consistency in the terms used surrounding this concept. Resting and feeding behaviours were the most studied (in 122 and 86 papers, respectively), as opposed to grooming and exploratory behaviours that were only studied in 16 and 9 studies, respectively. By examining the timeline of when different behavioural categories were studied, it was revealed that the study of behavioural needs in adult dairy cows has evolved greatly in time and was guided by the priorities and the tools available at the time. This review summarises where the concept of behavioural needs fits within the study of dairy cows and highlights the gaps in the literature observed in the past and the present. It is imperative that these concepts be better understood and defined in the literature for the study of dairy cow welfare to move forward.
Amarioarei, Georgiana; Vliet, Rachel; Cellier, Marjorie; Aigueperse, Nadège; Diallo, Abdoulaye Baniré; Vasseur, Elsa
A scoping review of cognitive enrichment for young cattle and its implications for welfare and agricultural practices [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{amarioarei_scoping_2025,
title = {A scoping review of cognitive enrichment for young cattle and its implications for welfare and agricultural practices},
author = {Georgiana Amarioarei and Rachel Vliet and Marjorie Cellier and Nadège Aigueperse and Abdoulaye Baniré Diallo and Elsa Vasseur},
url = {https://www.ufaw.org.uk/ufaw-events/ufaw-international-animal-welfare-conference-2025},
year = {2025},
date = {2025-06-01},
urldate = {2025-06-01},
address = {Online},
abstract = {Cognitive enrichment satisfies the behavioral needs of animals by engaging their natural instincts, providing mental stimulation, and enhancing cognitive skills. It distinguishes itself from other types of enrichment by introducing cognitively challenging tasks to the environment and promoting cognitive stimulation. For young cattle, this form of enrichment is necessary for supporting healthy development, adaptability and overall well-being. This scoping review seeks to critically assess the current state of literature on the effects of cognitive enrichment on young domestic bovine within an agricultural environment. This is substantial for understanding benefits to individual animals in addition to possible implications for the efficiency and sustainability of farming practices. The scoping review specifically focuses on identifying gaps in knowledge related to methodology, terminology, practical implementation, and future directions of cognitive enrichment practices. A comprehensive search of databases Scopus and Web of Science was conducted using PRISMA guidelines to identify records published between 1970 and 2024. Studies were included if they met the criteria examining cognitive enrichment or related interventions for calves and heifers. After a multistep screening process, a total of 32 studies reduced from 13,195 were included in the final analysis. Results of the analysis showed inconsistent definitions of age classification of bovine developmental stages and missing fundamental cognitive enrichment terminology. These results highlight the difficulties in tracking and defining the emergence of cognitive enrichment for young bovines due to lack of standardization. Consequently, non-standardized methodologies impose limitations on cross-study comparisons and hinder the development of evidence-based recommendations for practical implementation. Secondly, methodological elements such as measures and practicality of tests were categorically organized to investigate the potential welfare implications of cognitive enrichment and were found to hold meaningful contributions towards the topic. As a whole, cognitive enrichment holds significant promise as a tool for enhancing the welfare and cognitive development of young bovine.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Cognitive enrichment satisfies the behavioral needs of animals by engaging their natural instincts, providing mental stimulation, and enhancing cognitive skills. It distinguishes itself from other types of enrichment by introducing cognitively challenging tasks to the environment and promoting cognitive stimulation. For young cattle, this form of enrichment is necessary for supporting healthy development, adaptability and overall well-being. This scoping review seeks to critically assess the current state of literature on the effects of cognitive enrichment on young domestic bovine within an agricultural environment. This is substantial for understanding benefits to individual animals in addition to possible implications for the efficiency and sustainability of farming practices. The scoping review specifically focuses on identifying gaps in knowledge related to methodology, terminology, practical implementation, and future directions of cognitive enrichment practices. A comprehensive search of databases Scopus and Web of Science was conducted using PRISMA guidelines to identify records published between 1970 and 2024. Studies were included if they met the criteria examining cognitive enrichment or related interventions for calves and heifers. After a multistep screening process, a total of 32 studies reduced from 13,195 were included in the final analysis. Results of the analysis showed inconsistent definitions of age classification of bovine developmental stages and missing fundamental cognitive enrichment terminology. These results highlight the difficulties in tracking and defining the emergence of cognitive enrichment for young bovines due to lack of standardization. Consequently, non-standardized methodologies impose limitations on cross-study comparisons and hinder the development of evidence-based recommendations for practical implementation. Secondly, methodological elements such as measures and practicality of tests were categorically organized to investigate the potential welfare implications of cognitive enrichment and were found to hold meaningful contributions towards the topic. As a whole, cognitive enrichment holds significant promise as a tool for enhancing the welfare and cognitive development of young bovine.
Cellier, Marjorie; Gisiger, Thomas; Diallo, Abdoulaye Baniré; Vasseur, Elsa
Enhancing Statistical Power in Animal Behavior Research: A Case Study on Enrichment in Dairy Cows Using Data Augmentation Methods [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{cellier_enhancing_2025,
title = {Enhancing Statistical Power in Animal Behavior Research: A Case Study on Enrichment in Dairy Cows Using Data Augmentation Methods},
author = {Marjorie Cellier and Thomas Gisiger and Abdoulaye Baniré Diallo and Elsa Vasseur},
url = {https://www.youtube.com/watch?v=vRVwPu_lO2Y&list=PLf1GMQ3ilgOwgkl5wD-_dH-E6TeUbj2zA&index=3},
doi = {10.1038/s41598-025-89891-4},
year = {2025},
date = {2025-05-01},
urldate = {2025-03-06},
volume = {15},
number = {1},
pages = {5917},
address = {Guelph, Canada},
abstract = {Research on animals often faces a critical limitation: small sample sizes limit statistical power, making it challenging to detect behavioral differences. Artificial data augmentation offers a promising solution by increasing dataset size. This study explores the potential of data augmentation in behavioral research through a case study on dairy cow enrichment in tie-stall housing.
Ten Holstein dairy cows were tested using a rossover design over two weeks, with a four-day washout
period. Two enrichment types were evaluated: a standard Kong and a modified version with additional visual (white tape) and tactile (chains) elements. Each device was suspended in the stall. Behavioral interactions with the enrichment were recorded using GoPro cameras for 45-minute sessions and annotated using The Observer XT software. A linear mixed model was applied, with interaction type, day, and treatment as fixed effects, and individuals nested within day as a random effect. Preliminary results showed no significant difference in the percentage of time spent interacting with the two Kong types (standard: 6.9±5.3% of their time; modified: 6.6±6.6% of their time},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Research on animals often faces a critical limitation: small sample sizes limit statistical power, making it challenging to detect behavioral differences. Artificial data augmentation offers a promising solution by increasing dataset size. This study explores the potential of data augmentation in behavioral research through a case study on dairy cow enrichment in tie-stall housing.
Ten Holstein dairy cows were tested using a rossover design over two weeks, with a four-day washout
period. Two enrichment types were evaluated: a standard Kong and a modified version with additional visual (white tape) and tactile (chains) elements. Each device was suspended in the stall. Behavioral interactions with the enrichment were recorded using GoPro cameras for 45-minute sessions and annotated using The Observer XT software. A linear mixed model was applied, with interaction type, day, and treatment as fixed effects, and individuals nested within day as a random effect. Preliminary results showed no significant difference in the percentage of time spent interacting with the two Kong types (standard: 6.9±5.3% of their time; modified: 6.6±6.6% of their time
Ten Holstein dairy cows were tested using a rossover design over two weeks, with a four-day washout
period. Two enrichment types were evaluated: a standard Kong and a modified version with additional visual (white tape) and tactile (chains) elements. Each device was suspended in the stall. Behavioral interactions with the enrichment were recorded using GoPro cameras for 45-minute sessions and annotated using The Observer XT software. A linear mixed model was applied, with interaction type, day, and treatment as fixed effects, and individuals nested within day as a random effect. Preliminary results showed no significant difference in the percentage of time spent interacting with the two Kong types (standard: 6.9±5.3% of their time; modified: 6.6±6.6% of their time
Arpin, Catherine; Cellier, Marjorie; Wolfe, Tania; Almeida, Hayda; Julliot, Célia; Robichaud, Marianne Villettaz; Diallo, Abdoulaye Baniré; Vasseur, Elsa
Hoofing it to New Homes: How the transition towards a new housing system is experienced by dairy cows. [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{arpin_hoofing_2025,
title = {Hoofing it to New Homes: How the transition towards a new housing system is experienced by dairy cows.},
author = {Catherine Arpin and Marjorie Cellier and Tania Wolfe and Hayda Almeida and Célia Julliot and Marianne Villettaz Robichaud and Abdoulaye Baniré Diallo and Elsa Vasseur},
url = {https://ccsaw.uoguelph.ca/upcoming-events/2025-isae-north-american-regional-meeting/},
year = {2025},
date = {2025-05-01},
urldate = {2025-05-01},
address = {Guelph, Canada},
abstract = {In response to a 2023 change in the Dairy Code of Practice that prohibits continuous tethering as of 2027, there is a predicted rise in the construction of free-stall and stall-free barns, and more cows transitioning between housing systems. The transition period of dairy cattle after a housing change is of interest from a welfare standpoint as there are behavioural adjustments, and potential stressors that could influence their ability to adapt to a new environment. Therefore, this study aims to understand how the transition towards a new barn is experienced by lactating dairy cows and is divided into 2 case studies: (1) Examining how the transition to a new housing system affects the cows' time-budget and how long it takes to stabilize, and (2) Assessing how previous housing system influences the behaviour of cows and their reactivity to milking once they have adapted to their new environment.
Thirty-eight cows from various farms were relocated to a bedded-pack barn on an enrollment basis. Initially, cows remained in their original groups for 10 days before being regrouped into larger groups. Data was collected after arrival via video recordings (2-3 hours, three times per day, three times a week, for 5 weeks) and live observations of milking reactivity at the parlour, conducted twice weekly. In the first case study, we will establish the time-budget of two groups (n=8 & 11 cows) using scan sampling at the group level. In the second case study, the individual time-budgets of a subsample of cows from four groups (n=8-11 per group) will be assessed by scan sampling for two weeks following regrouping, and analysed to explore how previous housing system affects the behaviours of dairy cows once adapted to their environment. As for milking reactivity, preliminary results showed that more tie-stall cows were reactive than free-stall cows on the first and last day of the experiment. Behaviours include kicking (1st day: tie stall=12.5% of cows},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
In response to a 2023 change in the Dairy Code of Practice that prohibits continuous tethering as of 2027, there is a predicted rise in the construction of free-stall and stall-free barns, and more cows transitioning between housing systems. The transition period of dairy cattle after a housing change is of interest from a welfare standpoint as there are behavioural adjustments, and potential stressors that could influence their ability to adapt to a new environment. Therefore, this study aims to understand how the transition towards a new barn is experienced by lactating dairy cows and is divided into 2 case studies: (1) Examining how the transition to a new housing system affects the cows' time-budget and how long it takes to stabilize, and (2) Assessing how previous housing system influences the behaviour of cows and their reactivity to milking once they have adapted to their new environment.
Thirty-eight cows from various farms were relocated to a bedded-pack barn on an enrollment basis. Initially, cows remained in their original groups for 10 days before being regrouped into larger groups. Data was collected after arrival via video recordings (2-3 hours, three times per day, three times a week, for 5 weeks) and live observations of milking reactivity at the parlour, conducted twice weekly. In the first case study, we will establish the time-budget of two groups (n=8 & 11 cows) using scan sampling at the group level. In the second case study, the individual time-budgets of a subsample of cows from four groups (n=8-11 per group) will be assessed by scan sampling for two weeks following regrouping, and analysed to explore how previous housing system affects the behaviours of dairy cows once adapted to their environment. As for milking reactivity, preliminary results showed that more tie-stall cows were reactive than free-stall cows on the first and last day of the experiment. Behaviours include kicking (1st day: tie stall=12.5% of cows
Thirty-eight cows from various farms were relocated to a bedded-pack barn on an enrollment basis. Initially, cows remained in their original groups for 10 days before being regrouped into larger groups. Data was collected after arrival via video recordings (2-3 hours, three times per day, three times a week, for 5 weeks) and live observations of milking reactivity at the parlour, conducted twice weekly. In the first case study, we will establish the time-budget of two groups (n=8 & 11 cows) using scan sampling at the group level. In the second case study, the individual time-budgets of a subsample of cows from four groups (n=8-11 per group) will be assessed by scan sampling for two weeks following regrouping, and analysed to explore how previous housing system affects the behaviours of dairy cows once adapted to their environment. As for milking reactivity, preliminary results showed that more tie-stall cows were reactive than free-stall cows on the first and last day of the experiment. Behaviours include kicking (1st day: tie stall=12.5% of cows
Amarioarei, Georgiana; Cellier, Marjorie; Aigueperse, Nadège; Wolfe, Tania; Montigny, Nicolas; Almeida, Hayda; Shepley, Elise; Diallo, Abdoulay Baniré
Stimulating Minds: Investigating the motivation of calves to engage with cognitive enrichment through participation measures [External] Miscellaneous
2025.
Abstract | Links | BibTeX | Tags: Presentation
@misc{amarioarei_stimulating_2025,
title = {Stimulating Minds: Investigating the motivation of calves to engage with cognitive enrichment through participation measures},
author = {Georgiana Amarioarei and Marjorie Cellier and Nadège Aigueperse and Tania Wolfe and Nicolas Montigny and Hayda Almeida and Elise Shepley and Abdoulay Baniré Diallo},
url = {https://ccsaw.uoguelph.ca/upcoming-events/2025-isae-north-american-regional-meeting/},
year = {2025},
date = {2025-05-01},
urldate = {2025-05-01},
address = {Guelph, Canada},
abstract = {Introducing cognitive enrichment from an early age has the potential to enhance an animal's capacity to learn both simple and complex tasks, promote neural plasticity, and support cognitive development. This is applicable for young cattle who are at a critical stage in their development and could benefit from the influence cognitive enrichment has on their ehavioral expression. This study aims to explore the motivation of weaned dairy calves to engage with cognitive enrichment by analyzing measures of participation over time. Furthermore, we will investigate whether calf performance in the form of measured success rates is a factor influencing the motivation to participate. We hypothesize that providing the opportunity to access cognitive enrichment will have an effect on the calves’ motivation to participate in cognitive challenges and engage with the nrichment. Additionally, we expect there to be a correlation between participation and ability to solve the cognitive tasks. Our study involved a total of four pairs of weaned calves (n=8). The experimental groups were presented with three variations of a puzzle box, each equipped with unique challenges that offer different solutions (push, slide, pull). These boxes were provided to the calves over the span of nine days in an isolated corridor behind their pen. We investigated participation through measuring the frequency to visit the enrichment and we measured the success rates of each calf by processing a predetermined pass/fail criteria of daily performance. Our preliminary results show that on day one of the experiment, the average participation was 2.7 ± 1.4 with a success rate of 58 ± 0.4%. By day nine, participation was 2.6 ± 1.6 with an average success rate of 76 ± 0.4%. These results indicate that there may be no notable change in participation level over time such that the animals are consistently motivated to access enrichment. Moreover, these results allude that there may be a progressive increase in performance success, possibly due to consistent participation. For future analysis, these results will be used in conjunction with the durations and latencies of use, to investigate further potential correlations between participation and success rate.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Introducing cognitive enrichment from an early age has the potential to enhance an animal's capacity to learn both simple and complex tasks, promote neural plasticity, and support cognitive development. This is applicable for young cattle who are at a critical stage in their development and could benefit from the influence cognitive enrichment has on their ehavioral expression. This study aims to explore the motivation of weaned dairy calves to engage with cognitive enrichment by analyzing measures of participation over time. Furthermore, we will investigate whether calf performance in the form of measured success rates is a factor influencing the motivation to participate. We hypothesize that providing the opportunity to access cognitive enrichment will have an effect on the calves’ motivation to participate in cognitive challenges and engage with the nrichment. Additionally, we expect there to be a correlation between participation and ability to solve the cognitive tasks. Our study involved a total of four pairs of weaned calves (n=8). The experimental groups were presented with three variations of a puzzle box, each equipped with unique challenges that offer different solutions (push, slide, pull). These boxes were provided to the calves over the span of nine days in an isolated corridor behind their pen. We investigated participation through measuring the frequency to visit the enrichment and we measured the success rates of each calf by processing a predetermined pass/fail criteria of daily performance. Our preliminary results show that on day one of the experiment, the average participation was 2.7 ± 1.4 with a success rate of 58 ± 0.4%. By day nine, participation was 2.6 ± 1.6 with an average success rate of 76 ± 0.4%. These results indicate that there may be no notable change in participation level over time such that the animals are consistently motivated to access enrichment. Moreover, these results allude that there may be a progressive increase in performance success, possibly due to consistent participation. For future analysis, these results will be used in conjunction with the durations and latencies of use, to investigate further potential correlations between participation and success rate.
Zhu, Junsheng
Student Panel - The Next Generation of AI Researchers [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{zhu_student_2024,
title = {Student Panel - The Next Generation of AI Researchers},
author = {Junsheng Zhu},
url = {https://www.mcgill.ca/cdsi/channels/event/mcgill-collaborative-ai-society-annual-symposium-361276},
year = {2024},
date = {2024-12-01},
urldate = {2024-12-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Naghashi, Vahid; Boukadoum, Mounir; Diallo, Baniré
Transformer-BEATS: A Transformer model for time series prediction of dairy milk production [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{naghashi_transformer-beats_2024,
title = {Transformer-BEATS: A Transformer model for time series prediction of dairy milk production},
author = {Vahid Naghashi and Mounir Boukadoum and Baniré Diallo},
url = {https://f1000research.com/posters/13-1400},
year = {2024},
date = {2024-11-01},
urldate = {2024-11-01},
address = {Medellin, Colombia},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Xu, Aimee; Cellier, Marjorie; Aigueperse, Nadège; Vasseur, Elsa
You've Got A Friend In Me: A Scoping Review on How Personality Traits in Dairy Cows Affects Social Enrichment Response [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{xu_youve_2024,
title = {You've Got A Friend In Me: A Scoping Review on How Personality Traits in Dairy Cows Affects Social Enrichment Response},
author = {Aimee Xu and Marjorie Cellier and Nadège Aigueperse and Elsa Vasseur},
url = {https://www.oplait.org/rvannuel},
year = {2024},
date = {2024-11-01},
urldate = {2024-11-01},
address = {Sherbrooke, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Diallo, Abdoulaye Baniré
Introduction à l’intelligence artificielle [External] Presentation
22.10.2024.
Links | BibTeX | Tags: Presentation
@misc{banire_diallo_introduction_nodate,
title = {Introduction à l’intelligence artificielle},
author = {Abdoulaye Baniré Diallo},
url = {https://event.fourwaves.com/2024rqrsymposium/pages},
year = {2024},
date = {2024-10-22},
urldate = {2024-10-22},
keywords = {Presentation},
pubstate = {published},
tppubtype = {presentation}
}
Diallo, Abdoulaye Baniré
Introduction à l’intelligence artificielle [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{banire_diallo_introduction_2024,
title = {Introduction à l’intelligence artificielle},
author = {Abdoulaye Baniré Diallo},
url = {https://event.fourwaves.com/2024rqrsymposium/pages},
year = {2024},
date = {2024-10-01},
urldate = {2024-10-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Massoua, Armand Bandiang
Using deep learning techniques for survival analyses [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{bandiang_massoua_using_2024,
title = {Using deep learning techniques for survival analyses},
author = {Armand Bandiang Massoua},
url = {https://event.fourwaves.com/2024rqrsymposium/pages},
year = {2024},
date = {2024-10-01},
urldate = {2024-10-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Naghashi, Vahid
Introduction to Deep Learning techniques with use case on dairy data [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{naghashi_introduction_2024,
title = {Introduction to Deep Learning techniques with use case on dairy data},
author = {Vahid Naghashi},
url = {https://event.fourwaves.com/2024rqrsymposium/pages},
year = {2024},
date = {2024-10-01},
urldate = {2024-10-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Xu, Aimee; Cellier, Marjorie; Aigueperse, Nadège; Vasseur, Elsa
Je suis ton amie : Identification des traits de personnalité chez les vaches laitières pour un regroupement optimisé comme forme d’enrichissement social [External] Miscellaneous
2024.
Abstract | Links | BibTeX | Tags: Presentation
@misc{xu_je_2024,
title = {Je suis ton amie : Identification des traits de personnalité chez les vaches laitières pour un regroupement optimisé comme forme d’enrichissement social},
author = {Aimee Xu and Marjorie Cellier and Nadège Aigueperse and Elsa Vasseur},
url = {https://www.youtube.com/watch?v=K5nIlA2I-Kc},
year = {2024},
date = {2024-10-01},
urldate = {2024-10-01},
address = {Drummondville, Canada},
abstract = {Dans de nombreuses fermes laitières canadiennes, les vaches sont regroupées par âge ou race, des critères pratiques, mais qui ne garantissent pas toujours l’harmonie. Tout comme les humains, ces critères ne garantissent pas une bonne entente entre les vaches. Notre projet explore le regroupement des vaches selon leur personnalité pour améliorer leurs liens sociaux et réduire les conflits. La personnalité, définie comme des différences comportementales stables au fil du temps et dans différents contextes joue un rôle clé : des études montrent que les vaches aux personnalités similaires s’entendent mieux, réduisant le stress et améliorant leur bien-être. Notre objectif est donc double : 1) valider les tests comportementaux appliqués aux vaches logées dans les stabulations entravées pour mesurer la personnalité; 2) déterminer des profils de personnalité afin de mieux comprendre les dynamiques sociales au sein du troupeau. Plusieurs tests comportementaux dont le Test d'Approche Humaine (HAT) et le Test de Soudaineté (SUD) ont été réalisés à trois périodes différentes, sur environ 70 vaches par période (HAT : n=44, SUD : n=48). Nous avons observé la cohérence des comportements des vaches à travers le temps et entre les tests. Nos résultats montrent une forte corrélation entre le HAT et le SUD sur les trois périodes, indiquant une réponse comportementale stable dans divers contextes. Cela apporte une information importante sur la validité des tests pour mesurer la personnalité chez les vaches laitières, en renforçant l’idée que ces deux tests mesurent un aspect partagé de la personnalité. Cependant, la réactivité aux stimuli soudains (SUD) semble plus stable que celle aux interactions humaines (HAT) du fait que ce dernier test mesure aussi l’affinité à l’humain qui peut être modifiée par l’expérience. Ces résultats sont un premier pas permettant de confirmer les tests classiquement utilisés pour mesurer la personnalité des animaux, en particulier la réactivité générale. L’implémentation et la validation d’autres tests complémentaires portant sur des traits de personnalité comme l’agressivité ou la curiosité sont nécessaires pour obtenir des profils plus complets. Cela permettra d’optimiser le regroupement des vaches, améliorant ainsi leur bien-être et pourrait avoir un impact sur leur santé et productivité, tout en répondant aux attentes sociétales pour des pratiques agricoles plus durables.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Dans de nombreuses fermes laitières canadiennes, les vaches sont regroupées par âge ou race, des critères pratiques, mais qui ne garantissent pas toujours l’harmonie. Tout comme les humains, ces critères ne garantissent pas une bonne entente entre les vaches. Notre projet explore le regroupement des vaches selon leur personnalité pour améliorer leurs liens sociaux et réduire les conflits. La personnalité, définie comme des différences comportementales stables au fil du temps et dans différents contextes joue un rôle clé : des études montrent que les vaches aux personnalités similaires s’entendent mieux, réduisant le stress et améliorant leur bien-être. Notre objectif est donc double : 1) valider les tests comportementaux appliqués aux vaches logées dans les stabulations entravées pour mesurer la personnalité; 2) déterminer des profils de personnalité afin de mieux comprendre les dynamiques sociales au sein du troupeau. Plusieurs tests comportementaux dont le Test d'Approche Humaine (HAT) et le Test de Soudaineté (SUD) ont été réalisés à trois périodes différentes, sur environ 70 vaches par période (HAT : n=44, SUD : n=48). Nous avons observé la cohérence des comportements des vaches à travers le temps et entre les tests. Nos résultats montrent une forte corrélation entre le HAT et le SUD sur les trois périodes, indiquant une réponse comportementale stable dans divers contextes. Cela apporte une information importante sur la validité des tests pour mesurer la personnalité chez les vaches laitières, en renforçant l’idée que ces deux tests mesurent un aspect partagé de la personnalité. Cependant, la réactivité aux stimuli soudains (SUD) semble plus stable que celle aux interactions humaines (HAT) du fait que ce dernier test mesure aussi l’affinité à l’humain qui peut être modifiée par l’expérience. Ces résultats sont un premier pas permettant de confirmer les tests classiquement utilisés pour mesurer la personnalité des animaux, en particulier la réactivité générale. L’implémentation et la validation d’autres tests complémentaires portant sur des traits de personnalité comme l’agressivité ou la curiosité sont nécessaires pour obtenir des profils plus complets. Cela permettra d’optimiser le regroupement des vaches, améliorant ainsi leur bien-être et pourrait avoir un impact sur leur santé et productivité, tout en répondant aux attentes sociétales pour des pratiques agricoles plus durables.
Vasseur, Elsa
Provision of automatic milking-system alerts for potential incidence of mastitis: a comparative study between stationary classification and time-series regression modelling [External] Presentation
Bologna, Italy, 01.09.2024.
Links | BibTeX | Tags: Presentation
@misc{vasseur_provision_2024-1,
title = {Provision of automatic milking-system alerts for potential incidence of mastitis: a comparative study between stationary classification and time-series regression modelling},
author = {Elsa Vasseur},
url = {https://www.ecplf2024.it/program/parallel-sessions-11-45-13-00/},
year = {2024},
date = {2024-09-01},
urldate = {2024-09-01},
address = {Bologna, Italy},
keywords = {Presentation},
pubstate = {published},
tppubtype = {presentation}
}
Vasseur, Elsa
Provision of automatic milking-system alerts for potential incidence of mastitis: a comparative study between stationary classification and time-series regression modelling [External] Presentation
Bologna, Italy, 01.09.2024.
Links | BibTeX | Tags: Presentation
@misc{vasseur_provision_2024,
title = {Provision of automatic milking-system alerts for potential incidence of mastitis: a comparative study between stationary classification and time-series regression modelling},
author = {Elsa Vasseur},
url = {https://www.ecplf2024.it/program/parallel-sessions-11-45-13-00/},
year = {2024},
date = {2024-09-01},
urldate = {2024-09-01},
address = {Bologna, Italy},
keywords = {Presentation},
pubstate = {published},
tppubtype = {presentation}
}
Samaké, Awa; Diallo, Abdoulaye Baniré
Predicting Dairy Milk Yields Using Privileged Information and Heteroscedastic Dropout [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{samake_predicting_2024,
title = {Predicting Dairy Milk Yields Using Privileged Information and Heteroscedastic Dropout},
author = {Awa Samaké and Abdoulaye Baniré Diallo},
url = {https://wscv-indaba.github.io/2024#program},
year = {2024},
date = {2024-09-01},
urldate = {2024-09-01},
address = {Dakar, Senegal},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Samaké, Awa; Diallo, Abdoulaye Baniré
Predicting Dairy Milk Yields Using Privileged Information and Heteroscedastic Dropout [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{samake_predicting_2024-1,
title = {Predicting Dairy Milk Yields Using Privileged Information and Heteroscedastic Dropout},
author = {Awa Samaké and Abdoulaye Baniré Diallo},
url = {https://deeplearningindaba.com/2024/},
year = {2024},
date = {2024-09-01},
urldate = {2024-09-01},
address = {Dakar, Senegal},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Almeida, H.; Grégoire, N. B.; Bilal, M.; Leduc, M.; Chorif, Y.; Zhao, X.; Dubuc, J.; Diallo, A. B.
Biomarker-based learning for disease prediction in precision dairy farming [External] Presentation
Montreal, Canada, 01.07.2024.
Links | BibTeX | Tags: Presentation
@misc{almeida_biomarker-based_2024-1,
title = {Biomarker-based learning for disease prediction in precision dairy farming},
author = {H. Almeida and N. B. Grégoire and M. Bilal and M. Leduc and Y. Chorif and X. Zhao and J. Dubuc and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/home},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {presentation}
}
Toure, François Gonothi; Diallo, Abdoulaye Baniré; Boukadoum, Mounir; Araujo,
Optimizing Dairy Cow Health and Productivity with Efficient Movement Monitoring using Data Augmentation and Computer Vision [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{gonothi_toure_optimizing_2024,
title = {Optimizing Dairy Cow Health and Productivity with Efficient Movement Monitoring using Data Augmentation and Computer Vision},
author = {François Gonothi Toure and Abdoulaye Baniré Diallo and Mounir Boukadoum and Araujo},
url = {https://blackscientists.ca/be-stemm-event/2024-en-stimm/},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Ottawa, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Orchi, H.; Diallo, A. B.; Vasseur, E.; Elbiaze, H.; Sabir, E.; Sadik, M.
Temporal Synchronization of Multi-View Video for Cattle Movement Analysis in Dynamic Farm Settings [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{orchi_temporal_2024,
title = {Temporal Synchronization of Multi-View Video for Cattle Movement Analysis in Dynamic Farm Settings},
author = {H. Orchi and A. B. Diallo and E. Vasseur and H. Elbiaze and E. Sabir and M. Sadik},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Naghashi, V.; Boukadoum, M.; Diallo, A. B.
Empowering Dairy Farmers: A Transformer-Based Framework for Informed Decision Making in Dairy Agriculture [External] Miscellaneous
2024.
Abstract | Links | BibTeX | Tags: Presentation
@misc{naghashi_empowering_2024,
title = {Empowering Dairy Farmers: A Transformer-Based Framework for Informed Decision Making in Dairy Agriculture},
author = {V. Naghashi and M. Boukadoum and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
abstract = {In precision livestock, the decision of animal replacement requires an estimation of the lifetime profit of the animal based on multiple factors and operational conditions. In dairy farms, this can be associated with the milk income corresponding to milk production, health condition and herd management costs, which in turn may be a function of other factors including genetics and weather conditions. Estimating the cumulative income from a cow's milk production can be posed as a multivariate time-series prediction task where a late-milk income of a cow can be predicted based on early dairy factors recorded in a sequence of time-steps (lactation months). Furthermore, the predicted milk income would serve as an input to a decision making system for deciding whether to keep or remove an animal in the next lactation period. In this work, a Transformer based model is proposed to predict the cumulative dairy income over the incoming lactation period and further a recommendation procedure is devised for decision making in order to reduce the farmers cost and save their time. In the Transformer, both temporal and inter-variable correlations are captured thanks to the temporal and spatial multi-head attention modules. The proposed framework is assessed using 47749 dairy cows corresponding to more than 5000 herds and the results are compared with the other state-of-the-art models. Our Transformer model outperforms the previous baselines and provides a promising prediction performance with the highest accuracy of 76%, opening the way of better resource management in the dairy industry.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
In precision livestock, the decision of animal replacement requires an estimation of the lifetime profit of the animal based on multiple factors and operational conditions. In dairy farms, this can be associated with the milk income corresponding to milk production, health condition and herd management costs, which in turn may be a function of other factors including genetics and weather conditions. Estimating the cumulative income from a cow's milk production can be posed as a multivariate time-series prediction task where a late-milk income of a cow can be predicted based on early dairy factors recorded in a sequence of time-steps (lactation months). Furthermore, the predicted milk income would serve as an input to a decision making system for deciding whether to keep or remove an animal in the next lactation period. In this work, a Transformer based model is proposed to predict the cumulative dairy income over the incoming lactation period and further a recommendation procedure is devised for decision making in order to reduce the farmers cost and save their time. In the Transformer, both temporal and inter-variable correlations are captured thanks to the temporal and spatial multi-head attention modules. The proposed framework is assessed using 47749 dairy cows corresponding to more than 5000 herds and the results are compared with the other state-of-the-art models. Our Transformer model outperforms the previous baselines and provides a promising prediction performance with the highest accuracy of 76%, opening the way of better resource management in the dairy industry.
Gisiger, T.; Cellier, M.; Vasseur, E.; Diallo, A. B.
Extracting meaningful video segments using a movement detection algorithm applied to dairy cow behavior study and welfare monitoring. [External] Miscellaneous
2024.
Abstract | Links | BibTeX | Tags: Presentation
@misc{gisiger_extracting_2024,
title = {Extracting meaningful video segments using a movement detection algorithm applied to dairy cow behavior study and welfare monitoring.},
author = {T. Gisiger and M. Cellier and E. Vasseur and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
abstract = {Precision dairy farming is essential to creating a food production system that is durable and respects animal welfare and the environment.
This approach requires gathering many hours of videos with still cameras, which are then used for research, welfare monitoring and tool training. Extracting the video segments with the most meaningful information would allow us to study larger fractions of the recordings taken while gathering the maximum number of observations. This can be framed as a movement detection problem, or, alternatively, a detection problem using traditional or deep learning techniques. However, the latter process of training in a cluttered farm environment might prove challenging.
Here, we propose an algorithm that estimates cow movement in a robust manner without the need for object detection or training. The resulting movement indices, paired with an independently set movement threshold, can then be used to partition videos into episodes where the cow is either immobile or displaying relevant movements and behaviours. This approach takes advantage of typical cow behaviour features and allows for factoring out video sections with repetitive or little/no movement.
The experimental setting consists of five 15-minute videos and focuses on measuring the extent to which discarding episodes with little to no movements speeds up the process of labelling behaviours by animal science experts.
This approach will allow for more complex experiments, novel angles of investigation and larger data-sets to study cow behaviour and interaction with their environment as well as monitoring for welfare status.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Precision dairy farming is essential to creating a food production system that is durable and respects animal welfare and the environment.
This approach requires gathering many hours of videos with still cameras, which are then used for research, welfare monitoring and tool training. Extracting the video segments with the most meaningful information would allow us to study larger fractions of the recordings taken while gathering the maximum number of observations. This can be framed as a movement detection problem, or, alternatively, a detection problem using traditional or deep learning techniques. However, the latter process of training in a cluttered farm environment might prove challenging.
Here, we propose an algorithm that estimates cow movement in a robust manner without the need for object detection or training. The resulting movement indices, paired with an independently set movement threshold, can then be used to partition videos into episodes where the cow is either immobile or displaying relevant movements and behaviours. This approach takes advantage of typical cow behaviour features and allows for factoring out video sections with repetitive or little/no movement.
The experimental setting consists of five 15-minute videos and focuses on measuring the extent to which discarding episodes with little to no movements speeds up the process of labelling behaviours by animal science experts.
This approach will allow for more complex experiments, novel angles of investigation and larger data-sets to study cow behaviour and interaction with their environment as well as monitoring for welfare status.
This approach requires gathering many hours of videos with still cameras, which are then used for research, welfare monitoring and tool training. Extracting the video segments with the most meaningful information would allow us to study larger fractions of the recordings taken while gathering the maximum number of observations. This can be framed as a movement detection problem, or, alternatively, a detection problem using traditional or deep learning techniques. However, the latter process of training in a cluttered farm environment might prove challenging.
Here, we propose an algorithm that estimates cow movement in a robust manner without the need for object detection or training. The resulting movement indices, paired with an independently set movement threshold, can then be used to partition videos into episodes where the cow is either immobile or displaying relevant movements and behaviours. This approach takes advantage of typical cow behaviour features and allows for factoring out video sections with repetitive or little/no movement.
The experimental setting consists of five 15-minute videos and focuses on measuring the extent to which discarding episodes with little to no movements speeds up the process of labelling behaviours by animal science experts.
This approach will allow for more complex experiments, novel angles of investigation and larger data-sets to study cow behaviour and interaction with their environment as well as monitoring for welfare status.
Araujo, V. M. De; Gisiger, T.; Gambs, S.; Vasseur, E.; Diallo, A. B.
Revolutionizing Livestock Monitoring: AI-Powered Cow Detection in Farm Environments [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{de_araujo_revolutionizing_2024,
title = {Revolutionizing Livestock Monitoring: AI-Powered Cow Detection in Farm Environments},
author = {V. M. De Araujo and T. Gisiger and S. Gambs and E. Vasseur and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Jacques, A. A. Boatswain; Lord, E.; Reinhartz, V.; Diallo, A. B.
Precision Farming for Profit: Leveraging Profitability Maps and ILPMZ to Optimize Return on Investment and Soil Conservation of Agricultural Fields [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{boatswain_jacques_precision_2024,
title = {Precision Farming for Profit: Leveraging Profitability Maps and ILPMZ to Optimize Return on Investment and Soil Conservation of Agricultural Fields},
author = {A. A. Boatswain Jacques and E. Lord and V. Reinhartz and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Almeida, H.; Grégoire, N. B.; Bilal, M.; Leduc, M.; Chorif, Y.; Zhao, X.; Dubuc, J.; Diallo, A. B.
Biomarker-based learning for disease prediction in precision dairy farming [External] Miscellaneous
2024.
Abstract | Links | BibTeX | Tags: Presentation
@misc{almeida_biomarker-based_2024,
title = {Biomarker-based learning for disease prediction in precision dairy farming},
author = {H. Almeida and N. B. Grégoire and M. Bilal and M. Leduc and Y. Chorif and X. Zhao and J. Dubuc and A. B. Diallo},
url = {https://www.iscb.org/ismb2024/programme-schedule/scientific-programme/digag},
year = {2024},
date = {2024-07-01},
urldate = {2024-07-01},
address = {Montreal, Canada},
abstract = {Metabolic diseases have great impact on dairy production and animal welfare [1, 2]. Metabolomic profiling has helped identify biomarkers to predict disease risk in dairy cows [3, 4]. Previous studies tend to overlook other biomarkers, like from milk production, which could help predict diseases in cows [5]. Our ensemble learner supports predicting disease risk based on heterogeneous biomarkers from metabolomic and health profiles, milk production history, and herd history.
Our datasets contain biomarkers for over 13,700 health events of 1,200 cows from 50 dairy farms in Canada. Biomarkers are captured for a health event e at timepoint t. Given an upcoming lactation Ln, base predictions are obtained for all health events et occurring during lactation Ln−1.
The ensemble learner averages base predictions for an animal and outputs disease probabilities for lactation Ln. Binary classes are disease or non-disease, based on a curated set of nine most common diseases in dairy cows. Classification performance was evaluated for multiple combinations of biomarkers and classifiers.
Classification models based on Logistic Regression and Random Forest classifiers yield best performances, with an average of 0.6 and 0.77 F-measure for disease and non-disease respectively.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Metabolic diseases have great impact on dairy production and animal welfare [1, 2]. Metabolomic profiling has helped identify biomarkers to predict disease risk in dairy cows [3, 4]. Previous studies tend to overlook other biomarkers, like from milk production, which could help predict diseases in cows [5]. Our ensemble learner supports predicting disease risk based on heterogeneous biomarkers from metabolomic and health profiles, milk production history, and herd history.
Our datasets contain biomarkers for over 13,700 health events of 1,200 cows from 50 dairy farms in Canada. Biomarkers are captured for a health event e at timepoint t. Given an upcoming lactation Ln, base predictions are obtained for all health events et occurring during lactation Ln−1.
The ensemble learner averages base predictions for an animal and outputs disease probabilities for lactation Ln. Binary classes are disease or non-disease, based on a curated set of nine most common diseases in dairy cows. Classification performance was evaluated for multiple combinations of biomarkers and classifiers.
Classification models based on Logistic Regression and Random Forest classifiers yield best performances, with an average of 0.6 and 0.77 F-measure for disease and non-disease respectively.
Our datasets contain biomarkers for over 13,700 health events of 1,200 cows from 50 dairy farms in Canada. Biomarkers are captured for a health event e at timepoint t. Given an upcoming lactation Ln, base predictions are obtained for all health events et occurring during lactation Ln−1.
The ensemble learner averages base predictions for an animal and outputs disease probabilities for lactation Ln. Binary classes are disease or non-disease, based on a curated set of nine most common diseases in dairy cows. Classification performance was evaluated for multiple combinations of biomarkers and classifiers.
Classification models based on Logistic Regression and Random Forest classifiers yield best performances, with an average of 0.6 and 0.77 F-measure for disease and non-disease respectively.
Diallo, A. B.; Theriault, P.; Hernandez, I.; Savard, M.; Blanc, C.
Panelist: Quelle agriculture de demain avec les technologies d'aujourd'hui ? [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{diallo_panelist_2024,
title = {Panelist: Quelle agriculture de demain avec les technologies d'aujourd'hui ?},
author = {A. B. Diallo and P. Theriault and I. Hernandez and M. Savard and C. Blanc},
url = {https://crimevent.zohobackstage.com/LesrencontresdelcosystmeClimatetIA#/?day=1},
year = {2024},
date = {2024-05-01},
urldate = {2024-05-01},
address = {Montreal, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Cellier, M.; Aigueperse, N.; Shepley, E.; Villetaz-Robichaud, M.; Vasseur, E.
Cow-conception - Intégrer l'expérience de l'animal lors de la conception du logement ou de nouvelles pratiques pour maximiser la réponse animale et l'expérience utilisateur (producteur) [External] Presentation
Rennes, France, 01.02.2024.
Abstract | Links | BibTeX | Tags: Presentation
@misc{cellier_cow-conception_2024-1,
title = {Cow-conception - Intégrer l'expérience de l'animal lors de la conception du logement ou de nouvelles pratiques pour maximiser la réponse animale et l'expérience utilisateur (producteur)},
author = {M. Cellier and N. Aigueperse and E. Shepley and M. Villetaz-Robichaud and E. Vasseur},
url = {https://idele.fr/rmt-one-welfare/?eID=cmis_download&oID=workspace%3A%2F%2FSpacesStore%2F46feb36a-548e-47dd-832d-088b886735db&cHash=669b6af9671fa7c4520d841a163c6d0e},
year = {2024},
date = {2024-02-01},
urldate = {2024-02-01},
address = {Rennes, France},
abstract = {https://idele.fr/rmt-one-welfare/publications/detail-article?tx_atolidelecontenus_publicationdetail%5Baction%5D=showArticle&tx_atolidelecontenus_publicationdetail%5Bcontroller%5D=Detail&tx_atolidelecontenus_publicationdetail%5Bpublication%5D=20574&cHash=82ad5b4ec8b6e5e22b85e6e032e37133},
keywords = {Presentation},
pubstate = {published},
tppubtype = {presentation}
}
https://idele.fr/rmt-one-welfare/publications/detail-article?tx_atolidelecontenus_publicationdetail%5Baction%5D=showArticle&tx_atolidelecontenus_publicationdetail%5Bcontroller%5D=Detail&tx_atolidelecontenus_publicationdetail%5Bpublication%5D=20574&cHash=82ad5b4ec8b6e5e22b85e6e032e37133
Cellier, M.; Aigueperse, N.; Shepley, E.; Villetaz-Robichaud, M.; Vasseur, E.
Cow-conception - Intégrer l'expérience de l'animal lors de la conception du logement ou de nouvelles pratiques pour maximiser la réponse animale et l'expérience utilisateur (producteur) [External] Miscellaneous
2024.
Links | BibTeX | Tags: Presentation
@misc{cellier_cow-conception_2024,
title = {Cow-conception - Intégrer l'expérience de l'animal lors de la conception du logement ou de nouvelles pratiques pour maximiser la réponse animale et l'expérience utilisateur (producteur)},
author = {M. Cellier and N. Aigueperse and E. Shepley and M. Villetaz-Robichaud and E. Vasseur},
url = {https://idele.fr/rmt-one-welfare/?eID=cmis_download&oID=workspace%3A%2F%2FSpacesStore%2F46feb36a-548e-47dd-832d-088b886735db&cHash=669b6af9671fa7c4520d841a163c6d0e},
year = {2024},
date = {2024-02-01},
urldate = {2024-01-01},
address = {Rennes, France},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Vasseur, Elsa; Diallo, Abdoulaye Banire
Capsule vidéo Well-E : Intelligence artificielle au service du bien-être des bovins laitiers [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{vasseur_capsule_2023,
title = {Capsule vidéo Well-E : Intelligence artificielle au service du bien-être des bovins laitiers},
author = {Elsa Vasseur and Abdoulaye Banire Diallo},
url = {https://www.craaq.qc.ca/Evenements-du-CRAAQ/symposium-sur-les-bovins-laitiers-2023/e/2646#tab_tab-en-detail},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Drummondville, Quebec},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Cellier, M.; Aigueperse, N.; Vasseur, E.
Alors, on sort? Caractérisation de l'anticipation des vaches laitières pour un accès à l'extérieure [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{cellier_alors_2023,
title = {Alors, on sort? Caractérisation de l'anticipation des vaches laitières pour un accès à l'extérieure},
author = {M. Cellier and N. Aigueperse and E. Vasseur},
url = {https://www.oplait.org/activites-galactinnov},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Quebec City, Canada},
abstract = {L'accès régulier à une aire d’exercice extérieure peut être vu comme un enrichissement, mais il n’est pas clair comment les vaches le perçoivent. Notre étude vise à caractériser les comportements d'anticipation de vaches avec opportunités de mouvement limitées quand un accès à une aire d’exercice extérieure leur est offert. Au cours de 2 essais indépendants, 32 vaches ont été divisées en 2 groupes : les contrôles (C; n=8 pour l’essai 1 (E1) et l’essai 2 (E2)) restant à l'intérieur, et les traitements (T; n=8 pour E1 et E2) sortant à l'extérieur 5j/sem pendant 8 semaines. Les sorties combinaient différentes durées de sortie et superficies d’aire. Des indices de sortie (e.g., préparation des allées) ont été mis en place 20min avant que les vaches ne sortent. Des enregistrements vidéo ont été faits pendant cette période pour relever les comportements liés à l’activité générale (ex : manger, boire), les comportements anticipatoires (ex : regarder la porte, piétiner) ainsi que les positions des oreilles (ex : en avant). Le nombre de changements comportementaux et de positions d’oreilles a également été calculé. Une analyse en composante principales a ensuite été réalisée sur les fréquences et les durées des comportements. Pour E1 et E2, les deux mêmes dimensions ont été trouvées : la dimension 1 opposant le repos à l’agitation, et la dimension 2 opposant la vigilance et le maintien de l’activité habituelle. Lors de E1, les C étaient plutôt au repos et maintenaient leurs activités habituelles tandis que les T étaient plutôt agitées et vigilantes. Au contraire, lors de E2, les C étaient plutôt agités (et maintenaient leurs activités habituelles) alors que les T étaient plutôt au repos (et vigilantes). Étudier des périodes ciblées (ex : les 5 premières minutes après les indices de sortie) ou utiliser d’autres types d’analyse (ex : les séries temporelles) pourrait permettre de mieux détailler et comprendre l'anticipation. Combinée à des indicateurs comme la motivation, elle participerait à évaluer la valence d’un enrichissement pour les vaches.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
L'accès régulier à une aire d’exercice extérieure peut être vu comme un enrichissement, mais il n’est pas clair comment les vaches le perçoivent. Notre étude vise à caractériser les comportements d'anticipation de vaches avec opportunités de mouvement limitées quand un accès à une aire d’exercice extérieure leur est offert. Au cours de 2 essais indépendants, 32 vaches ont été divisées en 2 groupes : les contrôles (C; n=8 pour l’essai 1 (E1) et l’essai 2 (E2)) restant à l'intérieur, et les traitements (T; n=8 pour E1 et E2) sortant à l'extérieur 5j/sem pendant 8 semaines. Les sorties combinaient différentes durées de sortie et superficies d’aire. Des indices de sortie (e.g., préparation des allées) ont été mis en place 20min avant que les vaches ne sortent. Des enregistrements vidéo ont été faits pendant cette période pour relever les comportements liés à l’activité générale (ex : manger, boire), les comportements anticipatoires (ex : regarder la porte, piétiner) ainsi que les positions des oreilles (ex : en avant). Le nombre de changements comportementaux et de positions d’oreilles a également été calculé. Une analyse en composante principales a ensuite été réalisée sur les fréquences et les durées des comportements. Pour E1 et E2, les deux mêmes dimensions ont été trouvées : la dimension 1 opposant le repos à l’agitation, et la dimension 2 opposant la vigilance et le maintien de l’activité habituelle. Lors de E1, les C étaient plutôt au repos et maintenaient leurs activités habituelles tandis que les T étaient plutôt agitées et vigilantes. Au contraire, lors de E2, les C étaient plutôt agités (et maintenaient leurs activités habituelles) alors que les T étaient plutôt au repos (et vigilantes). Étudier des périodes ciblées (ex : les 5 premières minutes après les indices de sortie) ou utiliser d’autres types d’analyse (ex : les séries temporelles) pourrait permettre de mieux détailler et comprendre l'anticipation. Combinée à des indicateurs comme la motivation, elle participerait à évaluer la valence d’un enrichissement pour les vaches.
Muszik, Jasmine; Cellier, M.; Aigueperse, N.; Vasseur, E.
Émotions Animales : Un Regard Approfondi sur la Motivation, l'Anticipation, et la Frustration [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{muszik_emotions_2023,
title = {Émotions Animales : Un Regard Approfondi sur la Motivation, l'Anticipation, et la Frustration},
author = {Jasmine Muszik and M. Cellier and N. Aigueperse and E. Vasseur},
url = {https://www.oplait.org/activites-galactinnov},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Quebec City, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Julliot, C.; Dallago, G. M.; Nejati, Amir; Diallo, A. B.; Vasseur, E.
Décrypter les secrets de la démarche : comment la cinématique peut prédire les scores de démarche chez la vache laitière [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{julliot_decrypter_2023,
title = {Décrypter les secrets de la démarche : comment la cinématique peut prédire les scores de démarche chez la vache laitière},
author = {C. Julliot and G. M. Dallago and Amir Nejati and A. B. Diallo and E. Vasseur},
url = {https://www.oplait.org/activites-galactinnov},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Quebec CIty, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Julliot, C.; Dallago, G. M.; Nejati, A.; Diallo, A. B.; Vasseur, E.
Utiliser la cinématique pour quantifier les attributs de la démarche et prédire un score de démarche chez les vaches laitières [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{julliot_utiliser_2023,
title = {Utiliser la cinématique pour quantifier les attributs de la démarche et prédire un score de démarche chez les vaches laitières},
author = {C. Julliot and G. M. Dallago and A. Nejati and A. B. Diallo and E. Vasseur},
url = {https://www.oplait.org/activites-galactinnov},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Quebec City, Canada},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Vliet, Rachel; Wolfe, Tania; Arsenault, Sandrine
WELL-E Posters [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{van_vliet_well-e_2023,
title = {WELL-E Posters},
author = {Rachel Vliet and Tania Wolfe and Sandrine Arsenault},
url = {https://www.craaq.qc.ca/Evenements-du-CRAAQ/symposium-sur-les-bovins-laitiers-2023/e/2646#tab_tab-en-detail},
year = {2023},
date = {2023-10-01},
urldate = {2023-10-01},
address = {Drummondville, Quebec},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Zhu, J.; Lacroix, R.; Wade, K.
Automated extraction of domain knowledge for transition-cow management [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{zhu_automated_2023,
title = {Automated extraction of domain knowledge for transition-cow management},
author = {J. Zhu and R. Lacroix and K. Wade},
url = {https://www.adsa.org/Portals/0/SiteContent/Docs/Meetings/2023ADSA/Abstracts_BOOK_2023.pdf},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Ottawa, Canada},
abstract = {In dairy cattle, the transition period (±3 weeks from calving) represents a challenging time for management. Vast changes in a cow’s physiology, housing, and feeding often result in metabolic or reproductive diseases, leading to a drop in production. Because most metabolic processes are intricately linked, dairy producers and their advisors may have difficulty drawing concise conclusions concerning transition-cow management. To help in this, machine-learning techniques and knowledge-graph theory were explored with a view to creating a decision-support system that could provide producers and their advisors with knowledge from domain literature. Specifically, knowledge was modeled as entities and relationships in knowledge-graph theory, and natural language models were developed to extract information as knowledge graphs. A data set comprising 1,152 sentences from 20 papers was created and split into 922 sentences for training and 230 sentences for testing. Two deep-learning models were trained to extract entities and relationships, respectively. For testing, a bi-lstm model was applied to the entity extraction task and obtained an F1 score of 80%. With regard to relationship extraction, a transformer-based model was deployed but yielded a low F1 of 23%. Therefore, a pre-trained transformer model with 80.8% accuracy was deployed. After feeding the domain literature into the deep-learning models, a knowledge graph of 1,576 nodes and 3,456 edges was constructed and stored in a Neo4j graph database. Subsequently, a semantic parsing method was used to allow users to query the knowledge graph using natural language. To determine the quality of the responses, answers were sampled and evaluated based on human evaluation. On average, the answers scored 7.5 out of 10 and proved informative with respect to the original literature. Although the final interactive results demonstrated a high degree of visualization and scalability, this study primarily sought to demonstrate its feasibility. For tailored industrial applications, further improvements could be implemented in specific knowledge-graph expansion and reasoning.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
In dairy cattle, the transition period (±3 weeks from calving) represents a challenging time for management. Vast changes in a cow’s physiology, housing, and feeding often result in metabolic or reproductive diseases, leading to a drop in production. Because most metabolic processes are intricately linked, dairy producers and their advisors may have difficulty drawing concise conclusions concerning transition-cow management. To help in this, machine-learning techniques and knowledge-graph theory were explored with a view to creating a decision-support system that could provide producers and their advisors with knowledge from domain literature. Specifically, knowledge was modeled as entities and relationships in knowledge-graph theory, and natural language models were developed to extract information as knowledge graphs. A data set comprising 1,152 sentences from 20 papers was created and split into 922 sentences for training and 230 sentences for testing. Two deep-learning models were trained to extract entities and relationships, respectively. For testing, a bi-lstm model was applied to the entity extraction task and obtained an F1 score of 80%. With regard to relationship extraction, a transformer-based model was deployed but yielded a low F1 of 23%. Therefore, a pre-trained transformer model with 80.8% accuracy was deployed. After feeding the domain literature into the deep-learning models, a knowledge graph of 1,576 nodes and 3,456 edges was constructed and stored in a Neo4j graph database. Subsequently, a semantic parsing method was used to allow users to query the knowledge graph using natural language. To determine the quality of the responses, answers were sampled and evaluated based on human evaluation. On average, the answers scored 7.5 out of 10 and proved informative with respect to the original literature. Although the final interactive results demonstrated a high degree of visualization and scalability, this study primarily sought to demonstrate its feasibility. For tailored industrial applications, further improvements could be implemented in specific knowledge-graph expansion and reasoning.
Mokhtarnazif, S.; Shepley, E.; Nejati, A.; Dallago, G. M.; Vasseur, E.
The effect of limited outdoor access frequencies on gait score, hoof lesions and hoof surface temperature of non-clinically lame cows housed in a movement-restricted environment. [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{mokhtarnazif_effect_2023,
title = {The effect of limited outdoor access frequencies on gait score, hoof lesions and hoof surface temperature of non-clinically lame cows housed in a movement-restricted environment.},
author = {S. Mokhtarnazif and E. Shepley and A. Nejati and G. M. Dallago and E. Vasseur},
url = {https://www.adsa.org/Portals/0/SiteContent/Docs/Meetings/2023ADSA/Abstracts_BOOK_2023.pdf},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Ottawa, Canada},
abstract = {Providing outdoor access is beneficial to the welfare of cows housed in restricted movement environments, but its daily implementation is challenging. The study objective was to evaluate the effect of low outdoor access frequency on gait and hoof health of non-clinically lame cows. Thirty-six lactating, tie-stall housed Holstein cows were blocked by parity and days in milk (6 blocks) and provided 1h of outdoor access in 2 treatments (1 d/wk or 3 d/wk) for 5 weeks (Nov-Dec 2022). Data were collected before (pre-trial), after (post-trial) and 8 weeks after (follow-up) outdoor provision. We assessed gait score using a 5-point visual scoring system and clinical claw lesion incidence in the trimming chute. For subclinical signs of lesions, the average hoof and claw surface temperatures were measured from the dorsal view of the coronary bands using an infrared thermography camera. Data were analyzed using mixed effect models with block, period, treatment, and the period × treatment interaction as fixed effects (α < 0.05), and either cow or claw nested within cow as random effect. The ambient and cow’s eye maximum temperature were added as covariates for analyzing thermography data. Average gait scores at pre-trial, post-trial, and follow-up were 2.13 ± 0.55, 1.98 ± 0.61 and 2.02 ± 0.51, respectively. Only non-severe lesions (sole},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Providing outdoor access is beneficial to the welfare of cows housed in restricted movement environments, but its daily implementation is challenging. The study objective was to evaluate the effect of low outdoor access frequency on gait and hoof health of non-clinically lame cows. Thirty-six lactating, tie-stall housed Holstein cows were blocked by parity and days in milk (6 blocks) and provided 1h of outdoor access in 2 treatments (1 d/wk or 3 d/wk) for 5 weeks (Nov-Dec 2022). Data were collected before (pre-trial), after (post-trial) and 8 weeks after (follow-up) outdoor provision. We assessed gait score using a 5-point visual scoring system and clinical claw lesion incidence in the trimming chute. For subclinical signs of lesions, the average hoof and claw surface temperatures were measured from the dorsal view of the coronary bands using an infrared thermography camera. Data were analyzed using mixed effect models with block, period, treatment, and the period × treatment interaction as fixed effects (α < 0.05), and either cow or claw nested within cow as random effect. The ambient and cow’s eye maximum temperature were added as covariates for analyzing thermography data. Average gait scores at pre-trial, post-trial, and follow-up were 2.13 ± 0.55, 1.98 ± 0.61 and 2.02 ± 0.51, respectively. Only non-severe lesions (sole
Julliot, C.; Dallago, G. M.; Nejati, A.; Diallo, A. B.; Vasseur, E.
Using kinematic to quantify gait attributes and predict gait score in dairy cows. [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{julliot_using_2023,
title = {Using kinematic to quantify gait attributes and predict gait score in dairy cows.},
author = {C. Julliot and G. M. Dallago and A. Nejati and A. B. Diallo and E. Vasseur},
url = {https://www.adsa.org/Portals/0/SiteContent/Docs/Meetings/2023ADSA/Abstracts_BOOK_2023.pdf},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Ottawa, Canada},
abstract = {Early identification of gait abnormalities could reduce lameness in dairy cows, but current assessment methods are laborious, making their use prohibitive for continuous monitoring. The objective of this pilot study was to predict the gait scores of dairy cows based on the gait attributes obtained using kinematic data. Data were collected from 12 Holstein cows between January 18 and February 12, 2021, using reflective markers attached to 20 anatomical locations. Cows were walked multiple times in a 7-m passageway corridor with 3 cameras on each side, totaling 69 passages of 3 steps long each. Five gait attributes of distance, duration, and velocity were calculated using 3D coordinates of the hoof markers. Range of motion was measured based on the angle between the stifle, hock, and fetlock markers of the rear legs. A trained observer scored cows’ gait on a 5-point numerical rating system (NRS), with scores ranging from 1 (sound) to 5 (severe lameness). Passages with NRS scores of 1.5 (n = 1) and 4 (n = 2) were removed since there were insufficient samples to represent these scores. The data were split into training and validation sets following a 70:30 ratio stratified by the distribution of the NRS scores. The machine learning algorithms RPART, GBM, XGBM, RF, and SVM with a radial basis kernel were trained using leave-one-out cross-validation. In addition to the original data, weighted classification and synthetic minority over-sampling technique (SMOTE) were also tested due to uneven NRS distribution. The models were evaluated according to their accuracy, sensitivity, specificity, F1 score, and balanced accuracy on the validation split. The best model was the XGBM trained using the original data, which achieved an overall accuracy of 0.66 (95% CI = 0.53–0.78). Conversely, the weighted RPART classification model had the lowest overall accuracy of 0.44 (95% CI = 0.31–0.57). The insights from this pilot study contribute to developing an automatic monitoring system to identify and treat cows that have impaired locomotion but are not yet clinically lame, allowing for improved welfare and profitability of dairy cows.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Early identification of gait abnormalities could reduce lameness in dairy cows, but current assessment methods are laborious, making their use prohibitive for continuous monitoring. The objective of this pilot study was to predict the gait scores of dairy cows based on the gait attributes obtained using kinematic data. Data were collected from 12 Holstein cows between January 18 and February 12, 2021, using reflective markers attached to 20 anatomical locations. Cows were walked multiple times in a 7-m passageway corridor with 3 cameras on each side, totaling 69 passages of 3 steps long each. Five gait attributes of distance, duration, and velocity were calculated using 3D coordinates of the hoof markers. Range of motion was measured based on the angle between the stifle, hock, and fetlock markers of the rear legs. A trained observer scored cows’ gait on a 5-point numerical rating system (NRS), with scores ranging from 1 (sound) to 5 (severe lameness). Passages with NRS scores of 1.5 (n = 1) and 4 (n = 2) were removed since there were insufficient samples to represent these scores. The data were split into training and validation sets following a 70:30 ratio stratified by the distribution of the NRS scores. The machine learning algorithms RPART, GBM, XGBM, RF, and SVM with a radial basis kernel were trained using leave-one-out cross-validation. In addition to the original data, weighted classification and synthetic minority over-sampling technique (SMOTE) were also tested due to uneven NRS distribution. The models were evaluated according to their accuracy, sensitivity, specificity, F1 score, and balanced accuracy on the validation split. The best model was the XGBM trained using the original data, which achieved an overall accuracy of 0.66 (95% CI = 0.53–0.78). Conversely, the weighted RPART classification model had the lowest overall accuracy of 0.44 (95% CI = 0.31–0.57). The insights from this pilot study contribute to developing an automatic monitoring system to identify and treat cows that have impaired locomotion but are not yet clinically lame, allowing for improved welfare and profitability of dairy cows.
Cellier, M.; Aigueperse, N.; Vasseur, E.
To go or not to go? Assessing anticipation for outdoor access in dairy cows [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{cellier_go_2023,
title = {To go or not to go? Assessing anticipation for outdoor access in dairy cows},
author = {M. Cellier and N. Aigueperse and E. Vasseur},
url = {https://www.adsa.org/Portals/0/SiteContent/Docs/Meetings/2023ADSA/Abstracts_BOOK_2023.pdf},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Ottawa, Canada},
abstract = {Regular outdoor access may enrich movement-restricted cows’ lives, but how they perceive it is unclear. Measures of cognitive processes, such as anticipation, can provide information on animal emotions. Our study aimed to characterize the anticipatory behaviors of cows living in restricted movement environments when an outing experience is provided as enrichment. In 2 independently analyzed trials, 32 tie-stallhoused lactating cows were blocked by parity and DIM and divided into 2 groups. Control cows (C; n = 8 in winter (W) and summer (S)) remained inside for the duration of the experiment, and outdoor cows (O; n = 8 in W and S) were provided with 5 d/week of outdoor access for 8 weeks. The outdoor areas combined different durations of outing (1 vs 2 h) and area sizes (20, 40, 60 and 80 m2). Exit cues, such as route preparation and halter placement in front of the outdoor cows, were set 20 min before the cows had outdoor access, and video recordings were taken during this period. The number of behavioral and ear position transitions were recorded and analyzed with 2 mixed models:1) to compare C vs. O cows with treatment, day, and their interaction as fixed effects, and cow nested within block as random effect; and 2) to test the influence of duration, area size, their interaction, and day as fixed effects, and cow nested within block as random effect. For W and S, no significant differences were found for anticipatory behaviors between C and O cows, nor for different durations and area sizes (overall mean ± SD: behavioral transitions: C: 23.9 ± 22.5, O: 20.9 ± 20.4; ear position transitions: C: 71.3 ± 22.5, O: 79.6 ± 20.4). Investigation of other behaviors before the exit (e.g., exploration, trampling, or lying) and having targeted observation periods (e.g., the first 5 min after setting exit cues) could allow for a more detailed study of the animal’s behavioral expression during anticipation. Combined with other indicators such as motivation, it would provide a basis for assessing emotional states in animals.},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Regular outdoor access may enrich movement-restricted cows’ lives, but how they perceive it is unclear. Measures of cognitive processes, such as anticipation, can provide information on animal emotions. Our study aimed to characterize the anticipatory behaviors of cows living in restricted movement environments when an outing experience is provided as enrichment. In 2 independently analyzed trials, 32 tie-stallhoused lactating cows were blocked by parity and DIM and divided into 2 groups. Control cows (C; n = 8 in winter (W) and summer (S)) remained inside for the duration of the experiment, and outdoor cows (O; n = 8 in W and S) were provided with 5 d/week of outdoor access for 8 weeks. The outdoor areas combined different durations of outing (1 vs 2 h) and area sizes (20, 40, 60 and 80 m2). Exit cues, such as route preparation and halter placement in front of the outdoor cows, were set 20 min before the cows had outdoor access, and video recordings were taken during this period. The number of behavioral and ear position transitions were recorded and analyzed with 2 mixed models:1) to compare C vs. O cows with treatment, day, and their interaction as fixed effects, and cow nested within block as random effect; and 2) to test the influence of duration, area size, their interaction, and day as fixed effects, and cow nested within block as random effect. For W and S, no significant differences were found for anticipatory behaviors between C and O cows, nor for different durations and area sizes (overall mean ± SD: behavioral transitions: C: 23.9 ± 22.5, O: 20.9 ± 20.4; ear position transitions: C: 71.3 ± 22.5, O: 79.6 ± 20.4). Investigation of other behaviors before the exit (e.g., exploration, trampling, or lying) and having targeted observation periods (e.g., the first 5 min after setting exit cues) could allow for a more detailed study of the animal’s behavioral expression during anticipation. Combined with other indicators such as motivation, it would provide a basis for assessing emotional states in animals.
Dallago, G. M.
Using 3D coordinates and machine learning to predict dairy cow locomotor ability Miscellaneous
2023.
BibTeX | Tags: Presentation
@misc{dallago_using_2023,
title = {Using 3D coordinates and machine learning to predict dairy cow locomotor ability},
author = {G. M. Dallago},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Online},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Dallago, G. M.
Panel discussion on opportunity and challenges related to use of sensors for health and welfare. Miscellaneous
2023.
BibTeX | Tags: Presentation
@misc{dallago_panel_2023,
title = {Panel discussion on opportunity and challenges related to use of sensors for health and welfare.},
author = {G. M. Dallago},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Online},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Vliet, Rachel; Aigueperse, Nadege; Vasseur, Elsa
CAREGIVERS ENGAGING IN MONKEY BUSINESS: EVALUATING THE USE OF CAREGIVERS AS SOCIAL ENRICHMENT FOR CAPTIVE RHESUS AND CYNOMOLGUS MACAQUES [External] Miscellaneous
2023.
Abstract | Links | BibTeX | Tags: Presentation
@misc{van_vliet_caregivers_2023,
title = {CAREGIVERS ENGAGING IN MONKEY BUSINESS: EVALUATING THE USE OF CAREGIVERS AS SOCIAL ENRICHMENT FOR CAPTIVE RHESUS AND CYNOMOLGUS MACAQUES},
author = {Rachel Vliet and Nadege Aigueperse and Elsa Vasseur},
url = {https://www.ufaw.org.uk/ufaw-events/ufaw-online-animal-welfare-conference-2023},
year = {2023},
date = {2023-06-01},
urldate = {2023-06-01},
address = {Online},
abstract = {Enrichment recommendations for captive non-human primates prioritize group-housing to provide them with social enrichment, which is often not possible in laboratory environments, due to infrastructural and experimental constraints. Our objective was to test if caregivers could serve as effective social enrichment for captive rhesus and cynomolgus macaques and to see if there was an effect of the type of social behaviours they employed. Eight female pair-housed macaques (rhesus n=5, cynomolgus n=3) were split evenly into two treatment groups: human behaviour (HB) and monkey-like behaviour (MB). For monkeys in the HB group, caregivers would use only human behaviours while interacting with them. For those in the MB group, caregivers would attempt to replicate macaque
social behaviours. Treatment occurred daily at 1pm, when the monkeys would receive fresh produce, and lasted three weeks. The caregiver would attempt to engage with each subject for 6 minutes: 2 minutes during feeding and 2 minutes before and after. Interactions were unstructured, and caregivers would gauge their behaviours
based on the monkey’s reactions. Our results showed that monkeys were more likely to engage in interactions before feeding. During interactions, MB monkeys interacted less with the caregivers, showing lower rates of affiliative, submissive, and abnormal behaviour than HB monkeys (P < 0.10). Monkeys from both groups spent most of their time oriented towards the caregivers and at the front of their cages, showed a 2.2 ± 0.69% (7.9 sec},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}
Enrichment recommendations for captive non-human primates prioritize group-housing to provide them with social enrichment, which is often not possible in laboratory environments, due to infrastructural and experimental constraints. Our objective was to test if caregivers could serve as effective social enrichment for captive rhesus and cynomolgus macaques and to see if there was an effect of the type of social behaviours they employed. Eight female pair-housed macaques (rhesus n=5, cynomolgus n=3) were split evenly into two treatment groups: human behaviour (HB) and monkey-like behaviour (MB). For monkeys in the HB group, caregivers would use only human behaviours while interacting with them. For those in the MB group, caregivers would attempt to replicate macaque
social behaviours. Treatment occurred daily at 1pm, when the monkeys would receive fresh produce, and lasted three weeks. The caregiver would attempt to engage with each subject for 6 minutes: 2 minutes during feeding and 2 minutes before and after. Interactions were unstructured, and caregivers would gauge their behaviours
based on the monkey’s reactions. Our results showed that monkeys were more likely to engage in interactions before feeding. During interactions, MB monkeys interacted less with the caregivers, showing lower rates of affiliative, submissive, and abnormal behaviour than HB monkeys (P < 0.10). Monkeys from both groups spent most of their time oriented towards the caregivers and at the front of their cages, showed a 2.2 ± 0.69% (7.9 sec
social behaviours. Treatment occurred daily at 1pm, when the monkeys would receive fresh produce, and lasted three weeks. The caregiver would attempt to engage with each subject for 6 minutes: 2 minutes during feeding and 2 minutes before and after. Interactions were unstructured, and caregivers would gauge their behaviours
based on the monkey’s reactions. Our results showed that monkeys were more likely to engage in interactions before feeding. During interactions, MB monkeys interacted less with the caregivers, showing lower rates of affiliative, submissive, and abnormal behaviour than HB monkeys (P < 0.10). Monkeys from both groups spent most of their time oriented towards the caregivers and at the front of their cages, showed a 2.2 ± 0.69% (7.9 sec
Vasseur, Elsa; banire Diallo, Abdoulaye
Présentation du projet WELL-E [External] Miscellaneous
2023.
Links | BibTeX | Tags: Presentation
@misc{vasseur_presentation_2023,
title = {Présentation du projet WELL-E},
author = {Elsa Vasseur and Abdoulaye banire Diallo},
url = {https://www.quebec.ca/gouvernement/politiques-orientations/strategie-quebecoise-sante-bien-etre-animaux},
year = {2023},
date = {2023-05-01},
urldate = {2023-05-01},
address = {Qc},
keywords = {Presentation},
pubstate = {published},
tppubtype = {misc}
}