The D4Dairy research consortium has a prominent member. Complexity researcher Peter Klimek from the Complexity Science Hub (CSH) Vienna and the Medical University (MedUni) Vienna has been named Austrian Scientist of the Year 2021. Congratulations!
by Kristina Linke (comments: 0)
...and all the best for 2022 wishes the D4Dairy team!
by Kristina Linke (comments: 0)
Following the D4Dairy Annual Meeting, some partners accepted the invitation and visited the Upper Austrian Insemination Station in Hohenzell.
by Kristina Linke (comments: 0)
The third D4Dairy Annual Meeting took place in Linz on 22 and 23 September 2021. It was hosted by the Chamber of Agriculture of Upper Austria and the Upper Austrian Insemination Station. After an exciting keynote lecture on "Open Science with Closed Data: Data Spaces and Data Circles" by Prof. Allan Hanbury, progress in all projects was presented and discussed with the numerous partners present. Participants also had the opportunity to see digitalisation on a farm and discuss it with the farmer and each other.
The aim would be to be able to generate new knowledge from this data diversity in the interaction of the most varied characteristics, which could subsequently be of great benefit for Austrian cattle breeding.
Since I use an automatic milking system in my farm, I use the associated herd management program (Lely). Currently, the herd management program and the LKV Herd Manager exchange data for performance control. The areas of birth reports and inseminations are not covered. Especially in this area I would like to see more intensive networking with the Rinderdatenverbund RDV via the LKV Herd Manager. Currently I have to more or less get into two different systems and make entries.
This network will strengthen not only the family farms, but also the business partners involved and Austria as a business location as a whole.
Thanks to all project partners and especially to the project team that managed to bundle the large number of partners for a joint project that ultimately benefits the farmer.
The application of new BigData methods to the extensive data set of the cattle industry holds enormous potential for the early detection and development of preventive models for diseases.
The simple, efficient and effective use of data for dairy farms is a major concern of our company. Our customers should spend as little time as possible on using our monitoring system. That is why we sought this cooperation through the D4Dairy project in order to create the greatest possible added value for our customers by combining data.
D4Dairy's overall goal is to provide digital support to dairy management via a data-driven, networked information system, exploiting the potential of advanced technologies (mid-infra-red spectra, genome information, ...) and advanced data analysis to further improve animal health, nutrition, animal welfare and product quality.
Building on the COMET project ADDA, the existing network along the milk value chain was expanded to include technology providers and scientific partners with a focus on new technologies to form the D4Dairy consortium.
The D4Dairy subprojects are divided into two areas:
Area 1: Data and Decision - Digitalisation, data integration and decision support
In recent years, there has been a dramatic increase in the generation, processing and storage of data on dairy farms. To help to utilize this data in a sensible way, the focus in Area 1 lies on integration of the data generated on farms (data from health-related sensors, feed automation, antibiotic use, housing climate data,…) and interoperability of the systems used in the farms and on process development and optimization, with the aim of generating decision support tools. The projects in this area include work on aspects of quality assurance, data exchange, data protection and research on the interrelationships between different characteristics. Based on the results and findings from the various research projects, tools for farmers and veterinarians to support decision-making will be developed. Different strategies to reduce the risk of antimicrobial resistance are an important part in this area as well. Furthermore, Area 1 brings together all results from D4Dairy, since it is here that the information and knowledge transfer takes place, and studies in the social context (knowledge transfer, acceptance) are carried out.
P1.1 tackles data exchange management and data networking between project partners. The aim is to create a database for D4Dairy research and for the development of applications for farmers. P1.1 describes the state of the art and the goals of various stakeholders (farmers/technology suppliers/organisations/…) dealing with automation systems applied in dairying. A focus is put on work organisation and process quality. A work package is dedicated to legal aspects of protection of data and intellectual property. Thus, newly developed applications are in line with the wishes of users and the current legal requirements. P1.1 develops the technical basis for an automated data exchange between the partners. This covers the routine data exchange between the central cattle database and the involved public and industry partners as well as the exchange of extended datasets for research purposes. Novel IT concepts for the quality assurance of sensor data and the efficient management of data usage and sharing rights will be investigated. New applications will be intensively tested on pilot farms. The insights of D4Dairy will be incorporated in concepts for the further development of novel traits and services.
The availability of integrated and novel data combined with advanced analytical methodologies allows earlier detection of the onset of potential health problems within dairy herds and thus enables timelier intervention. D4Dairy aims to develop possibilities for data-driven strategies for improving health and to support farmers in utilizing such strategies and tools so they can prevent or react to emerging health and welfare issues in a timely manner.
Data from digital technologies such as data from AMS (automated milking systems), from sensors and feeding systems will be used to derive new traits. The project will work on trait definition and their relationship to other traits of interest as well as sensor data evaluation for early detection of disease with focus on lameness, but also ketosis and others. Concepts for optimisation of feeding processes is another focus of work within this project. Outcome of the other projects within D4Dairy will incorporated in digital decision support tools developed within D4Dairy.
Project 1.3 comprises three separate subprojects, namely the harmonisation of antimicrobial susceptibility testing (AST) for mastitis pathogens (1.3.1), using data to drive recommendations for drying off strategies (1.3.2), and strategies to improve calf health and beef quality (1.3.3).
Working with laboratories throughout Austria, subproject 1.3.1 aims to standardise AST, improve comparability of resistance data as well as integrating the results of such testing into the central cattle database. AST data will then be more accessible to and evaluable for farmers and veterinarians. Subproject 1.3.2, together with a dairy and leading milk processor, will develop a decision-making tool for veterinarians to advise farmers on the best drying off strategy for their farm. Subproject 1.3.3. will analyse the effects of feeding calves milk containing antibiotic drug residues (“waste milk“) on the development of antimicrobial resistant bacteria on farm, and to suggest alternatives to this practice.
The use of digital tools, sensor technology and automation in modern agriculture is rapidly advancing and expanding worldwide. The application of digital technology changes the way farms are operated and managed and provides new perspectives and challenges for consultants and veterinarians. Therefore, it is essential for these groups to acquire knowledge in this field. Besides, consumers are becoming more concerned about animal health, animal welfare, food safety and the nutritional aspects of food. Although trust in the food produced can be strengthened through the use of modern technologies, social acceptance of dairying 4.0 is of great importance and should be taken into account. In this project we want to carry out quantitative and qualitative studies on the acceptance and assessment of new technologies among farmers, consultants, veterinarians and consumers. Furthermore, we will disseminate results via various communications channels to meet training needs of the farmers, consultants and veterinarians.
Area 2: Data and Detection - Data driven detection of risk factors and early predictors for improved health
Functional traits, in particular health (related) traits, have gained importance in livestock breeding during the last years. In this context, data from automation and sensors or other novel traits in combination with advanced analysis methodology offer new possibilities for herd management and breeding. Area 2 of D4Dairy covers the broad range of recording and validation of novel traits, the combination of novel and already available traits as well as detection of different risk factors for animal health by using novel methodology up to implementation in optimisation, mitigation and breeding strategies. This enables the improvement of animal health and welfare in dairy cattle farms. Both reduction of costs and increase of returns by improving both functional and performance traits will result in a higher profitability of the dairy cattle sector.
Many animal disorders arise from the combined interplay of genetic and environmental risk factors. Early detection and improved monitoring of these risk factors can contribute significantly to animal health and welfare. This project will study the risk factors for the development of diseases and develop parameters for early detection utilizing novel scientific approaches such as big data analyses.
Our aim is to develop a quantitative and predictive framework that enables us to disentangle how different genetic and environmental factors, as well as their interactions, contribute to animal health and welfare, based on a comprehensive large-scale database that contains time-resolved phenotypic and environmental information on as well as genetic information for animals. The project will work on the identification of early prognostic disease markers. Expected results are more meaningful parameters for herd management and breeding.
Midinfrared spectroscopy (MIR) is the established routine procedure of dairy milk analysis. Beeing quite recently mainly used for the analysis of the standard milk components fat, protein and lactose, it is now possible to determin the content of low concentration components like e.g. fatty acids, ketone bodies and indicators like lactoferrin with decent acuracy. Recent research has been able show that the status of a dairy cow with regard to health and nutrition can be strongly linked to the biochemical composition of milk. So milk MIR spectroscopy is a convenient method for early detection of health and nutrition problems in dairy herds. The goal of the MIR project is the development and evaluation of new robust milk MIR prediction models by using a large number of routine milk samples and closely linked diagnosis made by veterinarians. Here participating research partners can resort to the huge expertise and data collected by the milkrecording organisations which make up the EEIG EMR (European Milk Recording). Furthermore new reference analysis will help to improve existing models of health related milk components.
The intention ist to supply dairy farmers with new reliable tools for their herd management which help them to cope with growing requirements concerning performance, economic efficiency and animal welfare in a sustainable way.
The overall aim of this project is to enhance animal health, welfare, production efficiency and potentially improve product quality by improving environmental data collection and data usage on farms. This aim will be achieved by (1) initially deploying commercially available sensors and communication networks to establish an integrated on-farm monitoring system for detecting various environmental variables. Then (2) the research team will collaborate with farm managers to collect production related data and generate optimisation strategies based on the collected information. In addition, the project is aimed at (3) developing practical/affordable sensor applications in dairy production and (4) modelling environmental stressors and animal responses to develop forecasting algorithms in relation to production loss associated with sub-optimal environment. To ensure that the project is applied; the study is executed in close collaboration with producers, commercial companies (Pessl Instruments and AgHiTech), academic institutions (BOKU, Vetmeduni) and industry bodies (ZuchtData, LKV).
The project uses Big Data approaches to generate easily interpretable management tools from complex data sources. Besides management purposes, these data sources can also be used for breeding.
For breeding applications the ease of interpretation of the trait is less important than the repeatability and especially the correlation with target traits. Therefore, it might be required to establish specific phenotypes for this project.
New traits will be developed for the complexes metabolism, udder and claw health. The work packages include the estimation of heritabilities and genetic correlations, the establishment of conventional and genomic breeding value estimations and the search for interesting loci via genome-wide association studies.
Mycotoxins - secondary metabolites of moulds – can lead to adverse health effects in livestock and cause economic losses worldwide. Yet, the presence of mycotoxins in Austrian dairy feeds and its impact on dairy health and fertility is largely unknown.
In this project, we perform an extensive mycotoxin survey in Austrian dairy farms with feed components being analysed for more than 400 fungal metabolites. By matching mycotoxin occurrence with dairy health and performance data, potential interlinks will be revealed. The rumen microbiome might also interfere with mycotoxin contamination, wherefore we use next generation sequencing techniques to unravel the effects of relevant mycotoxins on the rumen microbiome in vitro (rumen simulation technique, RUSITEC).
Results will contribute to our understanding of mycotoxins as possible trigger factor for impaired dairy health. Since climate change effects mycotoxin contamination levels in feed, our findings will be of importance for dairying beyond the duration of this project.
In order to tackle these complex and interdisciplinary challenges, D4Dairy has established an internationally competitive, transdisciplinary network of universities, competence centres and research institutes in Austria and abroad, as well as companies along the milk value chain (farmers, breeding organisations, milk processors, animal health services, interest groups, etc.) and - last but not least - national and international technology providers (sensors, feeding, climate measurement, data processing). The consortium consists of 31 company partners and 13 scientific partners.
Österreichische Agentur für Gesundheit und Ernährungssicherheit GmbH (AGES)
CRA-W - Walloon Agricultural Research Center
CSH - Complexity Science Hub Vienna/ Medical University Vienna (Meduni)
FFoQSI - Austrian Competence Center for Feed and Food Quality, Safety and Innovation
HBLFA Raumberg-Gumpenstein - Agricultural Research and Education Centre
BLT Wieselburg - Federal Institute of Education and Research Francisco Josephinum
LFL Bayern - Bayerische Landesanstalt für Landwirtschaft, Institut für Tierernährung und Futterwirtschaft (ITE)
LFL Bayern - Bayerische Landesanstalt für Landwirtschaft, Institut für Tierzucht (ITZ)
TU Graz - Graz University of Technology/ Institute for Technical Informatics
University College for Agrarian and Environmental Pedagogy (HAUP)
University of Liège (Gembloux Agro-Bio Tech - GxABT)
University of Natural Resources and Life Sciences Vienna (BOKU) / Institute of Agricultural Engineering
University of Natural Resources and Life Sciences Vienna (BOKU) / Division of Livestock Sciences
University of Veterinary Medicine Vienna (Vetmeduni) / Institute for Veterinary Public Health (IVPH)
University of Veterinary Medicine Vienna (Vetmeduni) / Institute of Animal Nutrition and Functional Plant Compounds
University of Veterinary Medicine Vienna / University Clinic for Ruminants
AMA - Agrarmarkt Austria Marketing GesmbH
Animal Health Services / Tiergesundheitsdienste
Biomedica Medizinprodukte GmbH & Co KG
bioMérieux Austria GmbH
BIOMIN Holding GmbH
Cattle Breeding Associations - FIH
Cattle Breeding Associations - NÖ Genetik
Chambers of Agriculture – LK Ö, LK NÖ, LK OÖ
European Milk Recording EEIG (EMR EEIG)
LKV Austria Qualitätsmanagement GmbH - Federal Recording Association
LKV Baden-Württemberg – Recording Association/ Landesverband Baden- Württemberg für Leistungs- und Qualitätsprüfungen in der Tierzucht e.V.
LKV Bayern e.V.
ILV - Local Quality Laboratory of Carinthia
QLG - Local Quality Laboratory of Lower Austria
Milchprüfring Upper Austria - Local Quality Laboratory of Upper Austria
ÖFK – Österreichische Fleischkontrolle Ges.m.b.H
SCR by Allflex
smaXtec animal care GmbH
ZAR - Zentrale Arbeitsgemeinschaft österreichischer Rinderzüchter (Federation of Austrian Cattle Breeders)
ZuchtData - ZuchtData EDV-Dienstleistungen GmbH
Further cooperation partners for specific topics
Reviewed Scientific Journals
- Matzhold et al. (2021) A systematic approach to analyse the impact of farm-profiles on bovine health. Science Reports 11, 21152
- Lasser et al. (2021) Integrating diverse data sources to predict disease risk in dairy cattle – a machine learning ap-proach. Journal of Animal Science, skab294
- Christophe et al. (2021). Multiple Breeds and Countries’ Predictions of Mineral Contents in Milk from Milk Mid-Infrared Spectrometry. Foods, 10(9), 2235.
- Penagos-Tabares et al. (2021). Mycotoxins, Phytoestrogens and Other Secondary Metabolites in Austrian Pastures: Occurrences, Contamination Levels and Implications of Geo-Climatic Factors. Toxins, 13(7), 460.
- Firth et al. (2021). The Effects of Feeding Waste Milk Containing Antimicrobial Residues on Dairy Calf Health. Pathogens, 10 (2), 112
- Entenfellner & Drillich (2020). Survey on knowledge transfer in farm animal practice. Der Praktische Tierarzt 101, 1228–1235
- Rienesl et al. (2020). Prediction of pregnancy state from milk mid-infrared (MIR) spectroscopy in dairy cows. Acta Fytotechnica et Zootechnica, 23(5).
- Egger-Danner et al. (2020). Use of benchmarking to monitor and analyze effects of herd size and herd milk yield on cattle health and welfare in Austrian dairy farms. Journal of Dairy Science
- Rienesl et al. (2019). Mastitis Detection from Milk Mid-Infrared (MIR) Spectroscopy in Dairy Cows. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 67(5), 1221-1226.
- Firth et al. (2019). Comparison of Defined Course Doses (DCDvet) for Blanket and Selective Antimicrobial Dry Cow Therapy on Conventional and Organic Farms. Animals, 9(10), 707.
- Costa et al. (2019). On the genomic regions associated with milk lactose in Fleckvieh cattle. Journal of dairy science, 102(11), 10088-10099.
- Firth al. (2019) Relationship between the probability of veterinary-diagnosed bovine mastitis occurring and farm management risk factors on small dairy farms in Austria. Journal of dairy science, 102(5), 4452-4463.
Reviewed Conference Papers
- Klimek et al. (2021) Data driven technologies: overview of current research and possibilities related to producing proxies for new phenotypes important to a sustainable and responsible livestock production. EAAP
- Rienesl et al. (2021) Predicting mastitis with somatic cell count, differential somatic cell count and milk MIR spectra. EAAP
- Grand et al. (2021) Assessment of needs for data integration and digitalization in Austrian dairy cattle farms. EAAP
- Schodl et al. (2021) Strategies for deriving auxiliary traits for lameness prediction and breeding value estimation. EAAP
- Schodl et al. (2021). D4Dairy-project–How digitalization and data integration pave the way to dairy health improvement. Interbull Bulletin, (56), 1-6.
- Pfeiffer et al. (2021) Prediction of subclinical ketosis based on automated milking systems (AMS) in dairy cattle. Animal-science proceedings, 12(1), 109.
- Fuerst-Waltl et al. (2020). Genetic relationships between ketosis and potential indicator traits. British Society of Animal Science Annual Conference
- Hintringer et al. (2020). Risk factors associated with milk fever in Austrian Fleckvieh. EAAP
- Fuerst-Waltl et al. (2020). Potential auxiliary traits for ketosis based on MIR spectra. EAAP
- Dale et al. (2020). Ketosis and Energy Balance milk MIR spectral predictions - Practical Use. EAAP
- Dale et al. (2020). „MastiMIR“ - A mastitis early warning system based on MIR spectra. EAAP
- Matzhold et al. (2020). A systematic approach to quantify the impact of feeding and farm management practices on bovine health. EAAP
- Papst et al. (2020). Localization from activity sensor data. In Proceedings of the 18th Conference on Embedded Networked Sensor Systems (pp. 703-704).
- Papst (2020). Privacy-preserving machine learning for time series data: PhD forum abstract. In Proceedings of the 18th Conference on Embedded Networked Sensor Systems (pp. 813-814).
- Rienesl et al. (2020) Prediction of pregnancy state from milk mid-infrared (MIR) spectroscopy in dairy cows. Animal Science Days
- Papst et al. (2019). Embracing Opportunities of Livestock Big Data Integration with Privacy Constraints. IoT Conference, 22.-25.10.2019, Bilbao, Spain
- Costa et al. (2019) On the genomic regions affecting milk lactose content in dairy cattle. EAAP conference
- Koek et al. (2019) Genetic parameters for ketosis and newly developed ketosis risk indicators based on MIR spectra. EAAP conference.
- Firth et al. (2019) Added value of data integration to reduce the use of antimicrobials on dairy farms. EAAP conference.
- Firth et al. (27.6.2019) Does the feeding of discard milk to dairy calves lead to antimicrobial resistance on farm? ICPD Bern
- Egger-Danner et al. (2019) Opportunities and challenges of data integration with focus on claw health and metabolism for decision support in herdmanagement., EAAP conference
- Werner et al. (2019) "KetoMIR2” – modelling of ketosis risk using vets diagnosis and MIR spectra for dairy cows in early lactation. ICAR conference.
- Obritzhauser et al. (2019). Integrating bacteriological milk examination into decision support for reduced use of antimicrobials. ICAR conference.
- Egger-Danner et al. (2019). Internet of Cows-Opportunities and Challenges for Improving Health, Welfare and Efficiency in Dairying. ICAR conference.
- Grandl et al. (2019). Opportunities and Challenges of New Technologies for Performance Recording with Focus on Claw Health and Metabolism. ICAR Conference.
- Firth et al. (2019). Is there a link between antimicrobial use and the prevalence of MRSA and ESBL-producing Escherichia coli on Austrian dairy farms? ICPD.