Back to search

BIONÆR-Bionæringsprogram

TraWel: Applying transmission model techniques and biosensor data to improve animal health and welfare in Norwegian dairy herds

Alternative title: TraWel: Bruk av transmisjonsmodeller for å avdekke helseutfordringer hos norske mjølkekyr

Awarded: NOK 6.6 mill.

Project Number:

326400

Application Type:

Project Period:

2021 - 2026

Funding received from:

Partner countries:

Large amounts of data are collected in Norwegian livestock production. This can be used to detect disease at an early stage and to make predictions for better health and welfare. In order to be able to take advantage of these large amounts of data, knowledge is needed in both advanced mathematical analyzes and biological contexts. In addition, computer programs will be able to learn from patterns that appear over time (machine learning). In Norway, more than half of the milk comes from farms with automatic milking systems (AMS). These systems generate enormous amounts of data that we can utilize provided that we take part in the competence building. In TraWel, we will develop new methods for utilizing large amounts of data to detect disease at an early stage in cows. The large amounts of data in question here are biosensor data from milking machines - which are collected several times a day per cow, throughout the year. Our goal is to detect and monitor the infectious disease called subclinical mastitis (hidden udder infection). TraWel will increase the knowledge at NMBU in handling large amounts of data from modern milking systems and help to utilize these with the use of advanced mathematical and statistical analyzes. The knowledge created through TraWel, the Faculty of Veterinary Science can benefit from in its work with preventive health work, and contribute to knowledge-based development for Norwegian livestock production. At NMBU, we have a knowledge about biology, mathematics, robotics, and statistics, and we are therefore in a unique position to contribute to the green shift. For us veterinarians, this means that healthy food from healthy animals that experience a good life where animal welfare is taken care of. In other words, an ethical food production.

Detecting disease early and thereby improving animal health and indirectly safeguarding good animal welfare is essential for present and future agricultural production. With the project called TraWel, we will develop new methods for utilizing large amounts of data to detect disease at an early stage. The large amounts of data in question, are sensor data from automatic milking machines - which are collected several times a day per cow, throughout the year. The infectious disease we are going to detect is subclinical mastitis. This disease causes large losses in Norwegian milk production. By integrating sensor data from automatic milking systems in the Nordic Cattle Data Exchange (NCDX) database, we will develop a decision management tool for handling and early warning of subclinical mastitis at both herd and animal level. The decision-tool will be linked to an animal welfare indicator that TINE has developed. This indicator is updated monthly and is currently used to advise farmers to improve animal welfare. The animal welfare indicator consists of 9 sub-indicators. One of these is an udder health indicator. TraWel provides a unique opportunity to link an early warning system for subclinical mastitis to this sub-indicator of udder health, to reduce the spread of disease and thus improve animal welfare. "A TraWel through big data, machine learning (ML), advanced mathematical methods, developing new methods for utilizing large amounts of biosensor data to detect disease at an early stage"

Funding scheme:

BIONÆR-Bionæringsprogram