by George Looby
For decades dairymen, veterinarians and those in university laboratories have been fighting a battle against mastitis. Researchers in the UK have employed new technology for determining the origin of mastitis in a herd. For many years, it was considered either contagious or environmental – passing from one cow to another, most often at milking. In tracing possible environmental sources, one is most often looking at bedding and other material to which the cows are exposed.
The research team at the University of Nottingham in the UK incorporated algorithms as an aid in determining whether mastitis was of contagious or environmental origin in a study of samples from 1,000 herds. (An algorithm is a mathematical process to solve a problem using a finite number of steps; it is a set of instructions that defines not just what needs to be done, but how to do it.) Data to support the study were provided through the Agriculture and Horticulture Development Board (AHDB), whose mastitis control program gives support to dairy farmers in the UK.
This means a veterinarian can look at certain numbers and accurately determine where to begin their search for the source of an outbreak. One of the factors that precipitated this study is that greatly reduced rate of antibiotic use in the treatment of mastitis. With less antibiotics is use, the dairyman and the vet had to dig a little deeper into the sources of a given outbreak.
Mastitis classified as environmental is broken into two broad categories: acquired during the dry period and acquired during lactation. As the computer-assisted diagnosis has evolved, it has been able to achieve a high level of accuracy in diagnosing the origin of a mastitis problem. This ability to pinpoint a problem on a given farm is of great value to the practitioner. They now have an accurate guideline to follow.
The study conducted was designed to see if the decision-making process of a well-trained specialist could be matched by a machine using algorithms. It requires a significant amount of time for a clinician to analyze and interpret data gathered in the field and come up with a plan of action. A properly designed algorithm can do the same job in a matter of seconds. One advantage of this technology is that it would allow for much closer monitoring of a mastitis control program in a given herd. Problem areas could be identified more quickly and the success of treatment protocols could be evaluated. Regularly scheduled testing could be carried out in a more efficient fashion (and hopefully in a more cost-effective manner).
We may soon see our mastitis control programs undergoing some rather significant changes, with computer-driven programs taking over where only manual computations existed before.