Researchers to use supercomputers to predict the TB status of dairy cows

It is used regularly by global tech giants such as Google and Facebook, but now “deep learning” is set to become a key weapon in the fight against tuberculosis in cattle.

Livestock scientists at Scotland’s Rural College (SRUC) have been awarded £337,000 for a project which will use highly advanced computers to predict the bovine tuberculosis (bTB) status of dairy cows.

By analysing millions of milk samples with a greater degree of accuracy than ever before, it is hoped the project will strike a significant blow to a disease estimated to cost the UK dairy industry around £175 million a year.

Deep learning – a branch of machine learning – employs algorithms and techniques to make better use of increasingly huge datasets and advances in computer technology.

Previous research by SRUC and collaborators has already led to the development of the first national genetic and genomic evaluation for bTB resistance of the Holstein breed as well as the launch of TB Advantage by the Agricultural and Horticultural Development Board (AHDB).

By using mid-infrared spectroscopy (a scientific analytic tool which gives essentially gives a fingerprint of the substance tested) of milk samples from National Milk Records (NMR), SRUC researchers have previously confirmed significant genetic variation among individual animals for resistance to TB infection.

The funding follows a Defra review of its 25-year TB eradication strategy, which now specifically includes “improved genetic resistance and improved diagnostic tests”.

The new project will produce data that can be used for management and diagnostic purposes.

Prof. Mike Coffey, who is leading the study, said: “This is potentially a great step forward in the fight against TB.

Being able to use routinely collected milk samples to alert a cow that may have been exposed to infection will allow the farmer to take remedial action long before the normal testing regime comes in.

“The early removal of infected cows should lower the background level of TB and contribute to enabling more effective outcomes from the existing tools deployed to eradicate the disease.”