AI used to fight drug resistance
Scientists in the United Kingdom and China have announced plans to use artificial intelligence on chicken farms in order to combat the problem of antibiotic resistance in both farm animals and humans.
The new initiative will use machine learning to find ways to track and prevent disease on poultry farms, reducing the need for antibiotic treatment in chickens and therefore lowering the risk of antibiotic-resistant bacteria transferring to people.
The research will be led by animal health experts from the University of Nottingham and Nimrod Veterinary Products in the UK as well as two Chinese partners－New Hope Liuhe in Chengdu and the China National Center for Food Safety Risk Assessment.
"Antibiotic resistance is a worldwide problem and it's getting worse and worse. Some of these superbugs are resistant to everything, we don't know how to treat them," University of Nottingham veterinary professor Tania Dottorini told China Daily. "On farms, superbugs are not confined to animals, they spread to humans and to the environment, it's an exponential spread. If we don't understand how to stop this, it's going to be really bad."
Around 700,000 deaths a year stem from antibiotic resistance, according to a report commissioned by the UK government. If left unchecked, drug resistance could lead to 10 million deaths a year by 2050, which is more than the number of people who now die from cancer annually.
Antibiotics work by disrupting function in certain parts of a bacterial cell. Bacteria become resistant to antibiotics through genetic mutations that alter those areas of the cell, meaning the medication can no longer target them.
The more a strain of bacteria is exposed to an antibiotic, the more likely it is to become resistant. Large numbers of people and animals are given antibiotics when they don't need them, so reducing unnecessary consumption is crucial in the fight against so-called superbugs.
"When you have a large-scale data set, the human mind can't cope with that, it's too complex," Dottorini said of machine learning. "We need something that is able to understand the relationship across a big amount of information."
Dottorini said that, if successful, these methods should be transferable to other farm studies in China and abroad.