A new algorithm has been used to detect the life-threatening condition in dogs, with an accuracy rate “greater than 99%”, say US developers.
Researchers at the University of California, Davis (UC Davis) School of Veterinary Medicine have developed an algorithm using artificial intelligence (AI) to detect Addison’s disease in dogs.
Addison’s disease – also known as hypoadrenocorticism – is a rare, life-threatening condition that results in a lack of critical hormones needed to maintain health, but it is notoriously difficult to recognise.
The loss of hormones associated with Addison’s disease results in subtle irregularities in blood tests that can be confused with other disease processes.
Dogs have vague clinical signs that mimic other conditions, such as kidney and intestinal disease, causing US veterinarians to refer to the disease as “the great pretender”. Addison’s can go undetected for years.
The new AI programme – developed in conjunction with electrical and computer engineers – eliminates confusion to diagnose Addison’s with “an accuracy rate greater than 99%” and will be commercially available by the end of 2020.
Krystle Reagan, a board-certified small animal internist with the UC Davis veterinary hospital, said: “It has the potential to revolutionise the detection of Addison’s and save many dogs’ lives.”