Artificial intelligence to identify equine ocular disease
Artificial intelligence could lead to more horses getting an earlier diagnosis.
An artificial intelligence (AI) tool could be used to diagnose equine recurrent uveitus (ERU) in horses.
The inflammatory ocular disease, also known as ‘moon blindness’, can lead to blindness or loss of the affected eye.
ERU is one of the more common eye diseases in horses, and can have a major economic impact. A quick, correct diagnosis can minimise the lasting damage.
The research team created an AI tool, which was trained with photographs of diseases to identify the patterns which may lead to a diagnosis.
To assess the efficacy of their deep learning tool, researchers from the Ludwig-Maximilian-Universität München asked 150 veterinary surgeons to evaluate 40 photos of horses’ eyes. The pictures included a mixture of healthy eyes, eyes with ERU, and eyes with other diseases.
Equine veterinary surgeons completed the test with a 76 per cent success rate, meanwhile veterinary surgeons working in small animal or mixed practice identified the eye issues correctly 67 per cent of the time.
When the AI image analysis tool was given the same challenge, the probability of receiving the correct answer was 93 per cent.
The researchers say that, while the difference was not statistically significant, it proved that AI could reliably recognise ERU. This could support veterinary surgeons with the diagnosis of potential emergency cases.
This could lead to more horses getting an earlier diagnosis, increasing the likelihood of them receiving prompt treatment and saving affected eyes. It will also enable less experienced veterinary surgeons to differentiate between ERU and other opthalmic diseases.
The AI, deep learning tool is web-app based, and can be used through a smartphone device.
Professor Anna May, who led the research team, said: "It's not meant to replace veterinarians, but can help them reach the correct diagnosis.
“It is particularly valuable for less experienced professionals or for horse owners in regions where vets are few and far between,"
The full study can be found in the Equine Veterinary Journal.
Image © Shutterstock
The BEVA has opened two new roles on its Nurse Committee.
