Moon blindness, also known as equine recurrent uveitis (ERU), is a common inflammatory ocular disease in horses that can lead to blindness or loss of the affected eye. The disease has a major economic impact and correct and swift diagnosis is crucial to minimize lasting damage. A team of researchers led by Professor Anna May from the LMU Equine Clinic has developed a deep learning tool that can reliably diagnose ERU based on photos, offering support to veterinary doctors in making accurate diagnoses.
In a study conducted by the researchers, 150 veterinarians were asked to evaluate 40 photos showing a mixture of healthy eyes, eyes with ERU, and eyes with other diseases. The deep learning tool was then tasked with evaluating the same photos based on image analyses. The results showed that while specialized equine veterinarians correctly interpreted the pictures 76 percent of the time, other veterinarians from small animal or mixed practices were accurate 67 percent of the time. In comparison, the AI tool had a 93 percent probability of providing a correct diagnosis, demonstrating its reliability in recognizing ERU and its potential as a supportive tool for veterinary doctors.
The web-app-based tool developed by the researchers is simple to use and only requires a smartphone. While it is not intended to replace veterinary professionals, it can assist them in reaching the correct diagnosis, especially for less experienced professionals or horse owners in areas with limited access to veterinarians. Early detection of ERU through the tool can lead to quicker treatment for affected horses, potentially slowing down the progression of the disease and saving the eyes of afflicted animals.
The deep learning tool developed by the LMU Equine Clinic researchers has significant implications for the diagnosis and treatment of ERU in horses. With its high level of accuracy in recognizing the disease based on photos, the tool can aid in early detection and prompt treatment, ultimately improving the outcome for affected horses. By providing support to veterinary doctors in making accurate diagnoses, the tool has the potential to enhance the management of ERU and reduce the economic impact of the disease in the horse industry.
Professor Anna May and her team have successfully demonstrated the effectiveness of the deep learning tool in diagnosing ERU in horses, highlighting its reliability and potential as a valuable tool for veterinary professionals. The high probability of a correct diagnosis provided by the AI tool compared to human specialists emphasizes its significance in the field of equine medicine. With its user-friendly interface and accessibility, the tool offers a practical solution for supporting veterinary doctors in diagnosing and treating ERU in horses, contributing to improved outcomes for affected animals and enhanced management of the disease.
Overall, the development of the deep learning tool for diagnosing ERU in horses by the researchers at the LMU Equine Clinic represents a significant advancement in equine medicine. By harnessing the power of artificial intelligence, the tool offers a reliable and efficient solution for diagnosing the disease based on photos, providing valuable support to veterinary professionals. With its potential to improve the early detection and treatment of ERU in horses, the tool has the capacity to enhance the management of the disease, ultimately benefiting both horses and the equine industry as a whole.
