An artificial intelligence (AI) model created at Mass Eye and Ear was significantly more accurate than doctors in diagnosing pediatric ear infections in the first such study, the research team working to develop the model for clinical use has reported. According to the new study, the results of which were published in the journal Otolaryngology-Head and Neck Surgery, the model, called OtoDX, was more than 95% more accurate in diagnosing ear infections on a set of 22 test images compared with 65% accuracy for a group of physicians consisting of ENTs, pediatricians and primary care doctors who looked at the same images. When tested on a dataset of more than 600 inner ear images, the diagnostic accuracy of the AI model was more than 80%. The model uses a type of artificial intelligence called deep learning and was built from hundreds of photographs collected from children before surgery at Mass Eye and Ear for recurrent ear infections or fluid in the ears. The authors say the results are an important step toward developing a diagnostic tool that could one day be deployed in clinics to help doctors examine patients.  Ear infections are incredibly common in children but are often misdiagnosed, leading to treatment delays or unnecessary antibiotic prescriptions, said lead study author Matthew Crowson, MD, an otolaryngologist and head and neck surgery at Harvard Medical School. This model will not replace physicians' judgment, but it can add to their experience and help them be more confident in making treatment decisions, he said.