The model, called OtoDX, achieved over 95% accuracy in diagnosing ear infections on a set of 22 test images, according to a new study published in the journal Science. Otolaryngology-Head and Neck Surgery and primary care physicians examining the same images.
When tested on a dataset of more than 600 inner ear images, the AI model’s diagnostic accuracy was over 80 percent, a significant leap above the clinicians’ average accuracy.
The model was created from hundreds of photos collected from children before Mass Eye and Ear surgery for recurrent ear infections or ear fluids. According to the authors, the results are an important step towards the development of a diagnostic tool that could one day be deployed in clinics to help doctors examine patients. An AI-based diagnostic tool could provide doctors such as pediatricians and emergency room clinics with an additional test to help them better inform clinical decision-making.