Photo by Sage Levy
Ear infections, or the inflammation of the middle ear typically caused by bacteria, are a notorious nuisance in early childhood. The high incidence, recurrent nature, and non-specific presentation of ear infections in children have made it a feared diagnosis among parents. According to the National Institute on Deafness and Other Communication Disorders, ear infections are the most common reason for a trip to the pediatrician, with five out of six children experiencing at least one ear infection by age three, and 25% of children having repeated ear infections. Researchers at the University of Maryland (UMD), in collaboration with Johns Hopkins University (JHU), have developed an innovative technology that could save a trip to the doctor’s office by enabling virtually anyone to record and triage a child’s symptoms and status confidently and competently.
John Rzasa, chief engineer of UMD’s Robert E. Fischell Institute for Biomedical Devices, along with James H. Clark, assistant professor of otolaryngology at JHU and Mathias Unberath, assistant professor of computer science at JHU, teamed up to create a multi-pronged system to improve the diagnosis of pediatric ear infections.
Photo by Sage Levy
The technology consists of three distinct components: a novel video-based artificial intelligence (AI) algorithm that diagnoses ear disease from video; a unique, specialized digital otoscope capable of capturing a high-quality ear exam even with a novice user; and an application for a smart device, which can record the data from the digital otoscope, use the AI algorithm, and communicate with an electronic medical record. With this device, called “OtoPhoto,” physicians, medical assistants, and even parents at home can safely and confidently probe a child’s ears and get results quickly to confirm or rule out an ear infection.
While several direct-to-consumer digital otoscopes are commercially available, the OtoPhoto system surpasses its predecessors with improved video quality, compensation for user error, and sophisticated AI capabilities. OtoPhoto can also be used wirelessly, eliminating the need to be tethered by a cable to a screen, which makes current digital otoscopes ergonomically challenging to use and cumbersome to set up. This invention has the capability to reduce visits to the pediatrician, improve the diagnostic accuracy of ear infections, and subsequently reduce the need for antibiotics.
The team’s research was funded in part by a grant from the Leon Lowenstein Foundation in conjunction with JHU. Research on the video-based AI algorithm component of the invention has been published in Frontiers in Digital Health.
Rzasa, Clark, and Unberath’s invention was recognized as a finalist in the physical sciences category for the Invention of the Year awards, which were celebrated on May 3 at the 2022 Innovate Maryland event, a university-wide celebration of innovation and partnerships at UMD.