Swedish researchers have developed a method that can diagnose ear infections through image analysis using a smartphone and cloud technology. The technique provides low-cost diagnosis of middle ear infections, such as otitis media.
Researchers from Umea University in Sweden and the University of Pretoria of South Africa devised a software-based smartphone-compatible otoscope that analyzes images of the tympanic membrane.
The digital otoscope takes pictures of the eardrum and transmits high-resolution images to a smartphone. The software will then analyze the images and automatically categorize it into five diagnostic groups. The technique uploads the images to a cloud-based system.
The automated diagnosis using these images generated an accuracy rate of 80.6 percent from a commercial video-otoscope, while pictures generated on-site using low-cost video-otoscope achieved 78.7 percent accuracy.
"Using this method, health personnel can diagnose middle ear infections with the same accuracy as general practitioners and pediatricians," said Claude Laurent, a study co-author who is from the Department of Clinical Sciences at Umea University.
The study aims to make the diagnosis of ear infections, particularly otitis media or middle-ear infection, as accessible as possible.
Otitis media commonly occurs in childhood and affects half a billion children across the globe. In developing countries, many children with the said illness are misdiagnosed or not diagnosed at all because of the lack of accessible equipment. If the infection is left untreated, it could lead to complications, such as speech and language development problems or even permanent hearing loss.
With the newly developed method, the researchers have paved the way to making an easy diagnosis of otitis media. Their work suggests that illnesses and complications can be easily and accurately diagnosed through this method in the same way medical practitioners come up with a diagnosis.
"Since the system is cloud-based, meaning that the images can be uploaded and automatically analysed, it provides rapid access to accurate and low-cost diagnoses in developing countries," said Laurent.
The accuracy of diagnosis made by general practitioners ranges from 64 to 80 percent. Comparing it to the use of traditional otoscopes, the system produces highly accurate results.
Laurent added that this method ensures a low-cost diagnosis of otitis media in places where healthcare resources are scarce.
The project is profiled in the journal EBioMedicine.
