A telephone application developed by scientists with support for artificial intelligence made it possible to accurately diagnose ear infections or acute average otitis media.
According to a study conducted at the University of Pittsburgh in the United States, scientists gathered 1151 video eating membranes of 635 children who entered the university hospital in the period from 2018 to 2023 to develop an appendix that will increase the accuracy of an acute -degree of medium severity of the ear infection.
Researchers uploaded 921 videos into two different models of artificial intelligence to detect acute average otitis media by studying such features as the shape, location, color and transparency of the eardrum.
both models were able to distinguish ear infections from acute average otitis media with an accuracy of 93 percent. Depending on their level, the frequency of detection by doctors of these diseases ranges from 30 to 84 percent.
Researchers said that the recently developed telephone application with support for artificial intelligence can help doctors, since it is difficult to distinguish an acute infection of the middle ear, one of the most common children’s infections, from other ear diseases.
Professor Alejandro Soberman from the medical faculty of the University of Pittsburgh said that acute otitis media is often diagnosed incorrectly. “An erroneous diagnosis leads to unnecessary treatment with antibiotics, which can jeopardize the effectiveness of existing antibiotics,” he said.
Joberman noted that the developed application will help to make the correct diagnosis and direct it to the correct treatment.
The results of the study were published in Jama Pediatrics.