Home Medizin KI-Smartphone-App erweist sich als vielversprechend bei der Diagnose von Ohrenentzündungen bei Kindern

KI-Smartphone-App erweist sich als vielversprechend bei der Diagnose von Ohrenentzündungen bei Kindern

von NFI Redaktion

Researchers at UPMC and the University of Pittsburgh have developed a new mobile app that uses artificial intelligence (AI) to accurately diagnose ear infections or acute otitis media (AOM). This app could help reduce unnecessary antibiotic use in young children. The findings of this research were published today in JAMA Pediatrics.

AOM is one of the most common childhood infections for which antibiotics are prescribed. However, without extensive training, it can be difficult to distinguish AOM from other ear conditions. The new AI tool, which diagnoses by analyzing a short video of the eardrum taken by an otoscope connected to a mobile phone camera, provides a simple and effective solution that may be more accurate than trained physicians.

Acute otitis media is often misdiagnosed. Underdiagnosis results in inadequate care and overdiagnosis leads to unnecessary antibiotic treatment, which can compromise the effectiveness of currently available antibiotics. Our tool helps in making the correct diagnosis and initiating the proper treatment.

Alejandro Hoberman, MD, lead author, Professor of Pediatrics and Director of the Department of General Academic Pediatrics at Pitt’s School of Medicine and President of UPMC Children’s Community Pediatrics

According to Hoberman, about 70% of children experience an ear infection before their first birthday. While this condition is common, an accurate diagnosis of AOM requires a trained eye to detect subtle visual findings resulting from a brief look at the eardrum of a squirming baby. AOM is often confused with otitis media with effusion, a condition generally not involving bacteria and not benefiting from antimicrobial treatment.

To develop a practical tool to improve accuracy in diagnosing AOM, Hoberman and his team first created a training library with 1,151 videos of eardrums from 635 children who visited UPMC pediatric practices between 2018 and 2023. Two trained experts with extensive experience in AOM research reviewed the videos and made a diagnosis: AOM or not AOM.

The researchers used 921 videos from the training library to teach two different AI models to recognize AOM by examining eardrum features, including shape, position, color, and transparency. They then used the remaining 230 videos to test the models‘ performance.

Both models were highly accurate, with sensitivity and specificity values of over 93%, indicating low rates of false negatives and false positives. Previous studies with physicians have reported diagnostic accuracy rates of AOM between 30% and 84%, depending on the type of healthcare provider, level of training, and age of the children studied.

These results suggest that our tool is more accurate than that of many clinicians,“ said Hoberman. „It could be a turning point in primary healthcare if it helps doctors strictly diagnose AOM and make treatment decisions.“

„Another advantage of our tool is that the videos we capture can be stored in a patient’s medical record and shared with other providers,“ said Hoberman. „We can also show parents and trainees – medical students and residents – what we see and explain why we diagnose an ear infection or not. It is important as a teaching aid and to reassure parents that their child is receiving appropriate treatment.“

Hoberman hopes that their technology can soon be widely used in healthcare provider practices to improve the accurate diagnosis of AOM and support treatment decisions.

Other authors of the study included Nader Shaikh, MD, Shannon Conway, Timothy Shope, MD, Mary Ann Haralam, CRNP, Catherine Campese, CRNP, and Matthew Lee, all from UPMC and the University of Pittsburgh; Jelena Kovačević, Ph.D., from New York University; Filipe Condessa, Ph.D., from the Bosch Center for Artificial Intelligence; and Tomas Larsson, M.Sc., and Zafer Cavdar, both from Dcipher Analytics.

This research was supported by the Department of Pediatrics at the University of Pittsburgh School of Medicine.


Journal Reference:

Shaikh, N., et al. (2024). Development and validation of an automated classifier for diagnosing acute otitis media in children. JAMA Pediatrics. doi.org/10.1001/jamapediatrics.2024.0011.

Related Posts

Adblock Detected

Please support us by disabling your AdBlocker extension from your browsers for our website.