AI-powered healthcare company AIDOT announced that its deep learning-based otitis media diagnostic pipeline, ECHO dot AI, has been published in the peer-reviewed Journal of Clinical Medicine. The study highlights the potential of AI to standardize otitis media diagnosis and reduce diagnostic disparities in primary care settings.

The research was conducted in collaboration with the Department of Otorhinolaryngology at Hanyang University Hospital, using 2,964 otoscopic images to train and validate the AI algorithm. The model was designed to classify the four most common medical conditions:

- Normal

- Acute Otitis Media (AOM)

- Otitis Media with Effusion (OME)

- Chronic Otitis Media (COM)

The AI diagnostic pipeline mimics the clinical workflow with a four-stage structure:

1. Image quality assessment

2. Automatic segmentation of the tympanic membrane

3. Laterality classification (left vs. right ear)

4. Final disease classification

The study reported image quality classification accuracy of 98.8%, laterality classification accuracy of 99.1%, and overall disease classification accuracy of 88.7%. Detailed performance metrics included F1-scores of 0.78 for AOM, 0.87 for OME, and 0.92 for COM, demonstrating clinically applicable performance.

The research team emphasized, “Otitis media is extremely common in children, yet diagnostic variability among non-specialist clinicians is high. AI-assisted diagnosis can reduce skill-based disparities and enable earlier detection, helping prevent complications.”

Jaehoon Jung, CEO of AIDOT, added, “This study provides strong evidence that AI diagnostic assistance can be effectively applied in ear-related diseases, enhancing diagnostic consistency, accuracy, and healthcare accessibility. Regulatory approval with MFDS is currently underway, and the solution will expand to international markets alongside Cerviray AI®.”

Korea Economic TV / December 5, 2025 / Reporter Yang Jaejun