AI-Assisted Detection Support for Middle Ear Diseases Using Multimodal Large Language Models

Yuan-Chia Chu ^1,2, Kuan-Hsun Lin ^1,2, Yuan-Chen Chou ³

⁰Department of Information Management, Taipei Veterans General Hospital, Taipei, Taiwan.

Studies in Health Technology and Informatics +

|

August 8, 2025

View abstract on PubMed

Summary

This study introduces an AI system for detecting middle ear diseases from otoscopic images. The novel approach enhances diagnostic accuracy in primary care settings.

Area Of Science

Otolaryngology
Artificial Intelligence
Medical Imaging

Background

Accurate detection of middle ear diseases like otitis media is challenging in primary care.
Existing diagnostic methods may lack efficiency and accessibility in non-specialist settings.

Purpose Of The Study

To develop and evaluate an AI-powered system for the early detection of ten middle ear conditions using otoscopic images.
To improve the diagnostic capabilities for middle ear pathologies in primary care and telemedicine.

Main Methods

Development of an AI system leveraging Azure OpenAI's GPT-4 Vision, a multimodal large language model (LLM).
Implementation using a Model-View-Controller (MVC) architecture for efficient image processing.
Analysis of otoscopic images to identify ten distinct middle ear conditions.

Main Results

The AI system successfully analyzes otoscopic images and detects ten middle ear conditions.
Image processing time is under 5 seconds, ensuring rapid diagnostic support.
Bilingual (English/Chinese) reports are generated, including confidence scores and treatment recommendations.

Conclusions

The AI-powered system offers a scalable and efficient solution for diagnosing middle ear diseases.
Integration of this technology can significantly enhance diagnostic workflows in telemedicine and primary care.
This multimodal LLM approach represents a novel advancement in automated otoscopic image analysis.

Keywords:

Artificial Intelligence Detection Imaging Diagnostic Imaging Large Language Models Middle Ear Effusion Mobile Applications Otitis Media Otoendoscopy Point-of-Care Systems Telemedicine