Deep learning multi-classification of middle ear diseases using synthetic tympanic images

Yoshimaru Mizoguchi1,2, Taku Ito1, Masato Yamada2

  • 1Department of Otorhinolaryngology, Institute of Science Tokyo, Tokyo, Japan.

Acta Oto-Laryngologica
|January 11, 2025
PubMed
Summary

Generative adversarial networks (GANs) did not improve AI diagnostic accuracy for middle ear diseases when combined with real images. However, AI models trained solely on synthetic tympanic membrane images achieved 70% accuracy.