Deep learning models have achieved diagnostic accuracy rates up to 92% for nasopharyngeal carcinoma and over 95% for otologic pathology, matching expert performance in otolaryngology.

This review summarizes recent deep learning advances in otolaryngology, including diagnostic models with expert-level accuracy for nasopharyngeal carcinoma and otologic pathology.

This narrative review synthesizes recent deep-learning applications and proposes a framework for their integration into otolaryngology.