OBJECTIVES: Diagnosis of cochlear malformation on temporal bone CT images is often difficult because the imaging findings are frequently subtle. Our aim was to assess the utility of deep learning analysis in diagnosing cochlear malformation on temporal bone CT images.
No immediate practice change; the model is a research prototype and requires prospective clinical validation before it could replace or assist expert radiological review of cochlear malformations.
Accurate pre-surgical identification of cochlear malformations is essential for cochlear implant candidacy decisions and surgical planning, and AI-assisted CT reading could reduce missed diagnoses.
- 01A deep learning model was trained to classify cochlear malformations on temporal bone CT images.
- 02Cochlear malformations are often subtle and can be missed on standard radiological review.
- 03Automated discrimination could support pre-implant workup and surgical planning.
- 04Study is developmental/validation stage; clinical deployment readiness is not yet established.
- 05Published in the Brazilian Journal of Otorhinolaryngology.
Deep learning can discriminate cochlear malformations on temporal bone CT images with improved diagnostic accuracy.
studypartially supportedImaging findings in cochlear malformations are often subtle, leading to diagnostic challenges.
opinionsupported- PMID
- 42054756
- DOI
- 10.1016/j.bjorl.2026.101811.
- Journal
- Brazilian Journal of Otorhinolaryngology
- Publication type
- research_article
- Evidence level
- 4
- Population
- Temporal bone CT images of patients with cochlear malformations
- Intervention
- Deep learning model for cochlear malformation classification
Primary outcomes
Diagnostic accuracy of cochlear malformation discrimination on CT
