Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jul 16, 2026

Testing a Cochlear Implant Electrode Insertion Training System for Optimal Electrode Array Placement in Different Inner Ear Anatomies
07:34

Testing a Cochlear Implant Electrode Insertion Training System for Optimal Electrode Array Placement in Different Inner Ear Anatomies

Published on: February 6, 2026

Deep learning for subtype classification of inner ear malformations on temporal bone HRCT: Development and

Xiaoge Li1, Xing Zhao2, Qianyu Hao3

  • 1Senior Department of Otolaryngology Head and Neck Surgery, the Sixth Medical Center of Chinese PLA General Hospital/State Key Laboratory of Hearing and Balance Science/National Clinical Research Center for Otolaryngologic Diseases/Key Laboratory of Hearing Science, Ministry of Education/Beijing Key Laboratory of Hearing Impairment Prevention and Treatment, Beijing 100048, China.

European Journal of Radiology
|July 14, 2026
PubMed

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Thin-Layer NbOCl<sub>2</sub> Flat-Band Semiconductor for Efficient PEC Photodetection.

The journal of physical chemistry letters·2026
Same author

An in situ non-contact 3D microscopic measurement system for temporal bone anatomy based on stereo imaging.

Scientific reports·2026
Same author

Annealing of skyrmion lattice in van der Waals magnet via field modulation.

Nature communications·2026
Same author

Genotype-phenotype correlation-driven precision management in hereditary conductive and mixed hearing loss.

European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery·2026
Same author

Room-temperature two-dimensional multiferroic metal with voltage-controllable magnetic order.

Nature materials·2026
Same author

Letter to the Editor: Beyond general scales - parent-child relationships in cultural contexts and adolescent suicide risk.

Child and adolescent mental health·2026
Summary

A deep learning model accurately classifies inner ear malformations (IEMs) from CT scans, outperforming human experts. This AI tool aids in preoperative risk assessment and surgical planning for IEMs.

Area of Science:

  • Radiology
  • Artificial Intelligence
  • Medical Imaging

Background:

  • Accurate classification of inner ear malformations (IEMs) using temporal bone high-resolution computed tomography (HRCT) is crucial for surgical planning.
  • Expertise in subspecialty radiology is often required for precise IEM diagnosis.

Purpose of the Study:

  • To develop and externally validate a deep learning (DL) model for multiclass diagnosis of IEMs on temporal bone HRCT.
  • To assess the model's performance against otolaryngologists in classifying IEM subtypes.

Main Methods:

  • A weakly supervised DL framework using a Transformer-based multiple instance learning architecture was developed.
  • The model was trained and validated on a multicenter cohort of 9,182 temporal bone HRCT scans (3,161 IEMs, 6,021 normal).
Keywords:
CTClinical decision supportDeep learningInner ear malformationsMulticenter validationMultiple instance learningSubtype classification

More Related Videos

High-Speed Human Temporal Bone Sectioning for the Assessment of COVID-19-Associated Middle Ear Pathology
03:42

High-Speed Human Temporal Bone Sectioning for the Assessment of COVID-19-Associated Middle Ear Pathology

Published on: May 18, 2022

Related Experiment Videos

Last Updated: Jul 16, 2026

Testing a Cochlear Implant Electrode Insertion Training System for Optimal Electrode Array Placement in Different Inner Ear Anatomies
07:34

Testing a Cochlear Implant Electrode Insertion Training System for Optimal Electrode Array Placement in Different Inner Ear Anatomies

Published on: February 6, 2026

High-Speed Human Temporal Bone Sectioning for the Assessment of COVID-19-Associated Middle Ear Pathology
03:42

High-Speed Human Temporal Bone Sectioning for the Assessment of COVID-19-Associated Middle Ear Pathology

Published on: May 18, 2022

  • Performance was evaluated against senior and junior otolaryngologists, with explainability assessed via attention-based slice ranking and Grad-CAM.

    • Main Results:

    • The DL model achieved high accuracy (93.9%) and macro-F1 score (88.2%) on an independent external test set.
    • The model outperformed both senior (75.7%) and junior (62.9%) otolaryngologists in the reader study.
    • Interpretable outputs, including contributing slices and Grad-CAM heatmaps, were generated by the model.

    Conclusions:

    • The developed weakly supervised DL framework provides accurate and generalizable subtype-level diagnosis of IEMs from temporal bone HRCT.
    • This AI tool can serve as a practical decision-support system for standardized preoperative assessment, especially where subspecialty expertise is limited.