EUVIMEDOpdag EUVIMED
ÜbersichtsarbeitOpen access · CC BY 4.0

Clinical Applications of Multimodal Artificial Intelligence in Otolaryngology: A State‐of‐the‐Art Review

Klinische Anwendungen multimodaler künstlicher Intelligenz in der HNO-Heilkunde: State-of-the-Art-Review

Ying Jie Li iD, Flora Su, Norbert Banyi, Philip Edgcumbe, Andrew Thamboo, Ameen Amanian iD

Otolaryngology–Head and Neck Surgery · 2026-08 · 175 (2) · 304-315

DOI: 10.1002/ohn.70285 ↗

Auf einen Blick

Ein Review beschreibt 44 Studien zu multimodaler KI in der HNO-Heilkunde. Am häufigsten wurden Bild und Text kombiniert; die meisten Anwendungen dienten der Entscheidungsunterstützung.

Fragestellung

Welche klinischen Anwendungen multimodaler KI gibt es in der HNO-Heilkunde?

Methode

Review der Literatur in MEDLINE, EMBASE, SCOPUS, Cochrane Library, Web of Science und CINAHL; Suche von Beginn bis 4. März 2025 nach PRISMA-ScR.

Ergebnisse

44 Studien. Bild und Text waren die häufigste Kombination (80 %, 35/44). Kopf-Hals-Krebs war das häufigste Fachgebiet (75 %, 33/44). AUC 0,65–0,96; Genauigkeit 37 %–86 %.

Einschränkungen

Meist kleine Einzelzentrumsstudien; keine prospektive Validierung. Modellfehler, Datenbias und Sprachgrenzen wurden genannt.

Schlussfolgerung der Autor:innen

Die Anwendung multimodaler Large Language Models in der HNO-Heilkunde nimmt rasch zu. Für eine sichere, gerechte und wirksame Nutzung seien strenge Validierung und ethische Aufsicht nötig.

Kernaussagen

  • 44 Studien zu multimodaler KI in der HNO-Heilkunde
  • Häufigste Kombination: Bild und Text
  • Meist Entscheidungsunterstützung, oft ohne prospektive Validierung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Übersichtsarbeit in Otolaryngology–Head and Neck Surgery, published online on 2026-05-12. Written by Ying Jie Li and 5 co-authors. Institution: Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada. Major MEDLINE topics: Artificial Intelligence, Otolaryngology. Cited 3 times so far (Crossref); 79 references. The full text is freely available (open access, CC BY 4.0).

Auf einen BlickAbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Objective

Artificial intelligence (AI) has advanced to simultaneously process visual, auditory, and textual inputs, providing users with “multimodal” AI. Given the clinical integration potential of these tools, otolaryngologists must stay informed. This study reviews current literature on applications of multimodal AI in otolaryngology. Data Sources The MEDLINE, EMBASE, SCOPUS, Cochrane Library, Web of Science, and CINAHL databases. Review Methods Databases were searched from the date of inception to March 4, 2025, following Preferred Reporting Items for Systematic Reviews and Meta‐analyses extension for scoping reviews (PRISMA‐ScR) guidelines. Studies on any application of multimodal AI in otolaryngology were included.

Conclusions

Forty‐four studies were included, with 55% (24/44) published in 2024 and 18% (8/44) in 2025. Image and text were the most commonly combined modalities (80%, 35/44), with emerging combinations including video with vector data (2%,1/44) and omics with text and/or image (14%, 6/44). Head and neck cancer was the most common subspecialty of focus (75%, 33/44), followed by general ear, nose, and throat (ENT) (11%, 5/44). All studies applied the models for clinical education (9%, 4/44) or decision support (91%, 40/44), assessing performance in areas such as board‐style examination performance (accuracy: 37%‐86%) or disease classification and prognostication (area under the receiver operating characteristic curve [AUC] 0.65‐0.96). However, most studies were limited to small, single‐institution samples and lacked prospective validation. Model error, data set bias, and language limitations underscore the need for further refinement.

Implications for Practice

The application of multimodal large language models (LLMs) in otolaryngology is rapidly expanding. Clinicians must understand both the capabilities and limitations of these systems. Rigorous validation and ethical oversight will be essential to ensure the safe, equitable, and effective adoption in otolaryngologic care.

Subjects

MeSH terms (MEDLINE)

Artificial IntelligenceOtolaryngologyHumans

Author keywords

artificial intelligencedeep learningmultimodalnatural language processing

Authors and institutions

  • Faculty of Medicine, University of British Columbia, Vancouver, British Columbia, Canada.
  • Department of Science, University of British Columbia, Vancouver, British Columbia, Canada.
  • Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada.
  • Department of Radiology, University of British Columbia, Vancouver, British Columbia, Canada.

ORCID profiles: Ying Jie Li, Ameen Amanian

Cite

APA

Li, Y. J., Su, F., Banyi, N., Edgcumbe, P., Thamboo, A., & Amanian, A. (2026). Clinical Applications of Multimodal Artificial Intelligence in Otolaryngology: A State‐of‐the‐Art Review. Otolaryngology–Head and Neck Surgery, 175(2), 304-315. https://doi.org/10.1002/ohn.70285

Vancouver

Li YJ, Su F, Banyi N, Edgcumbe P, Thamboo A, Amanian A. Clinical Applications of Multimodal Artificial Intelligence in Otolaryngology: A State‐of‐the‐Art Review. Otolaryngology–Head and Neck Surgery. 2026;175(2):304-315. doi:10.1002/ohn.70285

BibTeX
@article{Li2026,
  author = {Li, Ying Jie and Su, Flora and Banyi, Norbert and Edgcumbe, Philip and Thamboo, Andrew and Amanian, Ameen},
  title = {Clinical Applications of Multimodal Artificial Intelligence in Otolaryngology: A State‐of‐the‐Art Review},
  journal = {Otolaryngology–Head and Neck Surgery},
  year = {2026},
  volume = {175},
  number = {2},
  pages = {304--315},
  doi = {10.1002/ohn.70285}
}
Start for free →