EUVIMEDEUVIMED entdecken
Crossref

Laryngeal Cancer Screening During Flexible Video Laryngoscopy Using Large Computer Vision Models

Ishwarya S Mamidi, Michael E Dunham iD, Lacey K Adkins, Andrew J McWhorter, Zhide Fang, Britney T Banh

Annals of Otology, Rhinology & Laryngology · 2024-08 · 133 (8) · 720-728

DOI: 10.1177/00034894241253376 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of Otology, Rhinology & Laryngology, online erschienen am 16.05.2024. Verfasst von Ishwarya S Mamidi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Otolaryngology-Head and Neck Surgery, Louisiana State University Health Sciences Center, New Orleans, LA, USA. Hauptthemen laut MEDLINE: Laryngeal Neoplasms, Laryngoscopy, Video Recording, Early Detection of Cancer. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 35 Literaturangaben. Gefördert von: Louisiana State University Health Sciences Center Resident Research Grant.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

Develop an artificial intelligence assisted computer vision model to screen for laryngeal cancer during flexible laryngoscopy.

Methods

Using laryngeal images and flexible laryngoscopy video recordings, we developed computer vision models to classify video frames for usability and cancer screening. A separate model segments any identified lesions on the frames. We used these computer vision models to construct a video stream annotation system. This system classifies findings from flexible laryngoscopy as “potentially malignant” or “probably benign” and segments any detected lesions. Additionally, the model provides a confidence level for each classification.

Results

The overall accuracy of the flexible laryngoscopy cancer screening model was 92%. For cancer screening, it achieved a sensitivity of 97.7% and a specificity of 76.9%. The segmentation model attained an average precision at a 0.50 intersection-over-union of 0.595. The confidence level for positive screening results can assist clinicians in counseling patients regarding the findings.

Conclusion

Our model is highly sensitive and adequately specific for laryngeal cancer screening. Segmentation helps endoscopists identify and describe potential lesions. Further optimization is required to enable the model’s deployment in clinical settings for real-time annotation during flexible laryngoscopy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Laryngeal NeoplasmsLaryngoscopyVideo RecordingEarly Detection of CancerArtificial IntelligenceHumansSensitivity and Specificity

Stichwörter der Autor:innen

artificial intelligencecancer screeningcomputer visionflexible laryngoscopylaryngeal cancer

Autor:innen und Institutionen

  • Department of Otolaryngology-Head and Neck Surgery, Louisiana State University Health Sciences Center, New Orleans, LA, USA.
  • Biostatistics Program, School of Public Health, Louisiana State University Health Sciences Center, New Orleans LA, USA.
  • Our Lady of the Lake Voice Center, Our Lady of the Lake Regional Medical Center, Baton Rouge, LA, USA.

ORCID-Profile: Michael E Dunham

Förderung

  • Louisiana State University Health Sciences Center Resident Research Grant

Zitieren

APA

Mamidi, I. S., Dunham, M. E., Adkins, L. K., McWhorter, A. J., Fang, Z., & Banh, B. T. (2024). Laryngeal Cancer Screening During Flexible Video Laryngoscopy Using Large Computer Vision Models. Annals of Otology, Rhinology & Laryngology, 133(8), 720-728. https://doi.org/10.1177/00034894241253376

Vancouver

Mamidi IS, Dunham ME, Adkins LK, McWhorter AJ, Fang Z, Banh BT. Laryngeal Cancer Screening During Flexible Video Laryngoscopy Using Large Computer Vision Models. Annals of Otology, Rhinology & Laryngology. 2024;133(8):720-728. doi:10.1177/00034894241253376

BibTeX
@article{Mamidi2024,
  author = {Mamidi, Ishwarya S and Dunham, Michael E and Adkins, Lacey K and McWhorter, Andrew J and Fang, Zhide and Banh, Britney T},
  title = {Laryngeal Cancer Screening During Flexible Video Laryngoscopy Using Large Computer Vision Models},
  journal = {Annals of Otology, Rhinology & Laryngology},
  year = {2024},
  volume = {133},
  number = {8},
  pages = {720--728},
  doi = {10.1177/00034894241253376}
}
Kostenlos starten →