EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Machine Learning of Endoscopy Images to Identify, Classify, and Segment Sinonasal Masses

Lirit Levi iD, Kenan Ye, Maxime Fieux iD, Axel Renteria, Steven Lin, Lei Xing, Noel F Ayoub iD, Zara M Patel iD, Jayakar V Nayak iD, Peter H Hwang iD, Michael T Chang iD

International Forum of Allergy & Rhinology · 2025-05 · 15 (5) · 524-535

DOI: 10.1002/alr.23525 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Forum of Allergy & Rhinology, online erschienen am 08.01.2025. Verfasst von Lirit Levi und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Otolaryngology - Head and Neck Surgery, Stanford University School of Medicine, Stanford, California, USA. Hauptthemen laut MEDLINE: Machine Learning, Endoscopy, Nasal Polyps, Paranasal Sinus Neoplasms. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Foundation of the Hospices Civil de Lyon fellowship, Fondation Edmond Roudnitska, Philippe Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

We developed and assessed the performance of a machine learning model (MLM) to identify, classify, and segment sinonasal masses based on endoscopic appearance.

Methods

A convolutional neural network‐based model was constructed from nasal endoscopy images from patients evaluated at an otolaryngology center between 2013 and 2024. Images were classified into four groups: normal endoscopy, nasal polyps, benign, and malignant tumors. Polyps and tumors were confirmed with histopathological diagnosis. Images were annotated by an otolaryngologist and independently verified by two other otolaryngologists. We used high‐ and low‐quality images to mirror real‐world conditions. The models used for classification (EfficientNet‐B2) and segmentation (nnUNet) were trained, validated, and tested at an 8:1:1 ratio. The performance accuracy was averaged across a 10‐fold cross‐validation assessment. Segmentation accuracy was assessed via Dice similarity coefficients.

Results

A total of 1242 images from 311 patients were used. The MLM was trained, validated, and tested on 663 normal, 276 polyps, 157 benign, and 146 malignant tumors images. Overall, the model performed at 84.1 ± 4.3% accuracy in the validation set and 80.4 ± 1.7% in the test set. The model correctly identified the presence of a sinonasal mass at 90.5 ± 1.2% accuracy rate. The MLM accuracy performance rates were 86.2 ± 1.0% for polyps and 84.1 ± 1.8% for tumors. Benign and malignant tumor subclassification achieved 87.8 ± 2.1% and 94.0 ± 2.4% accuracy, respectively. Segmentation accuracies for polyps were 72.3% and 72.8% for tumors.

Conclusions

An MLM for nasal endoscopy images can perform with moderate to high accuracy in identifying, classifying, and segmenting sinonasal masses. Performance in future iterations may improve with larger and more diverse training datasets.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningEndoscopyNasal PolypsParanasal Sinus NeoplasmsParanasal SinusesNose NeoplasmsHumansNeural Networks, ComputerMaleFemaleMiddle Aged

Stichwörter der Autor:innen

artificial intelligencechronic rhinosinusitisparanasal sinus diseasespolyp

Autor:innen und Institutionen

  • Department of Otolaryngology - Head and Neck Surgery, Stanford University School of Medicine, Stanford, California, USA.
  • Hospices Civils de Lyon, Centre Hospitalier Lyon Sud, Service d'ORL, d'otoneurochirurgie et de Chirurgie Cervico-Faciale, Pierre Bénite Cedex, Lyon, France.
  • Université de Lyon, Université Lyon 1, Lyon, France.
  • Department of Medicine, Stanford University School of Medicine, Stanford, California, USA.
  • Department of Radiation Oncology-Radiation Physics, Stanford University School of Medicine, Stanford, California, USA.

ORCID-Profile: Lirit Levi, Maxime Fieux, Noel F Ayoub, Zara M Patel, Jayakar V Nayak, Peter H Hwang, Michael T Chang

Förderung

  • Foundation of the Hospices Civil de Lyon fellowship
  • Fondation Edmond Roudnitska
  • Philippe Foundation
  • France-Stanford Center
  • Association Française de Rhinologie
  • Department of Otolaryngology - Head & Neck Surgery, Stanford University School of Medicine

Zitieren

APA

Levi, L., Ye, K., Fieux, M., Renteria, A., Lin, S., Xing, L., Ayoub, N. F., Patel, Z. M., Nayak, J. V., Hwang, P. H., & Chang, M. T. (2025). Machine Learning of Endoscopy Images to Identify, Classify, and Segment Sinonasal Masses. International Forum of Allergy & Rhinology, 15(5), 524-535. https://doi.org/10.1002/alr.23525

Vancouver

Levi L, Ye K, Fieux M, Renteria A, Lin S, Xing L, et al. Machine Learning of Endoscopy Images to Identify, Classify, and Segment Sinonasal Masses. International Forum of Allergy & Rhinology. 2025;15(5):524-535. doi:10.1002/alr.23525

BibTeX
@article{Levi2025,
  author = {Levi, Lirit and Ye, Kenan and Fieux, Maxime and Renteria, Axel and Lin, Steven and Xing, Lei and Ayoub, Noel F and Patel, Zara M and Nayak, Jayakar V and Hwang, Peter H and Chang, Michael T},
  title = {Machine Learning of Endoscopy Images to Identify, Classify, and Segment Sinonasal Masses},
  journal = {International Forum of Allergy & Rhinology},
  year = {2025},
  volume = {15},
  number = {5},
  pages = {524--535},
  doi = {10.1002/alr.23525}
}
Kostenlos starten →