Related Experiment Video
Updated: Jul 25, 2025
08:08
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
106
Detection of laryngeal carcinoma during endoscopy using artificial intelligence
David J Wellenstein1, Jonathan Woodburn2, Henri A M Marres1
1Department of Otorhinolaryngology and Head and Neck Surgery, Radboud University Medical Center, Nijmegen, The Netherlands.
Head & Neck
|June 28, 2023
Summary
A new deep learning (DL) algorithm accurately detects and classifies vocal cord cancer and benign lesions in real-time during endoscopy. This AI tool shows promise for improving outpatient clinic diagnostics of laryngeal pathology.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Otolaryngology
Background:
- Vocal cord pathologies require accurate and timely diagnosis.
- Current diagnostic methods can be time-consuming and may require specialized expertise.
- The need for efficient tools for real-time laryngeal pathology detection is significant.
Purpose of the Study:
- To evaluate a self-developed deep learning (DL) algorithm for real-time localization and classification of vocal cord carcinoma and benign lesions.
- To assess the performance of the DL algorithm in a clinical setting.
Main Methods:
- The DL algorithm was trained and validated using a combination of in-house collected videos/photos and the open-access "Laryngoscope8" dataset.
- The algorithm's performance was assessed for localization and classification accuracy of vocal cord lesions.
Main Results:
- The algorithm demonstrated high sensitivity in classifying vocal cord carcinoma (71-78%) and benign vocal cord lesions (70-82%).
- The DL algorithm achieved an average frame rate of 63 frames per second, enabling real-time analysis.
- The system proved effective for real-time detection of laryngeal pathology in an outpatient clinic setting.
Conclusions:
- The developed deep learning algorithm is capable of localizing and classifying both benign and malignant laryngeal pathologies during endoscopic procedures.
- This AI-powered tool has the potential to enhance diagnostic capabilities in real-time during laryngoscopy.
