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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Global diversities of disability caused by breast cancer in 204 countries and territories: a secondary analysis from global burden of disease study 2021.

BMC public health·2026
Same author

Trimethylamine N-oxide detection for early prediction of renal function utilizing a three-dimensional localized electronic structure (3DLES) biosensor.

Mikrochimica acta·2025
Same author

Advances in Polyimide Membranes for Gas Separation: Synthesis, Modification, and Application.

Molecules (Basel, Switzerland)·2025
Same author

Pixel array-based urine biosensor for detection of trimethylamine N-oxide and glucose for early detection of diabetic kidney disease.

Analytica chimica acta·2025
Same author

Application of plasma cell-free DNA in screening of advanced colorectal adenoma.

European journal of medical research·2025
Same author

Multi-Instance Learning for Vocal Fold Leukoplakia Diagnosis Using White Light and Narrow-Band Imaging: A Multicenter Study.

The Laryngoscope·2024

Related Experiment Video

Updated: Jun 27, 2025

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.5K

Multi-instance learning based artificial intelligence model to assist vocal fold leukoplakia diagnosis: A multicentre

Mei-Ling Wang1, Cheng-Wei Tie2, Jian-Hui Wang3

  • 1Department of Endoscopy, National Cancer Center, National Clinical Research Center for Cancer, Cancer Hospital & Shenzhen Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Shenzhen, China.

American Journal of Otolaryngology
|May 4, 2024
PubMed
Summary

An artificial intelligence (AI) system using multi-instance learning (MIL) improved the diagnosis of vocal fold leukoplakia (VFL). This AI-assisted tool enhances otolaryngologists' ability to differentiate benign from malignant VFL, aiding clinical decisions.

Keywords:
Artificial intelligenceLaryngoscopyMulti-instance learningSegmentationVocal fold leukoplakia

More Related Videos

Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications
06:24

Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications

Published on: January 5, 2024

828
Investigating the Three-dimensional Flow Separation Induced by a Model Vocal Fold Polyp
09:58

Investigating the Three-dimensional Flow Separation Induced by a Model Vocal Fold Polyp

Published on: February 3, 2014

8.5K

Related Experiment Videos

Last Updated: Jun 27, 2025

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.5K
Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications
06:24

Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications

Published on: January 5, 2024

828
Investigating the Three-dimensional Flow Separation Induced by a Model Vocal Fold Polyp
09:58

Investigating the Three-dimensional Flow Separation Induced by a Model Vocal Fold Polyp

Published on: February 3, 2014

8.5K

Area of Science:

  • Artificial Intelligence in Medicine
  • Otolaryngology
  • Medical Image Analysis

Background:

  • Vocal fold leukoplakia (VFL) diagnosis requires differentiating benign from malignant lesions.
  • Accurate diagnosis is crucial for effective treatment planning and patient outcomes.
  • Current diagnostic methods can be subjective and may benefit from objective AI assistance.

Purpose of the Study:

  • To develop and validate a multi-instance learning (MIL) based artificial intelligence (AI) model for VFL diagnosis using laryngoscopic images.
  • To assess the AI system's performance in differentiating benign and malignant VFL.
  • To evaluate the impact of AI assistance on the diagnostic performance of otolaryngologists.

Main Methods:

  • Development and validation of an AI system using 5362 laryngoscopic images from 551 patients across three hospitals.
  • Utilized automated region of interest (ROI) segmentation for image-level features and MIL for patient-level feature fusion.
  • Compared AI-assisted diagnosis performance against otolaryngologists (senior and junior) using real-time video analysis and a human-machine comparison database.

Main Results:

  • The MIL-based AI system achieved a maximum area under the curve (AUC) of 0.869 for image-level segmentation and 0.851 for patient-level diagnosis in external validation.
  • Real-time video diagnosis using the AI system reached an AUC of 0.850.
  • AI assistance significantly improved diagnostic AUC for both senior (0.720 to 0.808) and junior (0.647 to 0.807) otolaryngologists.

Conclusions:

  • The developed MIL-based AI-assisted diagnosis system demonstrates significant potential in improving VFL diagnostic accuracy.
  • The AI system enhances the diagnostic performance of otolaryngologists, leading to more informed clinical decisions.
  • This AI tool offers a valuable adjunct for the accurate differentiation of benign and malignant vocal fold lesions.