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

Topical Therapy for Atopic Dermatitis: What is New and the New Paradigm.

Immunology and allergy clinics of North America·2026
Same author

Normal Hearing Thresholds in Korean Adults Aged 20-79 Years: Establishing Reference Values and Comparison With International Organization for Standardization 7029:2017.

American journal of audiology·2026
Same author

Development of Hearing Information Booklet for Dementia Healthcare Professionals.

Journal of audiology & otology·2026
Same author

Engineering MSC Migration: Roles of Nanoparticles in Activating Migratory Pathways and Functions.

International journal of molecular sciences·2026
Same author

Post-traumatic benign paroxysmal positional vertigo: mechanisms, clinical phenotypes, and a structured clinical pathway for management.

Frontiers in neurology·2026
Same author

Feasibility study on noise attenuation and stability of active noise cancelling headphones for mobile hearing screening.

Hearing research·2026

Related Experiment Video

Updated: Sep 19, 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

1.6K

Digitizing audiograms with deep learning: structured data extraction and pseudonymization for hearing big data.

Sunghwa You1, Chanbeom Kwak2, Chul Young Yoon1

  • 1Research Institute of Hearing Enhancement, Yonsei University Wonju College of Medicine, Wonju, South Korea; Department of Medical Informatics and Biostatistics, Yonsei University Wonju College of Medicine, Wonju, South Korea.

Hearing Research
|June 18, 2025
PubMed
Summary

This study introduces a deep learning system to digitize pure-tone audiometry (PTA) images, converting them into structured data for hearing big data initiatives. The AI model significantly improves efficiency and accuracy, enabling better integration into electronic medical records.

Keywords:
AudiogramDeep learning algorithmsDigitizationHearing big dataPseudonymization

More Related Videos

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K
Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique
11:39

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Published on: September 7, 2022

2.3K

Related Experiment Videos

Last Updated: Sep 19, 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

1.6K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.3K
Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique
11:39

Assessment of Audio-Tactile Sensory Substitution Training in Participants with Profound Deafness Using the Event-Related Potential Technique

Published on: September 7, 2022

2.3K

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Healthcare
  • Audiology

Background:

  • Pure-tone audiometry (PTA) is crucial for diagnosing hearing loss, but audiogram images are often unstructured.
  • This lack of structure hinders integration into electronic medical records (EMRs) and common data models (CDMs), limiting large-scale data analysis.
  • Developing automated digitization methods is essential for advancing hearing big data research.

Purpose of the Study:

  • To develop and validate a deep learning-based system for the automated digitization of audiograms.
  • To convert unstructured audiogram images into structured, numerical data suitable for EMRs and CDMs.
  • To enhance the accessibility and scalability of hearing data for clinical and research applications.

Main Methods:

  • A convolutional neural network (CNN) was trained to extract frequency and threshold values from audiograms.
  • The system incorporated modules for preprocessing, pattern classification, image analysis, and post-processing.
  • Optical character recognition (OCR) was used for patient data extraction, followed by pseudonymization to ensure data privacy.

Main Results:

  • The deep learning model achieved high accuracy: 95.01% for the right ear and 98.18% for the left ear.
  • Digitization speed increased by 17.72 times compared to manual methods, reducing processing time per audiogram from 63.27s to 3.57s.
  • The resulting structured data format facilitates seamless integration into big data platforms and CDMs, adhering to privacy regulations.

Conclusions:

  • The developed system significantly improves the efficiency and accuracy of audiogram digitization.
  • This facilitates the creation of comprehensive hearing big data, supports AI-driven diagnostics, and enables large-scale hearing data analysis.
  • The framework ensures structured numerical data output, overcoming limitations of previous classification-focused studies and complying with data pseudonymization requirements.