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

AI-surrogate structure and dose correlation for left anterior descending artery in lung stereotactic body radiotherapy.

Journal of applied clinical medical physics·2026
Same author

Research through Evaluation for Large Language Model in Patient-Clinician Communications.

Research square·2026
Same author

Proton Therapy for Head and Neck Adenoid Cystic Carcinoma: A Multi-institutional Review.

International journal of particle therapy·2026
Same author

Quality-of-Life Recovery in Cushing's Disease: A Cluster Analysis.

The Canadian journal of neurological sciences. Le journal canadien des sciences neurologiques·2026
Same author

Pulsed Electrical Field Ablation Plus Intratumoral Immunotherapy for Hepatocellular Carcinoma.

Journal of immunotherapy (Hagerstown, Md. : 1997)·2026
Same author

Development and Validation of NICVD-NQ Score for Predicting Short-Term Outcomes After Primary PCI.

Catheterization and cardiovascular interventions : official journal of the Society for Cardiac Angiography & Interventions·2026

Related Experiment Video

Updated: Jan 14, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

Evaluating the impact of different deface algorithms on deep learning segmentation software performance.

Ali Ammar1, Libing Zhu1, Shep Bryan1

  • 1Department of Radiation Oncology, Mayo Clinic, Phoenix, AZ, United States.

Frontiers in Oncology
|October 23, 2025
PubMed
Summary
This summary is machine-generated.

Defacing medical images for AI training impacts on-face structure segmentation but minimally affects off-face organs crucial for radiation oncology. Further research is needed to improve AI model robustness for privacy-preserving data sharing.

Keywords:
CARINA AIDeIdentifiercomputed tomography (CT)defacinghead and neckmri_refaceradiation therapysegmentation

More Related Videos

Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.5K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K

Related Experiment Videos

Last Updated: Jan 14, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Deep Learning-Based Segmentation of Cryo-Electron Tomograms
10:25

Deep Learning-Based Segmentation of Cryo-Electron Tomograms

Published on: November 11, 2022

10.5K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K

Area of Science:

  • Medical Imaging Analysis
  • Artificial Intelligence in Radiation Oncology
  • Data Privacy in Medical Research

Background:

  • Data sharing is vital for advancing radiation oncology, especially for training AI models using medical imaging.
  • Patient privacy necessitates de-identification of medical images, including facial feature removal (defacing).
  • The impact of defacing on AI-driven organ segmentation in head-and-neck (HN) CT images requires evaluation.

Purpose of the Study:

  • To assess the effect of two defacing algorithms on AI-based organ segmentation in HN CT scans.
  • To evaluate the clinical implications of defacing-induced segmentation changes on radiation therapy planning.

Main Methods:

  • Two defacing algorithms (DeIdentifier, mri_reface_0.3.3) were applied to 50 patient CT scans.
  • Segmentation accuracy was measured using Dice Similarity Coefficient (DSC), Hausdorff Distance (HD95), and Surface Dice Similarity Coefficients (SDSC) with two AI tools (INTContour, AccuContour).
  • Dose differences (D0.01cc) were calculated to assess clinical impact; paired t-tests were used for statistical analysis.

Main Results:

  • Defacing significantly reduced segmentation accuracy for on-face structures (e.g., oral cavity, eyes) but minimally affected off-face structures (e.g., brainstem, spinal cord).
  • DeIdentifier preserved Hounsfield Units (HU) and anatomical consistency better than mri_reface, which introduced more variability.
  • Dose distribution analysis showed minimal differences (<20 cGy) for most structures, with slight variations in the Brainstem, Lips_NRG, and Brain.

Conclusions:

  • Defacing alters segmentation accuracy in facial regions but has a minimal overall impact on off-face structures and radiation therapy planning.
  • Domain adaptation techniques are recommended to enhance AI model robustness for defaced and non-defaced medical image datasets.
  • Balancing data privacy through defacing with maintaining segmentation integrity is crucial for AI in medical imaging.