EyeModelHub
Home Subspecialties Oculoplastics Model card
§Model Card

Orbital Volume Segmentation AI (Human-in-the-Loop U-Net)

Automatic bony-orbit segmentation for orbital volume measurement

Chang Y.J. · 2025 · Republic of Korea · doi:10.1016/j.jcms.2025.01.007
Open DOI
Segmentation CNN CT Internally validated
More attributes
Anatomy · OrbitConditions · Orbital volume assessmentSingle-centern = 349 patients · 349 CT images images
§01Overview

Brief overview

This study developed a U-Net–based deep learning platform with multiple pretrained CNN encoders (including MobileNet-V2, ResNet34/50, InceptionResNet-V2, and EfficientNet-B0) to segment orbital volume from 3D facial CT scans. Across 349 retrospectively collected CTs from a single Korean institution, the final MobileNet-V2 U-Net achieved Dice coefficients up to 94.1% with inference times of ~33 ms per axial slice. The authors also evaluated a human-in-the-loop semi-automated workflow, showing faster annotation times than manual segmentation without loss of accuracy. No external validation, prospective testing, public code release, or clinical deployment was reported. The model was designed as a research-stage decision-support tool for orbital surgical planning and volumetric assessment.

§02Performance
Headline performance

Manual-only dataset (178 CTs): Dice ≈ 0.90.

Full metrics — as reported
Manual-only dataset (178 CTs): Dice ≈ 0.90.
Full dataset with AI-assisted refinement (349 CTs): best Dice = 0.94 (MobileNetV2 U-Net).
Reader study: AI-assisted segmentation improved Dice (90.3% vs 88.7%) and reduced annotation time by 22–80%.
Calibration · Uncertainty
Not reported.
Bias · Fairness
Not reported.
§03Data & Validation
Study designRetrospective cohort
Center typeSingle-center
Patients349
Images / eyes349 CT images
Split~60/20/20
Reporting frameworks None stated
Age range>19
Sex (% female)Not reported
Race / ethnicity reportedN
Subgroup performanceN
§04Deployment
Deployment maturity
  1. Concept
  2. Prototype
  3. Internal val.
  4. External val.
  5. Regulator
  6. Clinical
Regulatory status
None No FDA / Health Canada / CE clearance.
Artifacts & license
Code repository none
Pretrained weights none
Live demo none
License Not specified
§05Quality Review · APPRAISE-AI
Total score
48 /100 Moderate
ClinicalDataMethodRobust.Report.Repro.
Sub-domain scores
Clinical relevance 4/4 100%
Data quality 14/24 58%
Methodological conduct 8/20 40%
Robustness of results 5/20 25%
Reporting quality 11/12 92%
Reproducibility 6/20 30%
Why this score?

The APPRAISE-AI tool grades AI/ML medical studies across 24 items in 6 subdomains, scored independently by two reviewers. The total reflects the methodological transparency of the publication — not the underlying clinical value of the model.

How to read this: high scores in clinical relevance and reporting with low scores in methodological conduct, robustness, or reproducibility is the dominant pattern in ophthalmic AI today.

Quality bands: Low <40 · Moderate 40–59 · High 60–79 · Very High 80+. Most published ophthalmic AI sits in the Low–Moderate band; studies that break into High typically combine multi-center datasets, prospective evaluation, and open-source release.

§06Full Metadata
Citation & identification
Bibliographic record
Model IDoculoplastics_chang_2025
Full author listCho, J., Shon, B., Choi, K. Y., Jeong, S., & Ryu, J. Y.
JournalJ Craniomaxillofac Surg
DOI10.1016/j.jcms.2025.01.007
ConditionsOrbital volume assessment
Architecture
Model internals
Architecture familyCNN
Architecture detailU-Net with ImageNet-pretrained encoders, paired with a human-in-the-loop workflow (AI pre-segmentation + clinician refinement).
Funding & conflicts
Disclosures
FundingMSIT(Ministry of Science and ICT), Korea, under the ITRC(Information Technology Research Center) support program(IITP-2025-RS-2020-II201808) supervised by the IITP(Institute of Information & Communications Technology Planning & Evaluation); Korea Health Technology R&D Project grant (grant number: HR22C1832)
Conflicts of interestNR
Card lifecycle
Status, version, reviewers
Card statusPublished
Card versionv1.0.0
First added2026-04-28
Last reviewed2026-04-28
Reviewersattribution opt-in pending
Update history
Card changelog
  • 2026-04-28 initial publication (v1.0.0)
← Back to Oculoplastics