Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 9 min ago
Peer-reviewedCC BYFrontiers in medicine

Beyond tessellation: AI-derived choroidal vessel density on multi-channel cSLO as a surrogate biomarker for OCT choroidal parameters.

  • Yuou Yao1,2
  • Xi Wang3
  • Tengfei Long4
  • Chaohong Li5
  • Saiguang Ling3
  • Enzhong Jin1,2
  • Yaoyao Sun1,2
  • Shirong Su1
  • Jianyu Xiong1
  • Jiyang Tang1,2
  • Xiang Meng1,2
  • Yi Cai1,2
3 more authors
  • Qiaozhu Zeng1,2
  • Heng Li6
  • Heng Miao1,2
6 affiliations
  1. 1Department of Ophthalmology, Peking University People's Hospital, Beijing, China.
  2. 2Beijing Key Laboratory of Ocular Disease and Optometry Science, Beijing, China.
  3. 3EVision Technology (Beijing) Co. Ltd., Beijing, China.
  4. 4Aerospace Information Research Institute, Chinese Academy of Sciences (CAS), Beijing, China.
  5. 5Suzhou Microclear Medical Instruments Co., Ltd., Suzhou, Jiangsu, China.
  6. 6Faculty of Biomedical Engineering, Shenzhen University of Advanced Technology, Shenzhen, China.
Read full textPDFDOICite

Abstract

Purpose

To develop an AI-based method for quantifying choroidal vessel density (CVD) from ultra-widefield confocal scanning laser ophthalmoscopy (UWF-cSLO) and to validate this metric against swept-source optical coherence tomography (SS-OCT) choroidal parameters.Methods

In this retrospective cross-sectional study, 633 eyes from 354 participants underwent simultaneous UWF-cSLO and SS-OCT imaging on an integrated platform, ensuring precise spatial co-registration. A TransUNet deep-learning model segmented choroidal vessels from red-channel (R; 665 nm) and infrared-channel (IR; 785 nm) images, yielding channel-specific metrics (R-CVD and IR-CVD) and a combined metric (IR + R CVD). Eight SS-OCT parameters-including choroidal thickness (CT), choroidal vascularity index (CVI), luminal area, stromal area (SA), total choroidal area (TCA), and vessel volume index (VVI)-served as reference measures. Associations were evaluated using Spearman correlations with participant-level clustered bootstrap confidence intervals, multivariable generalised estimating equations (GEE) adjusted for age and sex, and receiver operating characteristic (ROC) analysis.Results

Segmentation accuracy exceeded 0.95 (Dice coefficient) in both channels. R-CVD exhibited the strongest inverse correlations with structural and volumetric OCT parameters (SA: ρ = -0.619; CT: ρ = -0.615; TCA: ρ = -0.615; all q < 0.001), whereas IR + R CVD correlated most closely with proportion-based indices (VVI: ρ = -0.34; CVI: ρ = -0.29). These channel-dependent patterns persisted after adjustment for age and sex in multivariable GEE models. On ROC analysis, R-CVD discriminated thin from thick choroids with an area under the curve (AUC) of 0.789 (sensitivity 72.5%; specificity 74.8%), whereas IR-CVD alone showed near-chance performance (AUC 0.517).Conclusions

AI-derived CVD from UWF-cSLO correlates strongly with SS-OCT choroidal parameters in a wavelength-dependent manner and may serve as an accessible surrogate for choroidal structural assessment in settings without OCT.

Related

  1. Interocular symmetry of retinal and choroidal OCT parameters in precapillary pulmonary hypertension: a brief research report
    Diana A. Dmuchowska, Katarzyna Ptaszyńska, Joanna Konopińska +1
    Frontiers in medicinePeer-reviewedCC BYFull textGeneral medicineEurope PMC
  2. Multimodal Management of Isolated Pulmonary Metastasis From Choroidal Melanoma: A Case Report
    Francisca Abreu, Rita Amorim e Costa, Tiago Figueiredo +2
    CureusPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  3. Short-term choroidal thickness modulation with SAT-001 digital software intervention in pediatric myopia: a randomized pilot study
    So Young Han, Jarang Hahm, Mi Kyoung Bae +2
    Scientific ReportsPeer-reviewedCC BY-NC-ND 4.0Full textEpidemiology & public healthCrossref
  4. Postprocessing of Structural En Face OCT Images of the Choroidal Haller Layer for Quantitative Assessment: An Exploratory Pilot Study
    Anait S. Khalatyan, Yusef Yusef, Konstantin S. Avetisov +7
    BiomedicinesPeer-reviewedCC BYFull textGeneral medicineEurope PMC
  5. Spatial Choroidal Analysis under Mid-Phase Hyperfluorescent Plaques Demonstrates Unique Signatures in Central Serous Chorioretinopathy: Macula Society International Central Serous Chorioretinopathy Research Network Report Number 20
    Giulia Gregori, Niroj Kumar Sahoo, Nasiq Hasan +14
    Ophthalmology sciencePeer-reviewedCC BYFull textGeneral medicineEurope PMC

Cite

BibTeXRIS
BibTeX
@article{yao2026beyond,
  title = {{Beyond tessellation: AI-derived choroidal vessel density on multi-channel cSLO as a surrogate biomarker for OCT choroidal parameters.}},
  author = {Yuou Yao and Xi Wang and Tengfei Long and Chaohong Li and Saiguang Ling and Enzhong Jin and Yaoyao Sun and Shirong Su and Jianyu Xiong and Jiyang Tang and Xiang Meng and Yi Cai and Qiaozhu Zeng and Heng Li and Heng Miao},
  journal = {Frontiers in medicine},
  year = {2026},
  month = oct,
  doi = {10.3389/fmed.2026.1884151},
  url = {https://europepmc.org/article/MED/42909305},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Beyond tessellation: AI-derived choroidal vessel density on multi-channel cSLO as a surrogate biomarker for OCT choroidal parameters.
AU  - Yuou Yao
AU  - Xi Wang
AU  - Tengfei Long
AU  - Chaohong Li
AU  - Saiguang Ling
AU  - Enzhong Jin
AU  - Yaoyao Sun
AU  - Shirong Su
AU  - Jianyu Xiong
AU  - Jiyang Tang
AU  - Xiang Meng
AU  - Yi Cai
AU  - Qiaozhu Zeng
AU  - Heng Li
AU  - Heng Miao
PY  - 2026
DA  - 2026/10/07
JO  - Frontiers in medicine
DO  - 10.3389/fmed.2026.1884151
UR  - https://europepmc.org/article/MED/42909305
AB  - Purpose  To develop an AI-based method for quantifying choroidal vessel density (CVD) from ultra-widefield confocal scanning laser ophthalmoscopy (UWF-cSLO) and to validate this metric against swept-source optical coherence tomography (SS-OCT) choroidal parameters.Methods  In this retrospective cross-sectional study, 633 eyes from 354 participants underwent simultaneous UWF-cSLO and SS-OCT imaging on an integrated platform, ensuring precise spatial co-registration. A TransUNet deep-learning model segmented choroidal vessels from red-channel (R; 665 nm) and infrared-channel (IR; 785 nm) images, yielding channel-specific metrics (R-CVD and IR-CVD) and a combined metric (IR + R CVD). Eight SS-OCT parameters-including choroidal thickness (CT), choroidal vascularity index (CVI), luminal area, stromal area (SA), total choroidal area (TCA), and vessel volume index (VVI)-served as reference measures. Associations were evaluated using Spearman correlations with participant-level clustered bootstrap confidence intervals, multivariable generalised estimating equations (GEE) adjusted for age and sex, and receiver operating characteristic (ROC) analysis.Results  Segmentation accuracy exceeded 0.95 (Dice coefficient) in both channels. R-CVD exhibited the strongest inverse correlations with structural and volumetric OCT parameters (SA: ρ = -0.619; CT: ρ = -0.615; TCA: ρ = -0.615; all q < 0.001), whereas IR + R CVD correlated most closely with proportion-based indices (VVI: ρ = -0.34; CVI: ρ = -0.29). These channel-dependent patterns persisted after adjustment for age and sex in multivariable GEE models. On ROC analysis, R-CVD discriminated thin from thick choroids with an area under the curve (AUC) of 0.789 (sensitivity 72.5%; specificity 74.8%), whereas IR-CVD alone showed near-chance performance (AUC 0.517).Conclusions  AI-derived CVD from UWF-cSLO correlates strongly with SS-OCT choroidal parameters in a wavelength-dependent manner and may serve as an accessible surrogate for choroidal structural assessment in settings without OCT.
SN  - 2296-858X
ER  -