EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Deep learning-based segmentation of OCT images for choroidal thickness

Raman Prasad Sah, Nimesh B Patel, Hope M Queener, Pavan K Narra, Lisa A Ostrin

Journal of Optometry · 2025

DOI: 10.1016/j.optom.2025.100556 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Optometry, 2025. Verfasst von Raman Prasad Sah und 4 weiteren Autor:innen. Beteiligte Einrichtung: University of Houston College of Optometry, 4401 MLK Blvd, Houston, TX 77204, USA. Hauptthemen laut MEDLINE: Tomography, Optical Coherence, Deep Learning, Choroid, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NEI NIH HHS, Foundation for the National Institutes of Health, University of Houston.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: To develop and validate a custom deep learning-based automated segmentation for choroidal thickness of optical coherence tomography (OCT) scans. METHODS: An in-house automated algorithm was trained on a Deeplabv3+ network, based on ResNet50, using a training set of 10,798 manually segmented OCT scans (accuracy 99.25% and loss 0.0229). A test set of 130 unique scans was segmented using manual and in-house automated methods. For manual segmentation, the choroid-sclera border was delineated by the user. For in-house automated segmentation, all borders were automatically detected by the program and manually inspected. Bland-Altman analysis, intraclass correlation coefficient (ICC), and Deming regression compared the central 1-mm diameter and 3-mm and 6-mm annuli for the two methods. The in-house method was also compared with an open-source algorithm for the test set of 130 scans. RESULTS: Mean choroidal thicknesses obtained with manual and in-house automated methods were not significantly different for the three regions (P > 0.05 for all). The fixed bias between methods ranged from -2.41 to 3.49 µm. Proportional bias ranged from -0.04 to -0.12 (P < 0.05 for all). The two methods demonstrated excellent agreement across regions (ICC: 0.96 to 0.98, P < 0.001 for all). The open-source automated method consistently resulted in thinner choroidal thickness compared to manual and in-house automated methods. CONCLUSIONS: Custom in-house deep learning automated choroid segmentation demonstrated excellent agreement and strong positive linear relationship with manual segmentation. The automated approach holds distinct advantages for estimating choroidal thickness, being more objective and efficient than the manual approach.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Tomography, Optical CoherenceDeep LearningChoroidImage Processing, Computer-AssistedHumansAlgorithmsMaleFemaleMiddle AgedReproducibility of ResultsAdultAged

Stichwörter der Autor:innen

Automated segmentationChoroidal thicknessDeep learningNeural networkOptical coherence tomography

Autor:innen und Institutionen

  • University of Houston College of Optometry, 4401 MLK Blvd, Houston, TX 77204, USA.
  • University of Houston College of Optometry, 4401 MLK Blvd, Houston, TX 77204, USA. Electronic address: Lostrin@central.uh.edu.

Förderung

  • NEI NIH HHS
  • Foundation for the National Institutes of Health
  • University of Houston
  • National Eye Institute

Zitieren

APA

Sah, R. P., Patel, N. B., Queener, H. M., Narra, P. K., & Ostrin, L. A. (2025). Deep learning-based segmentation of OCT images for choroidal thickness. Journal of Optometry. https://doi.org/10.1016/j.optom.2025.100556

Vancouver

Sah RP, Patel NB, Queener HM, Narra PK, Ostrin LA. Deep learning-based segmentation of OCT images for choroidal thickness. Journal of Optometry. 2025. doi:10.1016/j.optom.2025.100556

BibTeX
@article{Sah2025,
  author = {Sah, Raman Prasad and Patel, Nimesh B and Queener, Hope M and Narra, Pavan K and Ostrin, Lisa A},
  title = {Deep learning-based segmentation of OCT images for choroidal thickness},
  journal = {Journal of Optometry},
  year = {2025},
  doi = {10.1016/j.optom.2025.100556}
}
Kostenlos starten →