June 2021
Volume 62, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2021
Automated Vessel Segmentation in Adaptive Optics – Optical Coherence Tomography Images
Author Affiliations & Notes
  • Christopher Le
    University of Maryland School of Medicine, Baltimore, Maryland, United States
  • Dongyi Wang
    University of Maryland at College Park, College Park, Maryland, United States
  • Ricardo Villanueva
    University of Maryland School of Medicine, Baltimore, Maryland, United States
  • Zhuolin Liu
    US Food and Drug Administration, Silver Spring, Maryland, United States
  • Daniel Hammer
    US Food and Drug Administration, Silver Spring, Maryland, United States
  • Osamah Saeedi
    Ophthalmology, University of Maryland School of Medicine, Baltimore, Maryland, United States
  • Footnotes
    Commercial Relationships   Christopher Le, None; Dongyi Wang, None; Ricardo Villanueva, None; Zhuolin Liu, Indiana University (P); Daniel Hammer, None; Osamah Saeedi, Heidelberg (F), Heidelberg (R), NIH/NEI (F), Vasoptic Medical, Inc. (F)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2021, Vol.62, 13. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Christopher Le, Dongyi Wang, Ricardo Villanueva, Zhuolin Liu, Daniel Hammer, Osamah Saeedi; Automated Vessel Segmentation in Adaptive Optics – Optical Coherence Tomography Images. Invest. Ophthalmol. Vis. Sci. 2021;62(8):13.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To demonstrate automated capillary segmentation in adaptive optics – optical coherence tomography (AO-OCT) images.

Methods : AO-OCT volumes were acquired from the FDA multimodal AO imager focused on the inner retina. A trained grader generated the vessel plexus projections from the averaged AO-OCT volumes. We trained a UNet-based convolutional neural network to segment retinal capillaries in en face projections of the superficial vascular plexus (SVP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP). The network was trained with random 128x128 pixel patches from 177 automatically contrast-corrected projections of each plexus from 18 eyes in 18 subjects with a 20% validation split. We trained four models: one with all-plexus images and three with plexus-specific images only. The models’ results were compared with a trained grader's manual capillary segmentation. We evaluated segmentation performance based on Dice coefficient (DC), pixel-wise precision, recall, and accuracy on a held-out 20% test split.

Results : Our all-plexus model achieved good segmentation results with an overall DC of 0.701, precision of 0.686, recall of 0.749, and accuracy of 0.929. All-plexus and plexus-specific model performance had comparable results for each plexus, with slightly better DC in the all-plexus model in SVP and ICP, but worse in DCP projections.

Conclusions : This is the first application of deep learning techniques for automated vessel segmentation in AO-OCT projections, achieving a good DC, precision, recall, and accuracy and demonstrating the applicability of segmentation techniques to this imaging modality. Difference in results across plexus-specific models may reflect a trade-off in training sample size and task-relevance; however, statistical significance of these results is not determined from this study. Future work is needed to further understand whether automated segmentation should be optimized for each plexus.

This is a 2021 ARVO Annual Meeting abstract.

 

Train and test results for models trained using images from all plexuses (with broken down results for each distinct plexus), SVP only, ICP only, and the DCP only. For ease of viewing, bolded results indicate improved performance over all-plexus or plexus-specific counterpart models.

Train and test results for models trained using images from all plexuses (with broken down results for each distinct plexus), SVP only, ICP only, and the DCP only. For ease of viewing, bolded results indicate improved performance over all-plexus or plexus-specific counterpart models.

 

Representative raw AO-OCT projections, contrast-corrected images, manually generated masks, and plexus-specific and all-plexus model results with each row representing the associated distinct plexus

Representative raw AO-OCT projections, contrast-corrected images, manually generated masks, and plexus-specific and all-plexus model results with each row representing the associated distinct plexus

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×