June 2023
Volume 64, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2023
Device adaptation of optical coherence tomography (OCT) retinal layer segmentation algorithm using unlabeled target data
Author Affiliations & Notes
  • Yusuke Kikuchi
    Genentech, Inc., South San Francisco, California, United States
  • Alvaro Gomariz
    F. Hoffmann-La Roche Ltd., Basel, Switzerland
  • Yun Li
    F. Hoffmann-La Roche Ltd., Basel, Switzerland
  • Huanxiang Lu
    F. Hoffmann-La Roche Ltd., Basel, Switzerland
  • Thomas Albrecht
    F. Hoffmann-La Roche Ltd., Basel, Switzerland
  • Daniela Ferrara
    Genentech, Inc., South San Francisco, California, United States
  • Footnotes
    Commercial Relationships   Yusuke Kikuchi Genentech, Inc., Code E (Employment), Roche, Code I (Personal Financial Interest); Alvaro Gomariz Roche, Code E (Employment), Roche, Code I (Personal Financial Interest); Yun Li Roche, Code E (Employment), Roche, Code I (Personal Financial Interest); Huanxiang Lu Roche, Code E (Employment), Roche, Code I (Personal Financial Interest); Thomas Albrecht Roche, Code E (Employment), Roche, Code I (Personal Financial Interest); Daniela Ferrara Genentech, Inc., Code E (Employment), Roche, Code I (Personal Financial Interest)
  • Footnotes
    Support  Genentech, Inc., a member of the Roche group, South San Francisco, CA, provided support for the study and participated in the study design; conducting the study; and data collection, management, and interpretation.
Investigative Ophthalmology & Visual Science June 2023, Vol.64, 1107. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yusuke Kikuchi, Alvaro Gomariz, Yun Li, Huanxiang Lu, Thomas Albrecht, Daniela Ferrara; Device adaptation of optical coherence tomography (OCT) retinal layer segmentation algorithm using unlabeled target data. Invest. Ophthalmol. Vis. Sci. 2023;64(8):1107.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Retinal layer segmentation deep learning (DL) models trained on data from a source OCT device usually do not generalize to another target device due to the differences in visual appearance. The alternative of labeling OCT images from the target device is time consuming and costly. This study evaluates a semi-supervised contrastive learning method that requires no labeled images from the target device, which can support more efficient development of DL models across multiple OCT devices.

Methods : SegCLR is a semi-supervised contrastive learning method with proven success in retinal fluid segmentation where supervised learning and contrastive learning are performed simultaneously. SegCLR was evaluated in a layer segmentation problem (source device: Heidelberg Spectralis; target device: ZEISS Cirrus). The data sets were obtained from 2 clinical trials evaluating neovascular age-related macular degeneration: Avenue (NCT02484690; Spectralis) and Harbor (NCT00891735; Cirrus). Inner limiting membrane, outer plexiform layer, retinal pigment epithelium, and Bruch’s membrane were manually annotated by expert graders in a subset of both trials (Table 1). The performance of SegCLR was compared against a baseline (BL) model trained on labeled data from Avenue, and an upper bound (UB) model trained on labeled data from Harbor. Labeled Harbor data were only used for evaluations and training of the UB model (Fig 1). The data split was done at the patient level. The Dice coefficient on Harbor test set averaged over the 4 layers is reported as a performance metric.

Results : The mean Dice coefficient (standard deviation [SD] across 5 replicates) in the Harbor test set was 70.15 (15.97) for the BL model, 83.77 (0.19) for the UB model, and 83.16 (0.75) for the SegCLR model. SegCLR improved the performance by more than 13% from the BL model and performed similarly to the UB model in mean Dice coefficient. Comparing SD, SegCLR largely gained stability from the BL model.

Conclusions : SegCLR shows the ability to learn from unlabeled target data in OCT layer segmentation, enabling the model to quickly and accurately adapt to new OCT devices for retinal layer segmentation in the future, which can contribute to bringing efficiency and reliability to the clinical assessment of OCT images.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

Fig 1. Relationship between training sets and models

Fig 1. Relationship between training sets and models

 

Table 1. Size of datasets

Table 1. Size of datasets

×
×

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.

×