July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Improvement on 360° view of Irido-Corneal Angle in Automated Gonioscopy
Author Affiliations & Notes
  • Mauro Campigotto
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Silvia Rossi
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Lorenzo Cappellari
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Andrea De Giusti
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Anna Paviotti
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Nikita Scattolin
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Andrea Giaretta
    NIDEK Technologies Srl, Albignasego, PD, Italy
  • Footnotes
    Commercial Relationships   Mauro Campigotto, NIDEK Technologies Srl (E); Silvia Rossi, NIDEK Technologies Srl (E); Lorenzo Cappellari, NIDEK Technologies Srl (E); Andrea De Giusti, NIDEK Technologies Srl (E); Anna Paviotti, NIDEK Technologies Srl (E); Nikita Scattolin, NIDEK Technologies Srl (E); Andrea Giaretta, NIDEK Technologies Srl (E)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 5555. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mauro Campigotto, Silvia Rossi, Lorenzo Cappellari, Andrea De Giusti, Anna Paviotti, Nikita Scattolin, Andrea Giaretta; Improvement on 360° view of Irido-Corneal Angle in Automated Gonioscopy. Invest. Ophthalmol. Vis. Sci. 2019;60(9):5555.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : NIDEK GS-1 (NIDEK CO., LTD Japan) allows the acquisition of 16 partially overlapped irido-corneal angle images, which can be automatically combined to obtain a 360° view of the angle called “stitching”. This image can present some discontinuities or artifacts, due to misalignments between patient’s eye and the device during exam. The aim of this work is to verify the use of additional anatomical information given by Trabecular Meshwork (TM) position to improve the entire angle view.

Methods : To build the stitching view, each single image is preprocessed in order to extract 2 Regions Of Interest (ROIs), used to register an image with the previous and the successive one, respectively. To achieve a correct registration, and thus a good stitching image, ROIs must contain matching features. In GS-1, on each image the ROI top-left corners are currently selected according to the brightest point (“peak”) along the oriented median line, i.e. the line passing through the image center, having the orientation of the image acquisition prism. The use of anatomical information like the TM position as the reference point for ROI selection potentially improves the overlapping between the ROIs. Of course, the reliability of TM detection is crucial to such purpose. In GS-1, the TM feature is automatically extracted for each image as the intersection point between the TM curve and the oriented median line. However, failures or false positives may occur. As part of the current work, a manual refinement of the TM selection has been introduced. The TM-based stitching algorithm has been applied to 30 sets of images and has been visually compared with the peak-based stitching results.

Results : The stitching quality has been assessed by verifying the absence of discontinuities in the TM line. In all the considered cases, the stitching result has been improved by the usage of reliable TM information (Fig. 1b, 2b) with respect to the standard approach (Fig. 1a, 2a).

Conclusions : A correct 360° view of the irido-corneal angle is important in the identification/localization of anatomical features, pathological evidence and implants like MIGS. Using the TM information in the ROIs computation for image registration reduces the occurrence of discontinuities in the stitching visualization. The primary assumption is the correctness of the detected TM points, which can be manually refined if automatic result is not satisfying.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

 

×
×

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.

×