Editors Selection IGR 22-2

Clinical Examination Methods: Diagnostic Performance of SS-OCT for Angle Closure

Sarah Zhou
Benjamin Xu

Comment by Sarah Zhou & Benjamin Xu on:

95236 Diagnostic accuracy of swept source optical coherence tomography classification algorithms for detection of gonioscopic angle closure, Tan SS; Tun TA; Sultana R et al., British Journal of Ophthalmology, 2021; 0:

Find related abstracts

Anterior segment ocular coherence tomography (AS-OCT) is a non-contact, non-invasive form of ocular imaging that supports qualitative and quantitative assessment of the anterior chamber and its anatomical structures.1 AS-OCT devices are based on time-domain OCT (TD-OCT) or swept-source OCT (SS-OCT) technology, with SS-OCT providing faster imaging speed and wider scanning range.1 Recent studies have investigated the potential of both TD-OCT and SS-OCT for detecting gonioscopic angle closure.2,3

In this study, Tan and colleagues tested the predictive performance of five different classification algorithms for detecting gonioscopic angle closure based on quantitative biometric analysis of SS-OCT images. Among these algorithms, a stepwise logistic regression algorithm using measurements of anterior chamber area (ACA), lens vault (LV), and iris curvature (IC) best predicted gonioscopic angle closure. Furthermore, using this algorithm, biometric measurements from horizontal meridian scans were more predictive than scans from other meridians or an average of all eight meridians. Finally, the combination of ACA, LV, and IC provided similar predictive performance as an expanded set of biometric parameters and greater predictive performance than any single parameter on its own.

Algorithms were developed primarily with data from Asian eyes and may not be generalizable to more diverse populations

The method described in this study provides a non-contact and reproducible alternative to gonioscopy for detecting angle closure. Interestingly, a previous stepwise logistic regression algorithm developed using TD-OCT images reported slightly better predictive performance than the current algorithm (AUC 0.96 vs 0.91), although an explanation for this difference is unclear.2 Both of these studies share similar limitations. Algorithms were developed primarily with data from Asian eyes and may not be generalizable to more diverse populations. It is also important to point out that quantitative biometric analysis of AS-OCT images is only semi-automated and remains time-consuming and expertise-dependent. Finally, the possibility of false positives using these algorithms should be considered given the already-low benefit of treating most eyes with gonioscopic angle closure.4


  1. Ang M, Baskaran M, Werkmeister RM, et al. Anterior segment optical coherence tomography. Prog Retin Eye Res. 2018;66:132-156.
  2. Nongpiur ME, Haaland BA, Friedman DS, et al. Classification Algorithms Based on Anterior Segment Optical Coherence Tomography Measurements for Detection of Angle Closure. Ophthalmology. 2013;120:48-54.
  3. Xu BY, Chiang M, Chaudhary S, Kulkarni S, Pardeshi AA, Varma R. Deep Learning Classifiers for Automated Detection of Gonioscopic Angle Closure Based on Anterior Segment OCT Images. Published online 2019. 4. He M, Jiang Y, Huang S, et al. Laser peripheral iridotomy for the prevention of angle closure: a single-centre, randomised controlled trial. Lancet. 2019;393(10181):1609-1618.

Issue 22-2

Change Issue