Elsevier

Ophthalmology

Volume 120, Issue 1, January 2013, Pages 48-54
Ophthalmology

Original article
Classification Algorithms Based on Anterior Segment Optical Coherence Tomography Measurements for Detection of Angle Closure

https://doi.org/10.1016/j.ophtha.2012.07.005Get rights and content

Objective

A recent study found that a combination of 6 anterior segment optical coherence tomography (ASOCT) parameters (anterior chamber area, volume, and width [ACA, ACV, ACW], lens vault [LV], iris thickness at 750 μm from the scleral spur, and iris cross-sectional area) explain >80% of the variability in angle width. The aim of this study was to evaluate classification algorithms based on ASOCT measurements for the detection of gonioscopic angle closure.

Design

Cross-sectional study.

Participants

We included 2047 subjects aged ≥50 years.

Methods

Participants underwent gonioscopy and ASOCT (Carl Zeiss Meditec, Dublin, CA). Customized software (Zhongshan Angle Assessment Program, Guangzhou, China) was used to measure ASOCT parameters in horizontal ASOCT scans. Six classification algorithms were considered (stepwise logistic regression with Akaike information criterion, Random Forest, multivariate adaptive regression splines, support vector machine, naïve Bayes' classification, and recursive partitioning). The ASOCT-derived parameters were incorporated to generate point and interval estimates of the area under the receiver operating characteristic (AUC) curves for these algorithms using 10-fold cross-validation as well as 50:50 training and validation.

Main Outcome Measures

We assessed ASOCT measurements and angle closure.

Results

Data on 1368 subjects, including 295 (21.6%) subjects with gonioscopic angle closure were available for analysis. The mean (± standard deviation) age was 62.4±7.5 years and 54.8% were females. Angle closure subjects were older and had smaller ACW, ACA, and ACV; greater LV; and thicker irides (P<0.001 for all). For both, the 10-fold cross-validation and the 50:50 training and validation methods, stepwise logistic regression was the best algorithm for detecting eyes with gonioscopic angle closure with testing set AUC of 0.954 (95% confidence interval [CI], 0.942–0.966) and 0.962 (95% CI, 0.948–0.975) respectively, whereas recursive partitioning had relatively the poorest performance with testing set AUC 0.860 (95% CI, 0.790–0.930) and 0.905 (95% CI, 0.876–0.933), respectively. This algorithm performed similarly well (AUC, 0.957) in a second independent sample of 200 angle closure subjects and 302 normal controls.

Conclusions

A classification algorithm based on stepwise logistic regression that used a combination of 6 parameters obtained from a single horizontal ASOCT scan identified subjects with gonioscopic angle closure >95% of the time.

Financial Disclosure(s)

The authors have no proprietary or commercial interest in any of the materials discussed in this article.

Section snippets

Methods

This was an analysis of prospectively collected data from a community-based study performed in Singapore.15 Approval for the study was granted by the Singapore Eye Research Institute institutional review board. The study was conducted in accordance with the Declaration of Helsinki, and written informed consent was obtained from all subjects before enrolment into the study. The study population consisted of subjects aged ≥50 years recruited from a government-run, community-based clinic facility

Results

We recruited 2047 subjects into the study, of whom 679 were excluded for the following reasons: 11 could not undergo gonioscopy; 97 had incomplete demographic or biometric measurements; 62 could not complete the ASOCT examination or had poor quality ASOCT images; 42 had software delineation errors; and 467 had scleral spurs that were not clearly visible on ASOCT images. Therefore, data from 1368 subjects (66.8%) were included in the analysis. Of these, 750 (54.8%) were female, and 1232 (90.0%)

Discussion

Classification algorithms looking only at a single horizontal scan of the anterior segment using the ASOCT were able to identify nearly all cases of gonioscopic angle closure with close to 80% specificity and an AUC of 0.95. The full variable set consisting of 4 biometric, demographic, and ASOCT-derived parameters, as well as the reduced set, which solely relied on 6 ASOCT parameters, performed similarly across the different classification algorithms and training validation data. Furthermore,

References (29)

  • R. Sihota et al.

    Ocular parameters in subgroups of angle closure glaucoma

    Clin Experiment Ophthalmol

    (2000)
  • T. Aung et al.

    Anterior chamber depth and the risk of primary angle closure in 2 East Asian populations

    Arch Ophthalmol

    (2005)
  • W.P. Nolan et al.

    YAG laser iridotomy treatment for primary angle closure in East Asian eyes

    Br J Ophthalmol

    (2000)
  • P.J. Foster et al.

    Detection of gonioscopically occludable angles and primary angle closure glaucoma by estimation of limbal chamber depth in Asians: modified grading scheme

    Br J Ophthalmol

    (2000)
  • Cited by (70)

    • Anterior segment optical coherence tomography (AS-OCT) image analysis methods and applications: A systematic review

      2022, Computers in Biology and Medicine
      Citation Excerpt :

      In contrast to previous studies, Xu et al. [45]. proposed a new ML approach based on a SVM linear classifier [13,29]. In this study, two types of features were extracted: visual and clinical.

    • Evaluation of the Diagnostic Performance of Swept-Source Anterior Segment Optical Coherence Tomography in Primary Angle Closure Disease

      2022, American Journal of Ophthalmology
      Citation Excerpt :

      With the addition of minimal clinical information, SS-OCT was able to distinguish PAC/PACG from PACS with 82.0% sensitivity and 62.8% specificity, confirmed through an independent validation dataset. Anterior segment imaging has demonstrated general agreement with gonioscopy,17,18 and SS-OCT in particular has the advantage of being able to rapidly image all 360° of the anterior chamber, enabling extraction of much more data than traditional AS-OCT. In this study, we obtained data from 8 scans per eye including horizontal, vertical, and oblique scans as the 80° to 260° scan was previously reported to have the highest correlation with gonioscopy.19

    View all citing articles on Scopus

    Manuscript no. 2012-322.

    Financial Disclosure(s): The authors have no proprietary or commercial interest in any of the materials discussed in this article.

    Supported by grants from the Biomedical Research Council Translational Clinical Research (TCR) Partnership (TCR0101674) Singapore and the National Medical Research Council Clinician Scientist Award (NMRC/CSA/004/2008), Singapore.

    View full text