Skip to main content

  • Portable Brain-Computer Device for Detecting Glaucoma

    By Lynda Seminara and selected by Neil M. Bressler, MD, and Deputy Editors
    Add to My Bookmarks

    Journal Highlights

    JAMA Ophthalmology, June 2017

    Download PDF

    Nakanishi et al. conducted a case-control study to evaluate the utility of the nGoggle—a portable, wireless brain-computer interface device—in assessing loss of visual function in patients with glaucoma. The authors determined that the nGoggle enables discrimination of glaucomatous eyes from healthy eyes and has good test-retest reliability.

    The nGoggle is a head-mounted unit that records and processes multi-focal, steady-state, visual-evoked po­tentials (mfSSVEPs). In the study, 33 patients (62 eyes) with glaucoma and 17 healthy controls (30 eyes) were eval­uated by standard automated perimetry and the nGoggle. The capacity of these modalities to distinguish glaucomatous eyes from healthy eyes was compared in terms of global and sectoral parame­ters. Diagnostic accuracy was expressed as area under the receiver operating characteristic curve (AUC). An AUC of 1.0 denotes perfect discrimination, whereas an AUC of 0.5 indicates chance discrimination.

    Repeatability of nGoggle measurements was determined by testing 20 glaucomatous eyes (10 patients) in 3 weekly sessions and computing intra­class correlation coefficients (ICCs) and coefficients of variation. A canon­ical correlation analysis (CCA) was performed as a spatial filter to yield mfSSVEP-CCA parameters that were robust to artifacts.

    In nGoggle tests, the mean global mfSSVEP of eyes with glaucoma was lower than that of healthy eyes (0.289 vs. 0.334, respectively). The AUC for nGoggle global mfSSVEP was 0.92; this exceeded global parameters obtained by standard automated perimetry, includ­ing mean deviation (AUC, 0.81), mean sensitivity (AUC, 0.80), and pattern standard deviation (AUC, 0.77). The modalities did not differ significantly in accuracy of sectoral measurements. For the nGoggle, the mean coefficient of variation of the mfSSVEP-CCA global parameter was 3.03%, and the average ICC was 0.92. (An ICC of ≥ 0.75 was considered indicative of good repro­ducibility.)

    The authors concluded that this device can differentiate glaucomatous eyes from unaffected eyes. The ade­quate repeatability suggests that the device may be suitable for objective longitudinal assessment of glaucoma progression.

    The original article can be found here.