In vivo identification of angle dysgenesis and its relation to genetic markers associated with glaucoma using artificial intelligence

PURPOSE

To predict the presence of angle dysgenesis on anterior-segment optical coherence tomography (ADoA) by using deep learning (DL) and to correlate ADoA with mutations in known glaucoma genes.

PARTICIPANTS

In total, 800 high-definition anterior-segment optical coherence tomography (AS-OCT) images were included, of which 340 images were used to build the machine learning (ML) model. Images used to build the ML model included 170 scans of primary congenital glaucoma (16 patients), juvenile-onset open-angle glaucoma (62 patients), and adult-onset primary open-angle glaucoma eyes (37 patients); the rest were controls (n = 85). The genetic validation dataset consisted of another 393 images of patients with known mutations that were compared with 320 images of healthy controls.

METHODS

ADoA was defined as the absence of Schlemm's canal, the presence of hyperreflectivity over the region of the trabecular meshwork, or a hyperreflective membrane. DL was used to classify a given AS-OCT image as either having angle dysgenesis or not. ADoA was then specifically looked for on AS-OCT images of patients with mutations in the known genes for glaucoma.

RESULTS

The final prediction, which was a consensus-based outcome from the three optimized DL models, had an accuracy of >95%, a specificity of >97%, and a sensitivity of >96% in detecting ADoA in the internal test dataset. Among the patients with known gene mutations, ( MYOC, CYP1B1, FOXC1, and LTBP2 ) ADoA was observed among all the patients in the majority of the images, compared to only 5% of the healthy controls.

CONCLUSION

ADoA can be objectively identified using models built with DL.

In Vivo Imaging of the Schlemm's Canal and the Response to Selective Laser Trabeculoplasty

OBJECTIVE

To evaluate the presence of angle dysgenesis on ASOCT (anterior segment optical coherence tomography) (ADoA) as a predictive factor in determining outcomes of selective laser trabeculoplasty (SLT).

DESIGN

A prospective clinical cohort study.

SUBJECTS

Patients with juvenile-onset open-angle glaucoma (JOAG) without angle dysgenesis on gonioscopy.

METHOD

JOAG patients with uncontrolled intraocular pressure (IOP), who were to undergo SLT, were evaluated for the presence or absence of ADoA, which was defined as the absence of Schlemm's canal (SC) and/or presence of a hyperreflective membrane (HM) over the trabecular meshwork, as identified on ASOCT before the SLT procedure. Furthermore, the number of ASOCT B-scans in which SC was identified as present, were then quantified. Success of SLT was defined as a reduction of IOP by 20% or more from pre-laser value at 6-month follow-up without any further IOP-lowering medication or surgery. Only 1 repeat SLT was admissible for defining SLT success over the 6-month period. A successful reduction in IOP at 6-month follow-up was correlated with the extent of ADoA.

RESULTS

In comparison to pre-SLT IOP, 57.1% eyes (20/35) showed more than 20% reduction in IOP at 6 months with a mean reduction of 7.6 ± 1.8 mm Hg (29.6%). When all 3 observers agreed, SC was identified in 90% eyes (18/20) with success vs 26.6% eyes (4/15) with failure (P < .001). All eyes (5/5) with presence of HM showed failure (P < .001). All eyes (19/19) in which SC was present in >50% ASOCT B scans (>25/50 scans/eye) showed success (P < .001). On a bias-reduced regression analysis, the identification of SC on any 2 consecutive scans increased the chances of success at 6 months by 8.3 times, whereas the identification of SC in >50% of ASOCT scans was associated with a 21.4 times greater chance of success.

CONCLUSIONS

The presence of SC on ASOCT is a strong predictor for successful IOP reduction after SLT in JOAG eyes.