Artificial intelligence for glaucoma: state of the art and future perspectives.

PURPOSE OF REVIEW: To address the current role of artificial intelligence (AI) in the field of glaucoma. RECENT FINDINGS: Current deep learning (DL) models concerning glaucoma diagnosis have shown consistently improving diagnostic capabilities, primarily based on color fundus photography and optical coherence tomography, but also with multimodal strategies. Recent models have also suggested that AI may be helpful in detecting and estimating visual field progression from different input data. Moreover, with the emergence of newer DL architectures and synthetic data, challenges such as model generalizability and explainability have begun to be tackled. SUMMARY: While some challenges remain before AI is routinely employed in clinical practice, new research has expanded the range in which it can be used in the context of glaucoma management and underlined the relevance of this research avenue.

Keratoconus tomographic indices in osteogenesis imperfecta.

PURPOSE: Osteogenesis imperfecta (OI) is a rare inherited disease affecting collagen-rich tissues. Ocular complications have been reported such as thin corneas, low ocular rigidity, keratoconus, among others. The purpose of this study is to characterize corneal tomographic features in OI patients compared to unaffected patients, with particular focus on commonly studied keratoconus indices. METHODS: Cross-sectional case-control study including 37 OI patients and 37 age-matched controls. Patients and controls underwent comprehensive ophthalmological examination including corneal Scheimpflug tomography with a Pentacam HR device (Oculus Optikgeräte GmbH, Wetzlar, Germany) to analyse and compare topometric, tomographic, pachymetric and Belin-Ambrósio Enhanced Ectasia Display III (BAD-D) data of both eyes of each patient. RESULTS: Most OI patients had type I disease (n = 24; 65%) but type III-VII patients were also included. Two patients had clinically overt bilateral keratoconus. OI patients had significantly higher maximum keratometry (45.2 ± 2.1 vs. 43.7 ± 1.2; p = 0.0416), front and back elevation (3.0 ± 3.3 vs. 2.1 ± 1.3, p = 0.0201; 11.1 ± 8.2 vs. 5.0 ± 3.7, p < 0.0001), index of surface variance (25.5 ± 13 vs. 17.4 ± 8.3; p = 0.0016), index of vertical asymmetry (0.21 ± 0.14 vs. 0.15 ± 0.06; p = 0.0215), index of height asymmetry (9.2 ± 14 vs. 6.0 ± 4.5; p = 0.0421), index of height decentration (0.02 ± 0.01 vs. 0.01 ± 0.01; p < 0.0001) and average pachymetric progression (1.01 ± 0.19 vs. 0.88 ± 0.14; p < 0.0001) readings. Thinnest corneal thickness and maximum Ambrósio relational thickness were significantly lower (477 ± 52 vs. 543 ± 26; 387 ± 95 vs. 509 ± 49; p < 0.0001). Two-thirds of OI patients had corneas with a minimum thickness < 500 µm. BAD-D value was significantly higher in OI patients (2.1 ± 1.4 vs. 0.9 ± 0.2; p < 0.0001). CONCLUSION: OI patients showed significant changes in corneal profiles compared with healthy subjects. A high proportion of patients had tomographically suspect corneas when using keratoconus diagnostic indices. Further studies are warranted to assess the true risk of corneal ectasia in OI patients.

Bleb morphology of glaucoma drainage devices on magnetic resonance imaging.

PURPOSE: To evaluate bleb morphology features of different glaucoma drainage devices (GDD) using magnetic resonance imaging (MRI). MATERIALS AND METHODS: Prospective cohort study of GDD and bleb morphology in consecutive glaucoma patients implanted with Ahmed Glaucoma Valve (AGV), Paul Glaucoma Implant (PGI), Baerveldt Glaucoma Implant (BGI) and Ahmed ClearPath (ACP) devices. Thirty-six eyes from 30 consecutive patients underwent standardized GDD implantation followed by MRI at least 1 month after surgery. Main outcomes included bleb volume and endplate position relative to the optic nerve. Secondary outcomes included intraocular pressure (IOP), medication and surgical complications during a 12-month follow-up. RESULTS: Seven eyes were implanted with the AGV (19%), 5 with BGI (14%), 16 with PGI (44%) and 8 with ACP (22%). MRI scans were obtained 85 ± 66 days after surgery. Mean total bleb volume was 563 ± 390 mm3 . This bleb volume was inversely correlated with early post-operative IOP (day 7; rs = -0.3326, p = 0.0475) but positively correlated with IOP at 12 months (rs = 0.3592, p = 0.0341). No significant difference in total bleb volume was found between GDD types (p = 0.1223). A double-layered bleb was observed in 34 eyes (94%). The inferior bleb volume was significantly larger in PGI devices versus other GDD types (380 ± 205 vs. 193 ± 161 mm3 ; p = 0.0043). Distance from the endplate to the optic nerve was 9.5 ± 4.0 mm, similar across GDDs (p = 0.2519). CONCLUSIONS: Double-layered blebs are a common finding with GDDs. Bleb volume showed different correlations with IOP at distinct timepoints and the PGI device formed larger blebs. A standardized GDD implantation technique ensures a safe distance from the GDD endplate to the optic nerve.

Agreement Analysis on Angle Characteristics With Automated Gonioscopy.

PRCIS: Automated gonioscopy is a recent method of recording angle pictures. Our study shows that agreement between observers is moderate in most categories used in clinical practice, underlying its clinical usefulness but also room for improvement. INTRODUCTION: Iridocorneal angle (ICA) imaging methods have been recently developed to record gonioscopic images. The purpose of this study was to perform an interrater agreement analysis of ICA photographs. METHODS: Multicentric cross-sectional observational study. Consecutive patients in 2 ophthalmology departments underwent automated gonioscopy. One picture per quadrant from each eye was selected for randomization. Pictures were graded and analyzed by 4 masked glaucoma specialists regarding angle opening, width, Shaffer and Spaeth gradings and other findings. Fleiss' κ statistics was performed to assess interrater agreement. RESULTS: A total of 50 patients were recruited, with the sample containing a variety of diagnosis including pseudoexfoliation (22%), angle-closure suspect (12%), and pigmentary glaucoma (4%). The sample included phakic (68%) and pseudophakic patients, as well as cases with gonio-implanted surgical devices (10%). There was moderate agreement on angle opening, angle width, presence of angle vessels, and peripheral anterior synechiae (κ=0.435, 0.511, 0.558, 0.488, respectively; P<0.01). Fair agreement was observed regarding Shaffer grading, site of iris root insertion, angle pigmentation features, and the presence of iris processes. Expectedly from a 2-dimensional picture, the configuration of peripheral iris was found to have a poor agreement. Subset analysis on higher quality pictures seemed to improve agreement on pigment but did not further improve on the overall good agreement on angle opening status. CONCLUSIONS: Our study shows that automated gonioscopy provides moderate agreement on most clinically relevant features. Binary characteristics such as angle opening or PAS seem to be more robust than more complex angle classification parameters.