An Artificial Intelligence Enabled System for Retinal Nerve Fiber Layer Thickness Damage Severity Staging.

Purpose: To develop an objective glaucoma damage severity classification system based on OCT-derived retinal nerve fiber layer (RNFL) thickness measurements. Design: Algorithm development for RNFL damage severity classification based on multicenter OCT data. Subjects and Participants: A total of 6561 circumpapillary RNFL profiles from 2269 eyes of 1171 subjects to develop models, and 2505 RNFL profiles from 1099 eyes of 900 subjects to validate models. Methods: We developed an unsupervised k-means model to identify clusters of eyes with similar RNFL thickness profiles. We annotated the clusters based on their respective global RNFL thickness. We computed the optimal global RNFL thickness thresholds that discriminated different severity levels based on Bayes' minimum error principle. We validated the proposed pipeline based on an independent validation dataset with 2505 RNFL profiles from 1099 eyes of 900 subjects. Main Outcome Measures: Accuracy, area under the receiver operating characteristic curve, and confusion matrix. Results: The k-means clustering discovered 4 clusters with 1382, 1613, 1727, and 1839 samples with mean (standard deviation) global RNFL thickness of 58.3 (8.9) µm, 78.9 (6.7) µm, 87.7 (8.2) µm, and 101.5 (7.9) µm. The Bayes' minimum error classifier identified optimal global RNFL values of > 95 µm, 86 to 95 µm, 70 to 85 µm, and < 70 µm for discriminating normal eyes and eyes at the early, moderate, and advanced stages of RNFL thickness loss, respectively. About 4% of normal eyes and 98% of eyes with advanced RNFL loss had either global, or ≥ 1 quadrant, RNFL thickness outside of normal limits provided by the OCT instrument. Conclusions: Unsupervised machine learning discovered that the optimal RNFL thresholds for separating normal eyes and eyes with early, moderate, and advanced RNFL loss were 95 µm, 85 µm, and 70 µm, respectively. This RNFL loss classification system is unbiased as there was no preassumption or human expert intervention in the development process. Additionally, it is objective, easy to use, and consistent, which may augment glaucoma research and day-to-day clinical practice. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Detecting glaucoma from multi-modal data using probabilistic deep learning.

Objective: To assess the accuracy of probabilistic deep learning models to discriminate normal eyes and eyes with glaucoma from fundus photographs and visual fields. Design: Algorithm development for discriminating normal and glaucoma eyes using data from multicenter, cross-sectional, case-control study. Subjects and participants: Fundus photograph and visual field data from 1,655 eyes of 929 normal and glaucoma subjects to develop and test deep learning models and an independent group of 196 eyes of 98 normal and glaucoma patients to validate deep learning models. Main outcome measures: Accuracy and area under the receiver-operating characteristic curve (AUC). Methods: Fundus photographs and OCT images were carefully examined by clinicians to identify glaucomatous optic neuropathy (GON). When GON was detected by the reader, the finding was further evaluated by another clinician. Three probabilistic deep convolutional neural network (CNN) models were developed using 1,655 fundus photographs, 1,655 visual fields, and 1,655 pairs of fundus photographs and visual fields collected from Compass instruments. Deep learning models were trained and tested using 80% of fundus photographs and visual fields for training set and 20% of the data for testing set. Models were further validated using an independent validation dataset. The performance of the probabilistic deep learning model was compared with that of the corresponding deterministic CNN model. Results: The AUC of the deep learning model in detecting glaucoma from fundus photographs, visual fields, and combined modalities using development dataset were 0.90 (95% confidence interval: 0.89-0.92), 0.89 (0.88-0.91), and 0.94 (0.92-0.96), respectively. The AUC of the deep learning model in detecting glaucoma from fundus photographs, visual fields, and both modalities using the independent validation dataset were 0.94 (0.92-0.95), 0.98 (0.98-0.99), and 0.98 (0.98-0.99), respectively. The AUC of the deep learning model in detecting glaucoma from fundus photographs, visual fields, and both modalities using an early glaucoma subset were 0.90 (0.88,0.91), 0.74 (0.73,0.75), 0.91 (0.89,0.93), respectively. Eyes that were misclassified had significantly higher uncertainty in likelihood of diagnosis compared to eyes that were classified correctly. The uncertainty level of the correctly classified eyes is much lower in the combined model compared to the model based on visual fields only. The AUCs of the deterministic CNN model using fundus images, visual field, and combined modalities based on the development dataset were 0.87 (0.85,0.90), 0.88 (0.84,0.91), and 0.91 (0.89,0.94), and the AUCs based on the independent validation dataset were 0.91 (0.89,0.93), 0.97 (0.95,0.99), and 0.97 (0.96,0.99), respectively, while the AUCs based on an early glaucoma subset were 0.88 (0.86,0.91), 0.75 (0.73,0.77), and 0.92 (0.89,0.95), respectively. Conclusion and relevance: Probabilistic deep learning models can detect glaucoma from multi-modal data with high accuracy. Our findings suggest that models based on combined visual field and fundus photograph modalities detects glaucoma with higher accuracy. While probabilistic and deterministic CNN models provided similar performance, probabilistic models generate certainty level of the outcome thus providing another level of confidence in decision making.

It Is All about Pressure.

Glaucoma is an ocular disease caused by elevated intraocular pressure that leads to progressive optic neuropathy [...].

Canaloplasty in Pseudoexfoliation Glaucoma. Can It Still Be Considered a Good Choice?

Purpose: The aim of this study was to assess the long-term outcomes of canaloplasty surgery in pseudoexfoliation glaucoma (PEXG) patients. Material and Methods: A total of 116 PEXG patients with an intraocular pressure (IOP) > 21 mm/Hg and maximum tolerated local medical therapy who underwent canaloplasty from February 2008 to January 2022 were considered. Every six months, all subjects underwent a complete ophthalmic examination. The period of follow-up ranged from 2 to 167 months. Inclusion criteria included only patients for whom the entire procedure could be completed with a follow-up of at least 2 years. Results: Amongst the 116 PEXG patients, the entire procedure could not be performed in 10 eyes (8.6%), and thus they were not considered in the analysis. Twenty-three patients did not reach the two-year follow-up and another 16 patients during this time period were lost. A total of 67 patients with a mean follow-up of 49 ± 32.3 months were considered in the analysis. The pre-operative mean IOP was 31.2 ± 8.7 mm/Hg (range 20−60). The mean IOP at the two-year follow-up was 17.2 ± 6.7 mmHg, with a mean reduction from baseline of 44.9%. After two years, the qualified success rates according to three different criteria (IOP ≤ 21, ≤18 and ≤16 mmHg) were 80.6%, 73.1% and 61.0%, respectively. The total number of medications used pre- and at the follow-up at 2 years was 3.5 ± 0.8 and 1.2 ± 1.4, respectively. Early complications included: hyphema, in about 30% of cases; Descemet membrane detachment (4.9%); and IOP spikes > 10 mmHg (9.7%). A late failure with an acute IOP rise of up to 50 mmHg was observed in 41 cases (61.2%) after 3 to 72 months. Conclusions: Long-term post-operative outcomes of canaloplasty in PEXG patients appear to be quite good on average; however, an acute rise in IOP can be observed in more than 60% of the cases after a long period of satisfactory IOP control. For this reason, canaloplasty may not be suitable in eyes with PEXG, especially in patients with severe functional damage.

360° Ab-Interno Schlemm's Canal Viscodilation with OMNI Viscosurgical Systems for Open-Angle Glaucoma-Midterm Results.

PURPOSE: To evaluate the effectiveness of ab-interno microcatheterization and 360° viscodilation of Schlemm's canal (SC) performed with OMNI viscosurgical system in open angle glaucoma (OAG) together or not with phacoemulsification. SETTING: Two surgical sites. DESIGN: Retrospective, observational. METHODS: Eighty eyes from 73 patients with mild to moderate OAG underwent ab- interno SC viscodilation performed with OMNI system. Fifty eyes (Group 1) underwent only SC viscodilation, while 30 eyes (Group 2) underwent glaucoma surgery + cataract extraction. Primary success endpoint at 12 months was an intraocular pressure (IOP) reduction higher than 25% from baseline with an absolute value of 18 mmHg or lower, either on the same number or fewer ocular hypotensive medications, without further interventions. Secondary effectiveness endpoints included mean IOP, number of medications and comparison of outcomes between groups. Safety endpoints consisted of best-corrected visual acuity (BCVA), adverse events (AEs), and subsequent surgical procedures. RESULTS: Primary success was achieved in 40.0% and 67.9% in Groups 1 and 2, respectively. Mean IOP at 12-month follow-up showed a significant reduction in both groups (from 23.0 to 15.6 mmHg, p < 0.001, and from 21.5 to 14.1, p < 0.001, in Groups 1 and 2, respectively). Mean medication number decreased in both groups (from 3.0 to 2.0, p < 0.001 and from 3.4 to 1.9, p < 0.001, in Groups 1 and 2, respectively). AEs included hyphema (2 eyes), mild hypotony (4 eyes), IOP spikes one month after surgery (1 eye). Twelve eyes (15.0%) required subsequent surgical procedures. No BCVA reduction was observed. CONCLUSIONS: Viscodilation of SC using OMNI viscosurgical systems is safe and relatively effective in reducing IOP in adult patients with OAG.

Global Glaucoma Staging System (GGSS): A New Method to Simultaneously Assess the Severity of Both Functional and Structural Damage in Glaucoma.

BACKGROUND: The classification of damage in glaucoma is usually based either on visual field or optical coherent tomography (OCT) assessment. No currently available method is able to simultaneously categorize functional and structural damage. MATERIAL AND METHODS: In this study, 283 patients with chronic open-angle glaucoma (OAG) at different stages and 67 healthy subjects were tested with both standard automated perimetry and spectral domain OCT for retinal nerve fiber layer (RNFL) assessment. The visual field data were classified using the Glaucoma Staging System 2, whereas OCT results were processed with the OCT Glaucoma Staging System. These data were used to create a new staging system (global glaucoma staging system, GGSS), in which the severity of visual field and RNFL damage is reported on the Y and X axis, respectively. The GGSS was tested in a different sample of 147 patients with manifest OAG, 56 with preperimetric glaucoma and 43 normal subjects. A six-stage clinical classification, based on the analysis of visual fields and optic disc appearance, was used as gold standard. RESULTS: The GGSS was able to correctly classify in the same stage or within the immediately adjacent stages 145 cases on 147 (98.6%). Fifty-four preperimetric cases (96.4%) were classified as borderline or abnormal (Stage 1 or 2). Here, 41 normal eyes out of 43 were correctly classified as Stage 0, with a specificity of 95.3%. CONCLUSIONS: Preliminary results from this study are encouraging and suggest that the new GGSS is able to provide information concerning the severity of structural and functional damage in an integrated manner.

How to Measure Intraocular Pressure: An Updated Review of Various Tonometers.

Intraocular pressure (IOP) is an important measurement that needs to be taken during ophthalmic examinations, especially in ocular hypertension subjects, glaucoma patients and in patients with risk factors for developing glaucoma. The gold standard technique in measuring IOP is still Goldmann applanation tonometry (GAT); however, this procedure requires local anesthetics, can be difficult in patients with scarce compliance, surgical patients and children, and is influenced by several corneal parameters. Numerous tonometers have been proposed in the past to address the problems related to GAT. The authors review the various devices currently in use for the measurement of intraocular pressure (IOP), highlighting the main advantages and limits of the various tools. The continuous monitoring of IOP, which is still under evaluation, will be an important step for a more complete and reliable management of patients affected by glaucoma.

Improving event-based progression analysis in glaucomatous visual fields.

Glaucoma is a progressive optic neuropathy with characteristic changes to the optic nerve head and the visual field (VF). Detecting progression of VF damage with Standard Automated Perimetry (SAP) is of paramount importance for clinical care. One common approach to detecting progression is to compare each new VF test to a baseline SAP test (event analysis). This comparison is made difficult by the test-retest variability of SAP, which increases with the level of VF damage, and the limited range of measurement, meaning that damage cannot be assessed below a certain level. We performed a prospective international multi-centre data collection of SAP data on 90 eyes from 90 people with glaucoma and different levels of VF damage over a short period of time (6 tests in 60 days). Data were collected using a fundus tracked perimeter (Compass, CenterVue). We used these data (minus the first test) to develop an improved event analysis that accounts for both the change in variability with damage and the lower bound on the measurement imposed by SAP. Using simulations, we show that our approach is more sensitive compared to previously developed methods, especially in the case of advanced glaucoma, while retaining similar specificity.

Do Additional Testing Locations Improve the Detection of Macular Perimetric Defects in Glaucoma?

PURPOSE: To evaluate the ability of additional central testing locations to improve detection of macular visual field (VF) defects in glaucoma. DESIGN: Prospective cross-sectional study. PARTICIPANTS: Four hundred forty healthy people and 499 patients with glaucomatous optic neuropathy (GON) were tested with a fundus tracked perimeter (CMP; CenterVue) using a 24-2 grid with 12 additional macular locations (24-2+). METHODS: Glaucomatous optic neuropathy was identified based on expert evaluation of optic nerve head photographs and OCT scans, independently of the VF. We defined macular defects as locations with measurements outside the 5% and 2% normative limits on total deviation (TD) and pattern deviation (PD) maps within the VF central 10°. Classification was based on the total number of affected macular locations (overall detection) or the largest number of affected macular locations connected in a contiguous cluster (cluster detection). Criteria based on the number of locations and cluster size were used to obtain equivalent specificity between the 24-2 grid and the 24-2+ grids, calculated using false detections in the healthy cohort. Partial areas under the receiver operating characteristic curve (pAUCs) were also compared at specificities of 95% or more. MAIN OUTCOME MEASURES: Matched specificity comparison of the ability to detect glaucomatous macular defects between the 24-2 and 24-2+ grids. RESULTS: At matched specificity, cluster detection identified more macular defects with the 24-2+ grid compared with the 24-2 grid. For example, the mean increase in percentage of detection was 8% (95% confidence interval [CI], 5%-11%) and 10% (95% CI, 7%-13%) for 5% TD and PD maps, respectively, and 5% (95% CI, 2%-7%) and 6% (95% CI, 4%-8%) for the 2% TD and PD maps, respectively. Good agreement was found between the 2 grids. The improvement measured by pAUCs was also significant but generally small. The percentage of eyes with macular defects ranged from about 30% to 50%. Test time for the 24-2+ grid was longer (21% increase) for both cohorts. Between 74% and 98% of defects missed by the 24-2 grid had at least 1 location with sensitivity of < 20 dB. CONCLUSIONS: Visual field examinations with additional macular locations can improve the detection of macular defects in GON modestly without loss of specificity when appropriate criteria are selected.

Canaloplasty in Pigmentary Glaucoma: Long-Term Outcomes and Proposal of a New Hypothesis on Its Intraocular Pressure Lowering Mechanism.

This study presents the long-term results on canaloplasty in a group of patients affected by pigmentary glaucoma, and studies the progression of the disease after surgery. MATERIAL AND METHODS: Twenty-nine eyes of 25 patients with pigmentary glaucoma in maximum tolerated medical therapy with significant visual field damage progression underwent canaloplasty and were followed up to 11 years (mean 59.8 ± 30.1 months). All patients underwent a complete ophthalmic examination every 6 months. RESULTS: The pre-operative mean intraocular pressure (IOP) was 31.8 mmHg ± 10.9 (range 21-70) with an average of 3.3 medications. After 1, 2, 3, and 4 years, the mean IOP was 15.9 ± 4.0, 14.4 ± 7.3, 14.1 ± 2.1, and 15.7 mmHg, respectively, with 0.4, 0.5, and 0.7 medications, respectively. Four patients underwent trabeculectomy after 3 to 30 months due to uncontrolled IOP. Gonioscopy showed a significant reduction of pigment in trabecular meshwork in all cases, starting from the sixth month. In some cases, the pigment was almost completely reabsorbed after two years, suggesting an accelerated transit and escape of the granules through the trabecular spaces. CONCLUSIONS: Canaloplasty seems to be a reasonable option in treating patients affected by progressive pigmentary glaucoma. The reabsorption of pigment granules from the trabecular meshwork could, at least in part, explain the relatively high success rate observed after this surgical procedure.

A New System for Assessing Visual Disability Using a Digital Visor.

BACKGROUND: Considering the lack of universally accepted visual requirements for driving and for defining various grades of visual disability, the aim of this study is to propose a new method that provides a numerical score resulting from a combined assessment of the visual field and visual acuity loss obtained using a digital technology visor. METHODS: This study presents a new system for calculating the percentage of visual disability by combining binocular visual acuity and binocular visual field assessments. A new Global Vision Evaluation System digital technology visor uses standardized, reproducible criteria to produce well-defined, numerically expressed test results. Through a specific algorithm, the device produces a numerical value expressing the percentage of visual disability. RESULTS: Eighty-six subjects with various types of visual impairment underwent visual acuity and visual field test examinations carried out employing both traditional methods and the new digital visor. The two methods provided homogeneously similar results regarding the positioning of the subjects on the visual disability scale. CONCLUSIONS: The new digital visor seems to be a valid method to ensure that visual disability assessments are more homogeneous and reliable, and that, consequently, the resources available for this purpose are more fairly distributed.

Effect of fundus tracking on structure-function relationship in glaucoma.

AIMS: To investigate the effect of fundus tracking perimetry on structure-function relationship in glaucoma. METHODS: Perimetric data were acquired with the Humphrey Field Analyzer (HFA) and the Compass fundus perimeter (CMP, equipped with fundus tracking). We included data from 696 eyes from 360 healthy people and 711 eyes from 434 patients with glaucoma from the original study, for which the circumpapillary retinal nerve fibre layer optical coherence tomography scan (Cp-RNFL) was available. We explored the structure-function relationship using both global indices (mean deviation and average Cp-RNFL thickness loss) and anatomically defined visual field clusters comparing the R2 values from mixed-effect models. We then measured the diagnostic ability of a combined Structure Function Index (SFI) using perimetric data from either perimeter. The comparisons were based on partial receiver operating characteristic curves with a minimum specificity of 75% and their areas under the curve. RESULTS: The R2 for the global structure-function relationship was 0.50 for the CMP and 0.48 for the HFA. When visual field clusters were included in the model, the R2 was 0.29 for CMP and 0.30 for HFA. Overall, the discrimination ability of the SFI was not significantly higher than the Cp-RNFL for either the CMP (p=0.07) or the HFA (p=0.14). However, it was significantly better in eyes with perimetric damage (p<0.001), in which the CMP-SFI performed significantly better than the HFA-SFI (p=0.03). CONCLUSIONS: Structure-function relationship is similar between the two perimeters. Fundus tracking might improve discrimination ability with a combined SFI. TRIAL REGISTRATION NUMBER: ISRCTN13800424.

A Comparison between the Compass Fundus Perimeter and the Humphrey Field Analyzer.

PURPOSE: To evaluate relative diagnostic precision and test-retest variability of 2 devices, the Compass (CMP, CenterVue, Padova, Italy) fundus perimeter and the Humphrey Field Analyzer (HFA, Zeiss, Dublin, CA), in detecting glaucomatous optic neuropathy (GON). DESIGN: Multicenter, cross-sectional, case-control study. PARTICIPANTS: We sequentially enrolled 499 patients with glaucoma and 444 normal subjects to analyze relative precision. A separate group of 44 patients with glaucoma and 54 normal subjects was analyzed to assess test-retest variability. METHODS: One eye of recruited subjects was tested with the index tests: HFA (Swedish interactive thresholding algorithm [SITA] standard strategy) and CMP (Zippy Estimation by Sequential Testing [ZEST] strategy), 24-2 grid. The reference test for GON was specialist evaluation of fundus photographs or OCT, independent of the visual field (VF). For both devices, linear regression was used to calculate the sensitivity decrease with age in the normal group to compute pointwise total deviation (TD) values and mean deviation (MD). We derived 5% and 1% pointwise normative limits. The MD and the total number of TD values below 5% (TD 5%) or 1% (TD 1%) limits per field were used as classifiers. MAIN OUTCOME MEASURES: We used partial receiver operating characteristic (pROC) curves and partial area under the curve (pAUC) to compare the diagnostic precision of the devices. Pointwise mean absolute deviation and Bland-Altman plots for the mean sensitivity (MS) were computed to assess test-retest variability. RESULTS: Retinal sensitivity was generally lower with CMP, with an average mean difference of 1.85±0.06 decibels (dB) (mean ± standard error, P < 0.001) in healthy subjects and 1.46±0.05 dB (mean ± standard error, P < 0.001) in patients with glaucoma. Both devices showed similar discriminative power. The MD metric had marginally better discrimination with CMP (pAUC difference ± standard error, 0.019±0.009, P = 0.035). The 95% limits of agreement for the MS were reduced by 13% in CMP compared with HFA in participants with glaucoma and by 49% in normal participants. Mean absolute deviation was similar, with no significant differences. CONCLUSIONS: Relative diagnostic precision of the 2 devices is equivalent. Test-retest variability of MS for CMP was better than for HFA.

Canaloplasty in the Treatment of Open-Angle Glaucoma: A Review of Patient Selection and Outcomes.

Canaloplasty is a relatively new non-penetrating surgery for the reduction of intraocular pressure in patients affected by glaucoma. The technique uses a microcatheter to perform a 360 º cannulation of Schlemm's canal and leaves in place a tension suture providing an inward distension. It aims to restore the physiological outflow pathways of the aqueous humour and is independent of external wound healing. Several studies have shown that canaloplasty is effective in reducing intraocular pressure and has a low rate of complications, especially compared with trabeculectomy, the gold standard for glaucoma surgery. Currently, canaloplasty is indicated in patients with open-angle glaucoma, having a mild to moderate disease, and the combination with cataract phacoemulsification may provide further intraocular pressure reduction. This article reviews canaloplasty indications, results and complications and analyses its outcomes compared with traditional penetrating and non-penetrating techniques.

Canaloplasty in Corticosteroid-Induced Glaucoma. Preliminary Results.

PURPOSE: to present the mid-term results of canaloplasty in a small cohort of corticosteroid glaucoma patients. MATERIAL AND METHODS: Nine eyes from seven patients with various types of corticosteroid glaucoma in maximum medical therapy underwent canaloplasty. Patients underwent complete ophthalmic examination every six months. Success was defined as: post-operative intraocular pressure (IOP) ≤ 21 mmHg and ≤ 16 mmHg without ("complete success"), and with/without medical treatment ("qualified success"). The IOP reduction had to be ≥ 20. The number of medications before and after surgery was considered. The follow-up mean period was 32.7 ± 20.8 months (range 14-72 months). RESULTS: The pre-operative mean IOP was 30.7 ± 7.2 mmHg (range: 24-45). The mean IOP at 6 and 12-month follow-up was 13.1 ± 2.6 mmHg, and 13.7 ± 1.9 mmHg, respectively. Qualified and complete success at 6 and 12 months was 100% for both of the two definitions. The number of medications used preoperatively and at the 12-month follow-up was 4.3 ± 0.7, and 0.2 ± 1.0, respectively. No serious complication was observed. CONCLUSIONS: The mid-term results of canaloplasty in patients with corticosteroid-induced glaucoma appear to be very promising. Canaloplasty should be considered as a possible alternative to filtering surgery in this form of glaucoma, when medical therapy is not sufficient to maintain the IOP within reasonable limits.

Adipose Derived Stem Cells for Corneal Wound Healing after Laser Induced Corneal Lesions in Mice.

The aim of our study was to assess the clinical effectiveness of topical adipose derived stem cell (ADSC) treatment in laser induced corneal wounds in mice by comparing epithelial repair, inflammation, and histological analysis between treatment arms. Corneal lesions were performed on both eyes of 40 mice by laser induced photorefractive keratectomy. All eyes were treated with topical azythromycin bid for three days. Mice were divided in three treatment groups (n = 20), which included: control, stem cells and basic serum; which received topical treatment three times daily for five consecutive days. Biomicroscope assessments and digital imaging were performed by two masked graders at 30, 54, 78, 100, and 172 h to analyze extent of fluorescein positive epithelial defect, corneal inflammation, etc. Immunohistochemical techniques were used in fixed eyes to assess corneal repair markers Ki67, α Smooth Muscle Actin (α-SMA) and E-Cadherin. The fluorescein positive corneal lesion areas were significantly smaller in the stem cells group on days 1 (p < 0.05), 2 (p < 0.02) and 3. The stem cell treated group had slightly better and faster re-epithelization than the serum treated group in the initial phases. Comparative histological data showed signs of earlier and better corneal repair in epithelium and stromal layers in stem cell treated eyes, which showed more epithelial layers and enhanced wound healing performance of Ki67, E-Cadherin, and α-SMA. Our study shows the potential clinical and histological advantages in the topical ADSC treatment for corneal lesions in mice.