Faculty Bibliography

BACKGROUND:This review explores the bioethical implementation of artificial intelligence (AI) in medicine and in ophthalmology. AI, which was first introduced in the 1950s, is defined as "the machine simulation of human mental reasoning, decision making, and behavior". The increased power of computing, expansion of storage capacity, and compilation of medical big data helped the AI implementation surge in medical practice and research. Ophthalmology is a leading medical specialty in applying AI in screening, diagnosis, and treatment. The first Food and Drug Administration approved autonomous diagnostic system served to diagnose and classify diabetic retinopathy. Other ophthalmic conditions such as age-related macular degeneration, glaucoma, retinopathy of prematurity, and congenital cataract, among others, implemented AI too. PURPOSE/OBJECTIVE:To review the contemporary literature of the bioethical issues of AI in medicine and ophthalmology, classify ethical issues in medical AI, and suggest possible standardizations of ethical frameworks for AI implementation. METHODS:Keywords were searched on Google Scholar and PubMed between October 2019 and April 2020. The results were reviewed, cross-referenced, and summarized. A total of 284 references including articles, books, book chapters, and regulatory reports and statements were reviewed, and those that were relevant were cited in the paper. RESULTS:Most sources that studied the use of AI in medicine explored the ethical aspects. Bioethical challenges of AI implementation in medicine were categorized into 6 main categories. These include machine training ethics, machine accuracy ethics, patient-related ethics, physician-related ethics, shared ethics, and roles of regulators. CONCLUSIONS:There are multiple stakeholders in the ethical issues surrounding AI in medicine and ophthalmology. Attention to the various aspects of ethics related to AI is important especially with the expanding use of AI. Solutions of ethical problems are envisioned to be multifactorial.

BACKGROUND:Venezuela is in the throes of a complex humanitarian crisis that is one of the worst in decades to impact any country outside of wartime. This case analysis describes the challenges faced by the ongoing Maracaibo Aging Study (MAS) during the deteriorating conditions in Venezuela. When the MAS began in 1997, it focused on memory-related disorders. Since then, strategic planning and proactive community participation allowed us to anticipate and address logistical, funding, and ethical challenges, and facilitated the enrollment and retention of more than 2500 subjects over 55 years of age. All participants, who are residents of the city of Maracaibo, Venezuela, underwent various assessments on several occasions. Here, we discuss how our approach to implementing a longitudinal, population-based study of age-related conditions has allowed our research program to continue throughout this period of political, economic, and social upheaval. DISCUSSION/CONCLUSIONS:As the social context in Venezuela became more complicated, new challenges emerged, and strategies to sustain the study and participation were refined. We identified five main mechanisms through which the evolving humanitarian crisis has affected implementation of the MAS: 1) community dynamics; 2) morale of researchers, staff, and participants; 3) financial feasibility; 4) components of the research process; and 5) impact on the health of staff, participants, and their families. Strategies to compensate for the impact on these components were implemented, based on inputs from community members and staff. Improved communication, greater involvement of stakeholders, broadening the scope of the project, and strengthening international collaboration have been the most useful strategies. Particular demands emerged, related to the increased mortality and comorbidities of participants and staff, and deterioration of basic services and safety. CONCLUSION/CONCLUSIONS:Although the MAS has faced numerous obstacles, it has been possible to continue a longitudinal research project throughout the humanitarian crisis, because our research team has engaged the community deeply and developed a sense of mutual commitment, and also because our project has provided funding to help keep researchers employed, somewhat attenuating the brain drain.

Glaucoma is a chronic neurodegenerative disease of the optic nerve and a leading cause of irreversible blindness, worldwide. While the experimental research using animal models provides growing information about cellular and molecular processes, parallel analysis of the clinical presentation of glaucoma accelerates the translational progress towards improved understanding, treatment, and clinical testing of glaucoma. Optic nerve axon injury triggers early alterations of retinal ganglion cell (RGC) synapses with function deficits prior to manifest RGC loss in animal models of glaucoma. For testing the clinical relevance of experimental observations, this study analyzed the functional correlation of localized alterations in the inner plexiform layer (IPL), where RGCs establish synaptic connections with retinal bipolar and amacrine cells. Participants of the study included a retrospective cohort of 36 eyes with glaucoma and a control group of 18 non-glaucomatous subjects followed for two-years. The IPL was analyzed on consecutively collected macular SD-OCT scans, and functional correlations with corresponding 10-2 visual field scores were tested using generalized estimating equations (GEE) models. The GEE-estimated rate of decrease in IPL thickness (R = 0.36, P<0.001) and IPL density (R = 0.36, P<0.001), as opposed to unchanged or increased IPL thickness or density, was significantly associated with visual field worsening at corresponding analysis locations. Based on multivariate logistic regression analysis, this association was independent from the patients' age, the baseline visual field scores, or the baseline thickness or alterations of retinal nerve fiber or RGC layers (P>0.05). These findings support early localized IPL alterations in correlation with progressing visual field defects in glaucomatous eyes. Considering the experimental data, glaucoma-related increase in IPL thickness/density might reflect dendritic remodeling, mitochondrial redistribution, and glial responses for synapse maintenance, but decreased IPL thickness/density might correspond to dendrite atrophy. The bridging of experimental data with clinical findings encourages further research along the translational path.

Two patients presented with angle closure many years after cataract extraction. The first patient presented with acute intraocular pressure (IOP) elevation and closed iridocorneal angle that resolved with a laser iridotomy. The second patient presented with an insidious course of high IOP and progressive narrowing of the iridocorneal angle, ultimately requiring a pars plana vitrectomy and glaucoma valve implant, with subsequent normalization of pressure and angle anatomy. While rare, angle closure in eyes with posterior chamber intraocular lenses is a dangerous complication that can occur many years after cataract extraction. Retained lens fragments, and perhaps repeated intravitreal injections, might place susceptible patients at risk.

PURPOSE/OBJECTIVE:To describe the development of a new algorithm for detecting changes in 10-2 visual field (VF) tests using event-based analysis and to test its validity in a second, independent glaucoma cohort. DESIGN/METHODS:Prospective cohort study. METHODS:Patients with established open-angle glaucoma from the Macular Assessment and Progression Study (MAPS, development cohort, n=151), and the African Descent and Glaucoma Evaluation Study (ADAGES, validation cohort, n=52) were evaluated. The 10-2 VF results from MAPS were obtained during four test-retest sessions within a four-month period. For the validation analysis, 10-2 VF results from ADAGES performed on at least five visits were used. The event-based pointwise changes on 10-2 tests in the validation cohort were determined using two progression criteria: at least three progressing VF locations on two or three consecutive tests ("possible" or "likely" progression). Linear mixed-effects models were used to evaluate VF progression. RESULTS:In the validation cohort, the mean (SD) follow-up time was 2.3 (0.7) years. The number of eyes experiencing 10-2 VF progression based on "possible" and "likely" progression was 36 (54.5%) and 11 (16.6%), respectively. Eyes experiencing "possible" progression had MD changes [-0.60 dB/year (95% CI: -0.93 to -0.28)] faster than those not meeting this criterion (P<0.001), whereas for those with "likely" progression the difference was -0.91 dB/year (95% CI: -1.26 to -0.56, P<0.001). CONCLUSIONS:A new event-based progression algorithm using the 10-2 VF can identify eyes experiencing more rapid MD progression and may be used as a tool to assess progressive macular functional changes in glaucoma.

PURPOSE/OBJECTIVE:The aim of this study was to assess the benefit and feasibility of the teleophthalmology GlobeChek kiosk in a community-based program. DESIGN/METHODS:Single-site, nonrandomized, cross-sectional, teleophthalmologic study. METHODS:Participants underwent comprehensive evaluation that consists of a questionnaire form, brief systemic evaluation, screening visual field (VF), and GlobeChek kiosk screening, which included but not limited to intraocular pressure, pachymetry, anterior segment optical coherence tomography, posterior segment optical coherence tomography, and nonmydriatic fundus photography. The results were evaluated by a store-and-forward mechanism and follow-up questionnaires were obtained through phone calls. RESULTS:A total of 326 participatents were screened over 4 months. One hundred thirty-three (40.79%) participants had 1 condition in either eye, and 47 (14.41%) had >1 disease. Seventy (21.47%) had glaucoma, 37 (11.34%) narrow-angles, 6 (1.84%) diabetic retinopathy, 4 (1.22%) macular degeneration, and 43 (13.10%) had other eye disease findings. Age >65, history of high blood pressure, diabetes mellitus, not having a dental examination >5 years, hemoglobn A1c measurement of ≥5.6, predibates risk score of ≥9, stage 2 hypertension, and low blood pressure were found to be significant risk factors. As for the ocular parameters, all but central corneal thickness, including an intraocular pressure >21 mm Hg, vertical cup to disc ratio >0.7, visual field abnormalities, and retinal nerve fiber layer thinning were found to be significant. CONCLUSIONS:GlobeChek kiosk is both workable and effective in increasing access to care and identifying the most common causes of blindness and their risk factors.

OBJECTIVE:To describe reoperations in the operating room for complications encountered within 90 days following glaucoma surgery at a single institution over a two-year period. DESIGN/METHODS:Retrospective case series. SUBJECTS/METHODS:Adult patients who have undergone glaucoma surgery including a tube shunt, trabeculectomy with mitomycin C, trabectome or transcleral cyclophotocoagulation from June 1, 2015 to August 30, 2017 at a single institution. METHODS:These patients were then examined for postoperative complications that required reoperations within the first 90 days including revision of the tube shunt, revision of the trabeculectomy, drainage of the choroidals or placement of a tube shunt. MAIN OUTCOME MEASURES/METHODS:Percentage of reoperations for complications within the first 90 days following glaucoma surgery and surgical indications for these reoperations. RESULTS:622 glaucoma procedures were performed on 600 eyes in 525 patients over a two-year period from June 1, 2015 to June 30, 2017 by four glaucoma surgeons at a single institution. Of these, 275 (44%) were trabeculectomy with mitomycin C, 253 (41%) were placement of a tube shunt, 33 (5%) were cyclophotocoagulation, and 61 (10%) were trabectome procedures. Postoperative complications requiring reoperations within 90 days developed in 15 patients (2.4%) overall including 7 patients (2.5%) in the trabeculectomy with mitomycin C group and 8 patients (3.1%) in the tube shunt group. Five patients developed bleb leaks, 3 patients developed serous choroidal effusions, 3 patients had tube exposure, 1 patient had tube retraction, 1 patient had persistent iritis from iris touching the tube, and 1 had encapsulation around the tube. The rate of reoperation for complications was similar between the tube group and the trabeculectomy group (P=0.67, χ test). There were no complications requiring reoperations in 90 days for transcleral cyclophotocoagulation or trabectome. CONCLUSION/CONCLUSIONS:Early postoperative complications requiring reoperations within the first 90 days following glaucoma surgery was low and comparable to previous studies. Common indications for reoperation within 90 days include wound leak and tube shunt-related issues.

Purpose/UNASSIGNED:Our aim was to evaluate and compare the clinical outcomes after implantation of the silicone-plate (model FP7) and porous polyethylene-plate (model M4) Ahmed Glaucoma Valves. Patients and Methods/UNASSIGNED:This was a prospective, multicenter, comparative series. A total of 52 eyes (52 patients) were treated with either the silicone or porous plate Ahmed Glaucoma Valve implant. Hypertensive phase was defined as intraocular pressure >21 mmHg during the first 3 months postoperatively. Success was defined as 5 mmHg ≤intraocular pressure ≤21 mmHg (with or without additional glaucoma medications), without loss of light perception and without additional glaucoma procedures. Patients were monitored for 1 year after surgery. Results/UNASSIGNED:The pre-operative intraocular pressure was 29.9 ± 6.6 mmHg and 33.8 ± 10.5 in the silicone-plate and porous-plate groups, respectively (P = 0.118). At 12 months after surgery, the mean intraocular pressure was 13.6 ± 4.7 mmHg in the silicone-plate group and 17.9 ± 10.9 mmHg in the porous-plate group (P = 0.141). The mean number of glaucoma medications at 12 months was 1.64 ± 1.40 mmHg and 1.89 ± 1.54 mmHg in the silicone- and porous-plate groups, respectively (P = 0.605). Hypertensive phase was not significantly different in the two groups (50.0% of the silicone-plate and 57.7% of the porous-plate groups, P = 0.578). At 12 months after surgery, the percent success for the silicone-plate and porous-plate groups was 88.5% and 53.8%, respectively (P = 0.005). Complications were similar in the two groups. Conclusion/UNASSIGNED:The porous-plate Ahmed Glaucoma Valve showed similar average intraocular pressure reduction compared with the silicone-plate model. At 12 months after surgery, there was a significantly lower success rate in the porous-plate compared with the silicone-plate group.

Purpose : Optical coherence tomography (OCT) has become an essential tool to screen for glaucomatous optic neuropathy. Here we compare glaucoma referrals based on a customized one-page OCT report [1] with a more comprehensive community-based glaucoma screening protocol.[2] Methods : A community-based glaucoma screening study was conducted in a high-risk population in NYC from 2007-2014 using a mobile van. Each participant underwent visual acuity testing, OCT imaging, tonopen intraocular pressure assessment, frequency doubling testing, and optic nerve cup/disk ratio (C/D) funduscopic measurement. Trained ophthalmology residents/optometrists used all the data available to determine if ophthalmologic referral was needed based on suspicion for glaucoma. Topographic onepage OCT reports with RNFL and RGC thickness and probability plots were subsequently generated from volume scans of the optic disc and macula.[1] A report specialist judged each eye using these reports plus the commercially available OCT reports to similarly decide if a glaucoma referral was needed. The report specialist was blind to all other clinical data. Results : 188 patients (438 eyes) were used for this analysis. The inter-grader agreement between the two methods was kappa= 0.487. Of the 438 eyes, 329 (75.1%) had no evidence of glaucomatous damage per either screening methods and 43 (9.8%) were marked for referral by both. The 59 eyes referred by the van, but not the report specialist, were based on a combination of factors, the most prevalent being increased C/D (30/59) and increased IOP (20/59). 7 eyes had subtle defects picked up on the report, but not by the van screening protocol. Interestingly, circumpapillary RNFL quadrant analysis from the commercially available report picked up only 22 of the 43 eyes referred by both protocols. Conclusions : When comparing a community-based glaucoma screening protocol and a customized OCT report, a consensus was reached in 75% of cases (kappa= 0.49). While this is comparable to the inter-grader agreement reported for interpreting disc photos, OCT, and parametric data,[3,4] it should be possible to increase agreement by training screeners to identify subtle abnormalities in a customized OCT report

PRéCIS:: Pegasus outperformed five of the six ophthalmologists in terms of diagnostic performance, and there was no statistically significant difference between the deep learning system and the "best case" consensus between the ophthalmologists. The agreement between Pegasus and gold standard was 0.715, while the highest ophthalmologist agreement with the gold standard was 0. 613. Furthermore, the high sensitivity of Pegasus makes it a valuable tool for screening patients with glaucomatous optic neuropathy.