A comprehensive review of diagnostic imaging technologies to evaluate the retina and the optic disk

Feasibility of cross-vendor linkage of ophthalmic images with electronic health record data: an analysis from the IRIS Registry®

Purpose

To link compliant, universal Digital Imaging and Communications in Medicine (DICOM) ophthalmic imaging data at the individual patient level with the American Academy of Ophthalmology IRIS® Registry (Intelligent Research in Sight).

Design

A retrospective study using de-identified EHR registry data.

Subjects Participants Controls

IRIS Registry records.

Materials and Methods

DICOM files of several imaging modalities were acquired from two large retina ophthalmology practices. Metadata tags were extracted and harmonized to facilitate linkage to the IRIS Registry using a proprietary, heuristic patient-matching algorithm, adhering to HITRUST guidelines. Linked patients and images were assessed by image type and clinical diagnosis. Reasons for failed linkage were assessed by examining patients' records.

Main Outcome Measures

Success rate of linking clinicoimaging and EHR data at the patient level.

Results

A total of 2 287 839 DICOM files from 54 896 unique patients were available. Of these, 1 937 864 images from 46 196 unique patients were successfully linked to existing patients in the registry. After removing records with abnormal patient names and invalid birthdates, the success linkage rate was 93.3% for images. 88.2% of all patients at the participating practices were linked to at least one image.

Conclusions and Relevance

Using identifiers from DICOM metadata, we created an automated pipeline to connect longitudinal real-world clinical data comprehensively and accurately to various imaging modalities from multiple manufacturers at the patient and visit levels. The process has produced an enriched and multimodal IRIS Registry, bridging the gap between basic research and clinical care by enabling future applications in artificial intelligence algorithmic development requiring large linked clinicoimaging datasets.

Non-invasive in vivo imaging of brain and retinal microglia in neurodegenerative diseases

Microglia play crucial roles in immune responses and contribute to fundamental biological processes within the central nervous system (CNS). In neurodegenerative diseases, microglia undergo functional changes and can have both protective and pathogenic roles. Microglia in the retina, as an extension of the CNS, have also been shown to be affected in many neurological diseases. While our understanding of how microglia contribute to pathological conditions is incomplete, non-invasive in vivo imaging of brain and retinal microglia in living subjects could provide valuable insights into their role in the neurodegenerative diseases and open new avenues for diagnostic biomarkers. This mini-review provides an overview of the current brain and retinal imaging tools for studying microglia in vivo. We focus on microglia targets, the advantages and limitations of in vivo microglia imaging approaches, and applications for evaluating the pathogenesis of neurological conditions, such as Alzheimer's disease and multiple sclerosis.

The future application of artificial intelligence and telemedicine in the retina: A perspective

The development of artificial intelligence (AI) and deep learning provided precise image recognition and classification in the medical field. Ophthalmology is an exceptional department to translate AI applications since noninvasive imaging is routinely used for the diagnosis and monitoring. In recent years, AI-based image interpretation of optical coherence tomography and fundus photograph in retinal diseases has been extended to diabetic retinopathy, age-related macular degeneration, and retinopathy of prematurity. The rapid development of portable ocular monitoring devices coupled with AI-informed interpretations allows possible home monitoring or remote monitoring of retinal diseases and patients to gain autonomy and responsibility for their conditions. This review discusses the current research and application of AI, telemedicine, and home monitoring devices on retinal disease. Furthermore, we propose a future model of how AI and digital technology could be implemented in retinal diseases.

Improving foveal avascular zone segmentation in fluorescein angiograms by leveraging manual vessel labels from public color fundus pictures

In clinical routine, ophthalmologists frequently analyze the shape and size of the foveal avascular zone (FAZ) to detect and monitor retinal diseases. In order to extract those parameters, the contours of the FAZ need to be segmented, which is normally achieved by analyzing the retinal vasculature (RV) around the macula in fluorescein angiograms (FA). Computer-aided segmentation methods based on deep learning (DL) can automate this task. However, current approaches for segmenting the FAZ are often tailored to a specific dataset or require manual initialization. Furthermore, they do not take the variability and challenges of clinical FA into account, which are often of low quality and difficult to analyze. In this paper we propose a DL-based framework to automatically segment the FAZ in challenging FA scans from clinical routine. Our approach mimics the workflow of retinal experts by using additional RV labels as a guidance during training. Hence, our model is able to produce RV segmentations simultaneously. We minimize the annotation work by using a multi-modal approach that leverages already available public datasets of color fundus pictures (CFPs) and their respective manual RV labels. Our experimental evaluation on two datasets with FA from 1) clinical routine and 2) large multicenter clinical trials shows that the addition of weak RV labels as a guidance during training improves the FAZ segmentation significantly with respect to using only manual FAZ annotations.

Randomized, Comparative Study of Full- and Half-Dose Fluorescein Angiography

Purpose:

This study determines whether fluorescein angiography (FA) with a 250-mg dose of fluorescein (half dose) is equal in quality to the standard 500-mg dose of fluorescein (full dose) when using digital ultra-widefield (UWF) technology.

Methods:

In a randomized, prospective study using a UWF imaging system, FAs performed with half dose were compared with angiograms performed with full dose. Imaging studies were reviewed by 4 reviewers based on 6 characteristics: dye transit, macrovasculature, macula detail, microvasculature, leakage, and overall quality. The scores for macrovasculature, macula detail, microvasculature, and overall quality were converted to a fuzzy rating score to confirm results.

Results:

Seventy-nine FAs from 67 patients were reviewed for this study, including 12 patients who had both half-dose and full-dose FAs. Of all the factors studied, only microvasculature received a significantly different score between full dose and half dose that was confirmed by the fuzzy rating scale (3.79 vs 3.53; P = .04). Among those eyes that received both full and half dose, there was no significant difference in any of the 6 factors.

Conclusions:

In a UWF imaging system, aside from looking at fine microvascular abnormalities, the 250-mg dose of fluorescein provided similar results to a 500-mg dose. The images were not significantly different in overall quality.

The Relation between Body Mass Index and Retinal Photoreceptor Morphology and Microvascular Changes Measured with Adaptive Optics (rtx1) High-Resolution Imaging

Background

Overweight and obese patients are at risk for diabetes, cardiovascular disorders, and microvascular complications. The rtx1^TM (Imagine Eyes, France) is a microscope that allows near histological visualizations of cones and retinal microcirculation.

Objective

This study analysed the cones and retinal microvascular changes in a group of 47 healthy women with different BMI values. Participants were divided into 2 groups: the BMI group (28 women with BMI >/25) and the control group (19 lean women with BMI <25).

Results

The lumen and diameter of retinal arteries were not significantly different between groups. There were significant differences in the thickness of arteriole walls. The WLR and WCSA values differed significantly between the control and BMI groups (for WLR 0.25 ± 0.03 vs. 0.29 ± 0.03, p < 0.001; for WCSA 4136.7 ± 1140.0 vs. 5217.3 ± 944.0, respectively, p < 0.001). In healthy eyes, cone density and morphology were not affected by weight.

Conclusions

Retinal image analysis with rtx1 offers a novel noninvasive measurement of early changes in retinal vasculature that are not detectable during routine clinical examination. Abnormalities of retinal arterioles found by rtx1™examination should be considered as a strong risk factor for cardiovascular changes resulting from overweight and obesity.

Artificial intelligence and deep learning in ophthalmology - present and future (Review)

Since its introduction in 1959, artificial intelligence technology has evolved rapidly and helped benefit research, industries and medicine. Deep learning, as a process of artificial intelligence (AI) is used in ophthalmology for data analysis, segmentation, automated diagnosis and possible outcome predictions. The association of deep learning and optical coherence tomography (OCT) technologies has proven reliable for the detection of retinal diseases and improving the diagnostic performance of the eye's posterior segment diseases. This review explored the possibility of implementing and using AI in establishing the diagnosis of retinal disorders. The benefits and limitations of AI in the field of retinal disease medical management were investigated by analyzing the most recent literature data. Furthermore, the future trends of AI involvement in ophthalmology were analyzed, as AI will be part of the decision-making regarding the scientific investigation, diagnosis and therapeutic management.

Philadelphia Telemedicine Glaucoma Detection and Follow-Up Study: Cataract Classifications Following Eye Screening

Background: Cataracts are a major cause of visual impairment and blindness in the United States and worldwide. Introduction: Risk factors for cataracts include age over 40 years, smoking, diabetes, low socioeconomic status, female sex, steroid use, ocular trauma, genetic factors, and exposure to ultraviolet-B light. Community-based telemedicine vision screenings can be an efficient method for detecting cataracts in underserved populations. The Philadelphia Telemedicine Glaucoma Detection and Follow-Up Study reports the prevalence and risk factors for cataracts in individuals screened and examined for glaucoma and other eye diseases. Materials and Methods: A total of 906 high-risk individuals were screened for glaucoma using telemedicine in seven primary care practices and four Federally Qualified Health Centers in Philadelphia. Participants with suspicious nerves or other abnormalities on fundus photographs, unreadable images, and ocular hypertension returned for an eye examination with an ophthalmologist at the same community location. Results: Of the participants screened through telemedicine, 347 (38.3%) completed a follow-up eye examination by an ophthalmologist. Of these, 267 (76.9%) were diagnosed with cataracts, of which 38 (14.2%) had visually significant cataracts. Participants who were diagnosed with visually significant cataract were more likely to be older (p < 0.001), have diabetes (p = 0.003), and worse visual acuity (p < 0.001). Discussion: Our study successfully detected and confirmed cataracts in a targeted, underserved urban population at high risk for eye disease. Conclusions: Telemedicine programs offer an opportunity to identify and refer individuals who would benefit from continuous follow-up eye care and treatment to improve visual function and quality of life.

Philadelphia Telemedicine Glaucoma Detection and Follow-up Study: Analysis of Unreadable Fundus Images

PURPOSE

The purpose of this study was to ascertain determinants of unreadable fundus images for participants enrolled in the Philadelphia Telemedicine Glaucoma Detection and Follow-up Study.

METHODS

Individuals were screened for glaucoma at 7 primary care practices and 4 Federally Qualified Health Centers using telemedicine. Screening (visit 1) included fundus photography, assessing family history of glaucoma, and intraocular pressure (IOP) measurements. Participants with an unreadable image in at least one eye were deemed unreadable and invited to return for a confirmatory eye examination (visit 2).

RESULTS

A total of 906 participants completed the visit 1 eye screening and 17.1% (n=155/906) were "unreadable." In the multivariable logistic regression analysis, older age, male sex, smoking, and worse visual acuity were significantly associated with an unreadable fundus image finding at the eye screening (P<0.05). Of the 89 participants who were invited for the confirmatory eye examination solely for unreadable images and attended visit 2, 58 (65.2%) were diagnosed with at least one ocular pathology. The most frequent diagnoses were cataracts (n=71; 15 visually significant, 56 nonvisually significant), glaucoma suspects (n=27), and anatomical narrow angle (n=10).

CONCLUSIONS

Understanding the causes of unreadable fundus images will foster improvements in telemedicine techniques to optimize the predictive accuracy, efficiency, and cost in ophthalmology. A high proportion of participants with unreadable images (65.2%) in our study were diagnosed with some ocular pathology, indicating that the finding of an unreadable fundus image warrants a referral for a comprehensive follow-up eye examination.

Comparison of vascular parameters between normal cynomolgus macaques and healthy humans by optical coherence tomography angiography

BACKGROUND

The metabolic activity of retina is higher than other human tissues and is crucial to the vision. Cynomolgus macaques is widely used in ophthalmic disease research. The evaluation and comparison of macular and optic disc vascular circulation parameters between normal adult cynomolgus macaques and healthy adult humans using OCT-A can promote better use of nonhuman primate models in studies of ophthalmic vascular disease.

METHODS

Twelve normal adult cynomolgus macaques with a mean age of 4.91 ± 0.43 years were studied for data collection. The macula of 28 adult healthy humans (14 males and 14 females), with a mean age of 25.11 ± 6.21 years and the optic discs of 9 adult healthy humans (4 males and 5 females) with a mean age of 28.56 ± 6.78 years were measured. The vessel density (VD) was measured using an RTVue XR with AngioVue. The scan sizes of the macular and optic discs were 3 × 3 mm and 4.5 × 4.5 mm, respectively.

RESULTS

OCT-A can image the superficial and deep capillary plexuses and radial peripapillary capillary network. In RPC layer of the optic disc, the VD in the nasal quadrant was lower than the VD in the inferior temporal quadrant. Similarities and significant differences in VD between healthy humans and cynomolgus macaques were obtained using OCT-A.

CONCLUSIONS

This study provides normal vascular parameters for adult cynomolgus macaques using OCT-A to help establish an optical parameter database for cynomolgus macaques and compare VD between healthy humans and cynomolgus macaques to promote choroid-retinopathy research.

TRIAL REGISTRATION

Current Controlled Trials NCT03692169 , retrospectively registered on 26 sept 2018.

Retinal α-synuclein deposits in Parkinson's disease patients and animal models

Despite decades of research, accurate diagnosis of Parkinson's disease remains a challenge, and disease-modifying treatments are still lacking. Research into the early (presymptomatic) stages of Parkinson's disease and the discovery of novel biomarkers is of utmost importance to reduce this burden and to come to a more accurate diagnosis at the very onset of the disease. Many have speculated that non-motor symptoms could provide a breakthrough in the quest for early biomarkers of Parkinson's disease, including the visual disturbances and retinal abnormalities that are seen in the majority of Parkinson's disease patients. An expanding number of clinical studies have investigated the use of in vivo assessments of retinal structure, electrophysiological function, and vision-driven tasks as novel means for identifying patients at risk that need further neurological examination and for longitudinal follow-up of disease progression in Parkinson's disease patients. Often, the results of these studies have been interpreted in relation to α-synuclein deposits and dopamine deficiency in the retina, mirroring the defining pathological features of Parkinson's disease in the brain. To better understand the visual defects seen in Parkinson's disease patients and to propel the use of retinal changes as biomarkers for Parkinson's disease, however, more conclusive neuropathological evidence for the presence of retinal α-synuclein aggregates, and its relation to the cerebral α-synuclein burden, is urgently needed. This review provides a comprehensive and critical overview of the research conducted to unveil α-synuclein aggregates in the retina of Parkinson's disease patients and animal models, and thereby aims to aid the ongoing discussion about the potential use of the retinal changes and/or visual symptoms as biomarkers for Parkinson's disease.

Angiopoietin-4-dependent venous maturation and fluid drainage in the peripheral retina

The maintenance of fluid homeostasis is necessary for function of the neural retina; however, little is known about the significance of potential fluid management mechanisms. Here, we investigated angiopoietin-4 (Angpt4, also known as Ang3), a poorly characterized ligand for endothelial receptor tyrosine kinase Tie2, in mouse retina model. By using genetic reporter, fate mapping, and in situ hybridization, we found Angpt4 expression in a specific sub-population of astrocytes at the site where venous morphogenesis occurs and that lower oxygen tension, which distinguishes peripheral and venous locations, enhances Angpt4 expression. Correlating with its spatiotemporal expression, deletion of Angpt4 resulted in defective venous development causing impaired venous drainage and defects in neuronal cells. In vitro characterization of angiopoietin-4 proteins revealed both ligand-specific and redundant functions among the angiopoietins. Our study identifies Angpt4 as the first growth factor for venous-specific development and its importance in venous remodeling, retinal fluid clearance and neuronal function.

Fluorescein Labeled Leukocytes for in vivo Imaging of Retinal Vascular Inflammation and Infiltrating Leukocytes in Laser-Induced Choroidal Neovascularization Model

Purpose: To study the effect of anti-VEGF treatment on retinal inflammation in a laser-induced CNV rodent model.Methods: Leukocytes labeled with 1% sodium fluorescein were injected into the laser-induced CNV (wild type C57BL/6) mice at days 4 (baseline), 7, 14, and 19. At baseline intravitreally 3 mice received 1× PBS, and 3 mice received anti-VEGF. FFA, OCT, and SLO were performed at each time point to assess the CNV pathophysiology and inflammatory response.Results: Fluorescein leakage, SRF, and leukocyte infiltration were observed at baseline in both the groups before injection. From days 7 to 19, leukocyte infiltration and SRF were noted in the 1× PBS group, but limited or no SRF and leukocyte infiltration was observed in the anti-VEGF group.Conclusions: Leukocyte infiltration was established as an in vivo imaging inflammatory marker and along with FFA and OCT showed response to anti-VEGF therapy in laser-induced CNV model.

Retinal vasculature development in health and disease

Development of the retinal vasculature is based on highly coordinated signalling between different cell types of the retina, integrating internal metabolic requirements with external influences such as the supply of oxygen and nutrients. The developing mouse retinal vasculature is a useful model system to study these interactions because it is experimentally accessible for intra ocular injections and genetic manipulations, can be easily imaged and develops in a similar fashion to that of humans. Research using this model has provided insights about general principles of angiogenesis as well as pathologies that affect the developing retinal vasculature. In this review, we discuss recent advances in our understanding of the molecular and cellular mechanisms that govern the interactions between neurons, glial and vascular cells in the developing retina. This includes a review of mechanisms that shape the retinal vasculature, such as sprouting angiogenesis, vascular network remodelling and vessel maturation. We also explore how the disruption of these processes in mice can lead to pathology - such as oxygen induced retinopathy - and how this translates to human retinopathy of prematurity.

Photoacoustic-Guided Surgery with Indocyanine Green-Coated Superparamagnetic Iron Oxide Nanoparticle Clusters

A common cause of local tumor recurrence in brain tumor surgery results from incomplete surgical resection. Adjunctive technologies meant to facilitate gross total resection have had limited efficacy to date. Contrast agents used to delineate tumors preoperatively cannot be easily or accurately used in the real-time operative setting. Although multimodal imaging contrast agents are developed to help the surgeon discern tumor from normal tissue in the operating room, these contrast agents are not readily translatable. This study has developed a novel contrast agent comprised solely of two Food and Drug Administration approved components, indocyanine green (ICG) and superparamagnetic iron oxide (SPIO) nanoparticles-with no additional amphiphiles or carrier materials, to enable preoperative detection by magnetic resonance (MR) imaging and intraoperative photoacoustic (PA) imaging. The encapsulation efficiency of both ICG and SPIO within the formulated clusters is ≈100%, and the total ICG payload is 20-30% of the total weight (ICG + SPIO). The ICG-SPIO clusters are stable in physiologic conditions; can be taken up within tumors by enhanced permeability and retention; and are detectable by MR. In a preclinical surgical resection model in mice, following injection of ICG-SPIO clusters, animals undergoing PA-guided surgery demonstrate increased progression-free survival compared to animals undergoing microscopic surgery.

Human photoreceptor cone density measured with adaptive optics technology (rtx1 device) in healthy eyes: Standardization of measurements

The anatomic structures of the anterior segment of the eye enable correct reception of stimuli by the retina, which contains receptors that receive light impulses and transmit them to the visual cortex. The aim of this study was to assess the effect of the size of the sampling window in an adaptive optics (AO) flood-illumination retinal camera (rtx1) on cone density measurements in the eyes of healthy individuals and to investigate the differences in cone density and spacing in different quadrants of the retina. Thirty-three subjects with no ophthalmic or systemic disease underwent a detailed ophthalmologic examination. Photographs of retinal fragments 3 degrees from the fovea were taken using the rtx1 AO retinal camera. We used sampling windows with 3 sizes (50 × 50, 100 × 100, and 250 × 250 μm). Cone density, spacing, and shape were determined using AOdetect software. The median (interquartile range) cone density was 19,269 (4964) cones/mm. There were statistically significant differences between measurements taken with the 50/50 and 250/250-m windows. There were no significant differences in the cone spacing results between any of the windows examined, but the measurements differed according to location between the superior and temporal quadrants. The most common cone shape was hexagonal (47.6%) for all window sizes and locations. These findings may help in the development of a normative database for variation in cone density in healthy subjects and to allow the best window to be chosen for obtain the most correct values for eccentricity measurements of 3 degrees. In our study, the optimal sampling window was 100 × 100 μm.

Whole-globe biomechanics using high-field MRI

The eye is a complex structure composed of several interconnected tissues acting together, across the whole globe, to resist deformation due to intraocular pressure (IOP). However, most work in the ocular biomechanics field only examines the response to IOP over smaller regions of the eye. We used high-field MRI to measure IOP induced ocular displacements and deformations over the whole globe. Seven sheep eyes were obtained from a local abattoir and imaged within 48 h using MRI at multiple levels of IOP. IOP was controlled with a gravity perfusion system and a cannula inserted into the anterior chamber. T2-weighted imaging was performed to the eyes serially at 0 mmHg, 10 mmHg, 20 mmHg and 40 mmHg of IOP using a 9.4 T MRI scanner. Manual morphometry was conducted using 3D visualization software to quantify IOP-induced effects at the globe scale (e.g. axial length and equatorial diameters) or optic nerve head scale (e.g. canal diameter, peripapillary sclera bowing). Measurement sensitivity analysis was conducted to determine measurement precision. High-field MRI revealed an outward bowing of the posterior sclera and anterior bulging of the cornea due to IOP elevation. Increments in IOP from 10 to 40 mmHg caused measurable increases in axial length in 6 of 7 eyes of 7.9 ± 5.7% (mean ± SD). Changes in equatorial diameter were minimal, 0.4 ± 1.2% between 10 and 40 mmHg, and in all cases less than the measurement sensitivity. The effects were nonlinear, with larger deformations at normal IOPs (10-20 mmHg) than at elevated IOPs (20-40 mmHg). IOP also caused measurable increases in the nasal-temporal scleral canal diameter of 13.4 ± 9.7% between 0 and 20 mmHg, but not in the superior-inferior diameter. This study demonstrates that high-field MRI can be used to visualize and measure simultaneously the effects of IOP over the whole globe, including the effects on axial length and equatorial diameter, posterior sclera displacement and bowing, and even changes in scleral canal diameter. The fact that the equatorial diameter did not change with IOP, in agreement with previous studies, indicates that a fixed boundary condition is a reasonable assumption for half globe inflation tests and computational models. Our results demonstrate the potential of high-field MRI to contribute to understanding ocular biomechanics, and specifically of the effects of IOP in large animal models.

Rates of Vitrectomy among Enrollees in a United States Managed Care Network, 2001-2012

PURPOSE

To determine whether vitrectomy surgery rates have changed over the past decade and factors affecting the odds of undergoing this procedure.

DESIGN

Retrospective, longitudinal cohort study.

PARTICIPANTS

All enrollees 21 years of age or older between 2001 and 2012 in a United States managed care network.

METHODS

Claims data from a managed care network were analyzed to identify all enrollees who underwent 1 vitrectomy or more each year from 2001 through 2012. Rates of vitrectomy per 1000 enrollees were computed each year from 2001 through 2012 for the entire group and separately for patients with and without diabetes mellitus. Multivariate logistic regression assessed factors affecting the odds of undergoing vitrectomy surgery.

MAIN OUTCOME MEASURES

Annual rates of vitrectomy surgery from 2001 through 2012 and odds ratios (ORs) of undergoing a vitrectomy with 95% confidence intervals (CIs).

RESULTS

Among the 11 161 907 eligible enrollees, 40 892 (0.4%) underwent vitrectomy over the 12-year period. The average age of those undergoing vitrectomy was 57±13 years. Overall vitrectomy rates increased 31% from 2001 to 2012 (from 1.47 to 1.92 per 1000 patients). During this same period, the vitrectomy rate among persons with diabetes mellitus decreased by 43% (from 5.84 to 3.31 per 1000 patients with diabetes). Women had 24% decreased odds of undergoing vitrectomy (adjusted odds ratio [OR], 0.76; 95% confidence interval [CI], 0.72-0.79). The odds of undergoing a vitrectomy were 17% greater for black persons (adjusted OR, 1.17; 95% CI, 1.07-1.27) and 7% higher for persons with diabetes (adjusted OR, 1.07; 95% CI, 1.01-1.14).

CONCLUSIONS

Overall, we observed an increase in the vitrectomy rates per 1000 enrollees in this large managed care network over the course of the past decade. However, among persons with diabetes mellitus, vitrectomy rates declined substantially over this period. These changes may be explained, in part, by advances in surgical instrumentation and imaging methods to detect retinal diseases changing indications for surgery, improvements in diabetes care, and alternative treatment options for managing retinal conditions. These results may be useful for future planning of manpower needs and highlight the need for aggressive prevention of complications in black persons with diabetes.