Performance of retinal fluid monitoring in OCT imaging by automated deep learning versus human expert grading in neovascular AMD

PURPOSE

To evaluate the reliability of automated fluid detection in identifying retinal fluid activity in OCT scans of patients treated with anti-VEGF therapy for neovascular age-related macular degeneration by correlating human expert and automated measurements with central retinal subfield thickness (CSFT) and fluid volume values.

METHODS

We utilized an automated deep learning approach to quantify macular fluid in SD-OCT volumes (Cirrus, Spectralis, Topcon) from patients of HAWK and HARRIER Studies. Three-dimensional volumes for IRF and SRF were measured at baseline and under therapy in the central millimeter and compared to fluid gradings, CSFT and foveal centerpoint thickness (CPT) values measured by the Vienna Reading Center.

RESULTS

41.906 SD-OCT volume scans were included into the analysis. Concordance between human expert grading and automated algorithm performance reached AUC values of 0.93/0.85 for IRF and 0.87 for SRF in HARRIER/HAWK in the central millimeter. IRF volumes showed a moderate correlation with CSFT at baseline (HAWK: r = 0.54; HARRIER: r = 0.62) and weaker correlation under therapy (HAWK: r = 0.44; HARRIER: r = 0.34). SRF and CSFT correlations were low at baseline (HAWK: r = 0.29; HARRIER: r = 0.22) and under therapy (HAWK: r = 0.38; HARRIER: r = 0.45). The residual standard error (IRF: 75.90 µm; SRF: 95.26 µm) and marginal residual standard deviations (IRF: 46.35 µm; SRF: 44.19 µm) of fluid volume were high compared to the range of CSFT values.

CONCLUSION

Deep learning-based segmentation of retinal fluid performs reliably on OCT images. CSFT values are weak indicators for fluid activity in nAMD. Automated quantification of fluid types, highlight the potential of deep learning-based approaches to objectively monitor anti-VEGF therapy.

Automatic segmentation of multitype retinal fluid from optical coherence tomography images using semisupervised deep learning network

BACKGROUND/AIMS

To develop and validate a deep learning model for automated segmentation of multitype retinal fluid using optical coherence tomography (OCT) images.

METHODS

We retrospectively collected a total of 2814 completely anonymised OCT images with subretinal fluid (SRF) and intraretinal fluid (IRF) from 141 patients between July 2018 and June 2020, constituting our in-house retinal OCT dataset. On this dataset, we developed a novel semisupervised retinal fluid segmentation deep network (Ref-Net) to automatically identify SRF and IRF in a coarse-to-refine fashion. We performed quantitative and qualitative analyses on the model's performance while verifying its generalisation ability by using our in-house retinal OCT dataset for training and an unseen Kermany dataset for testing. We also determined the importance of major components in the semisupervised Ref-Net through extensive ablation. The main outcome measures were Dice similarity coefficient (Dice), sensitivity (Sen), specificity (Spe) and mean absolute error (MAE).

RESULTS

Our model trained on a handful of labelled OCT images manifested higher performance (Dice: 81.2%, Sen: 87.3%, Spe: 98.8% and MAE: 1.1% for SRF; Dice: 78.0%, Sen: 83.6%, Spe: 99.3% and MAE: 0.5% for IRF) over most cutting-edge segmentation models. It obtained expert-level performance with only 80 labelled OCT images and even exceeded two out of three ophthalmologists with 160 labelled OCT images. Its satisfactory generalisation capability across an unseen dataset was also demonstrated.

CONCLUSION

The semisupervised Ref-Net required only la few labelled OCT images to generate outstanding performance in automate segmentation of multitype retinal fluid, which has the potential for providing assistance for clinicians in the management of ocular disease.

COMPARATIVE EVALUATION OF OUTCOMES OF DRAINAGE TECHNIQUES IN VITRECTOMY FOR RHEGMATOGENOUS RETINAL DETACHMENT

PURPOSE

To compare the anatomical and functional outcomes of drainage through posterior retinotomy versus perfluorocarbon liquid (PFCL)-assisted drainage in vitreoretinal surgery for rhegmatogenous retinal detachment and to study intraoperative and postoperative complications.

METHODS

This was a prospective randomized study of 52 cases who underwent vitreoretinal surgery for rhegmatogenous retinal detachment. Group 1 underwent PFCL-assisted drainage through preexisting break, whereas Group 2 had posterior retinotomy to drain subretinal fluid. Cases were evaluated for retinal reattachment rates, visual outcomes, optical coherence tomography parameters, and postoperative metamorphopsia. The patients were followed up for minimum period of 3 months.

RESULTS

Two groups were comparable in terms of demographic and preoperative parameters. Both groups had single surgery success rate of 100% by the end of follow-up. Final best-corrected visual acuity in Group 1 was 0.61 ± 0.33 and 0.61 ± 0.32 in Group 2 (P = 0.77). Optical coherence tomography parameters (foveal contour, retinal layers, central macular thickness, and epiretinal membrane formation) were similar between the two groups. Subjective metamorphopsia was present in 30.77% (8 of 26) patients in Group-1 and 69.23% (18 of 26) patients in Group-2 (P = 0.034). One eye had retained subretinal PFCL away from the macula in Group 1.

CONCLUSION

Anatomical and functional outcomes were similar in vitrectomy using PFCL-assisted drainage versus posterior retinotomy drainage. Postoperative metamorphopsia was lesser in patients who underwent PFCL-assisted drainage through the pre-existing break.

IMPACT OF RESIDUAL SUBRETINAL FLUID VOLUMES ON TREATMENT OUTCOMES IN A SUBRETINAL FLUID-TOLERANT TREAT-AND-EXTEND REGIMEN

PURPOSE

To investigate associations between residual subretinal fluid (rSRF) volumes, quantified using artificial intelligence and treatment outcomes in a subretinal fluid (SRF)-tolerant treat-and-extend (T&E) regimen in neovascular age-related macular degeneration.

METHODS

Patients enrolled in the prospective, multicenter FLUID study randomized in an SRF-tolerant T&E regimen were examined by spectral-domain optical coherence tomography and tested for best-corrected visual acuity (BCVA). Intraretinal fluid and SRF volumes were quantified using artificial intelligence tools. In total, 375 visits of 98 patients were divided into subgroups: extended intervals despite rSRF and extended intervals without fluid. Associations between BCVA change, SRF volume, subgroups, and treatment intervals were estimated using linear mixed models.

RESULTS

In extended intervals despite rSRF, increased SRF was associated with reduced BCVA at the next visit in the central 1 mm (-0.138 letters per nL; P = 0.014) and 6 mm (-0.024 letters per nL; P = 0.049). A negative association between increased interval and BCVA change was found for rSRF in 1 mm and 6 mm (-0.250 and -0.233 letter per week interval, respectively; both P < 0.001). Extended intervals despite rSRF had significantly higher SRF volumes in the central 6 mm at the following visit (P = 0.002).

CONCLUSION

Artificial intelligence-based analysis of extended visits despite rSRF demonstrated increasing SRF volumes associated with BCVA loss at the consecutive visit. This negative association contributes to the understanding of rSRF volumes on treatment outcomes in neovascular age-related macular degeneration.

Predominantly Persistent Subretinal Fluid in the Comparison of Age-Related Macular Degeneration Treatments Trials

OBJECTIVE

To describe predominantly persistent subretinal fluid (SRF) in eyes receiving ranibizumab or bevacizumab for neovascular age-related macular degeneration and to compare visual acuity (VA) to eyes with nonpersistent SRF.

DESIGN

Cohort within randomized clinical trial.

PARTICIPANTS

Comparison of Age-related Macular Degeneration Treatments Trials patients assigned to pro re nata treatment.

METHODS

Graders evaluated monthly OCT scans for SRF. Predominantly persistent SRF through week 12 was defined as SRF at baseline and weeks 4, 8, and 12. Predominantly persistent SRF through 1 or 2 years was defined as SRF in 80% or more of visits by years 1 or 2, respectively. Linear regression models including baseline predictors of VA and predominantly persistent intraretinal fluid (IRF) were used to evaluate mean differences in vision outcomes.

PRIMARY OUTCOME MEASURES

Predominantly persistent SRF through year 1, adjusted VA score and VA change, and foveal SRF thickness.

RESULTS

Among 406 eyes with baseline SRF, SRF persisted in 108 eyes (26.6%) through week 12, in 94 eyes (23.2%) through year 1, and in 77 eyes (19.0%) through year 2. Adjusted VA means at year 1 were similar between eyes with predominantly persistent versus non persistent SRF by week 12 (68.1 vs. 70.2 letters; P = 0.18), year 1 (67.6 vs. 70.2 letters; P = 0.11), and year 2 (71.4 vs. 70.9 letters; P = 0.78). Adjusted changes in mean VA at year 1 were similar between eyes with predominantly persistent versus nonpersistent SRF by week 12 (6.3 vs. 7.6 letters; P = 0.38), year 1 (5.5 vs. 7.8 letters; P = 0.14), and year 2 (8.1 vs. 7.7 letters; P = 0.78). Among eyes with predominantly persistent SRF through year 1, foveal SRF was absent in 46 eyes (48.9%), thickness was 1 to 200 μm in 48 eyes (50.0%) and more than 200 μm in 1 eye (1.1%) at year 1.

CONCLUSIONS

Eyes with predominantly persistent and nonpersistent SRF through week 12, year 1, or year 2 showed similar VA outcomes after adjustment for baseline covariates and persistent IRF. At the foveal center, predominantly persistent SRF was most commonly absent or present in small quantities.

EXTERNAL DRAINAGE OF SUBRETINAL FLUID DURING RHEGMATOGENOUS RETINAL DETACHMENT REPAIR

PURPOSE

To describe the safety and efficacy of rhegmatogenous retinal detachment (RRD) repair with external drainage of subretinal fluid using a 28-gauge External Drainage and Depression device (Vortex Surgical, Chesterfield, MO).

METHODS

Retrospective review of patients who underwent primary rhegmatogenous retinal detachment repair with scleral buckle, pars plana vitrectomy, or scleral buckle/pars plana vitrectomy using the drainage device from August 2018 through March 2020, performed by four surgeons at two vitreoretinal practices.

RESULTS

Eighty-three eyes of 83 patients were included. At presentation, 28% had proliferative vitreoretinopathy. Surgery included 65% scleral buckle/pars plana vitrectomy, 33% pars plana vitrectomy, and 2% scleral buckle. There were no cases of retinal incarceration and two subretinal hemorrhages at the drainage site (both < 2 DD), 2 cases of recurrent RD with proliferative vitreoretinopathy (1 had proliferative vitreoretinopathy at presentation), and 6 (10%) new epiretinal membranes (3 were mild). There were no other complications. Mean follow-up was 274 days. Single operation success rate for those with ≥ 6-month follow-up was 97% (57/59).

CONCLUSION

External drainage of subretinal fluid during rhegmatogenous retinal detachment repair demonstrated a favorable safety profile with a high single operation success rate. Further study of the role of external drainage in rhegmatogenous retinal detachment repair is warranted.

Deep learning based joint segmentation and characterization of multi-class retinal fluid lesions on OCT scans for clinical use in anti-VEGF therapy

BACKGROUND

In anti-vascular endothelial growth factor (anti-VEGF) therapy, an accurate estimation of multi-class retinal fluid (MRF) is required for the activity prescription and intravitreal dose. This study proposes an end-to-end deep learning-based retinal fluids segmentation network (RFS-Net) to segment and recognize three MRF lesion manifestations, namely, intraretinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED), from multi-vendor optical coherence tomography (OCT) imagery. The proposed image analysis tool will optimize anti-VEGF therapy and contribute to reducing the inter- and intra-observer variability.

METHOD

The proposed RFS-Net architecture integrates the atrous spatial pyramid pooling (ASPP), residual, and inception modules in the encoder path to learn better features and conserve more global information for precise segmentation and characterization of MRF lesions. The RFS-Net model is trained and validated using OCT scans from multiple vendors (Topcon, Cirrus, Spectralis), collected from three publicly available datasets. The first dataset consisted of OCT volumes obtained from 112 subjects (a total of 11,334 B-scans) is used for both training and evaluation purposes. Moreover, the remaining two datasets are only used for evaluation purposes to check the trained RFS-Net's generalizability on unseen OCT scans. The two evaluation datasets contain a total of 1572 OCT B-scans from 1255 subjects. The performance of the proposed RFS-Net model is assessed through various evaluation metrics.

RESULTS

The proposed RFS-Net model achieved the mean F₁ scores of 0.762, 0.796, and 0.805 for segmenting IRF, SRF, and PED. Moreover, with the automated segmentation of the three retinal manifestations, the RFS-Net brings a considerable gain in efficiency compared to the tedious and demanding manual segmentation procedure of the MRF.

CONCLUSIONS

Our proposed RFS-Net is a potential diagnostic tool for the automatic segmentation of MRF (IRF, SRF, and PED) lesions. It is expected to strengthen the inter-observer agreement, and standardization of dosimetry is envisaged as a result.

RETINAL FLUID AND THICKNESS AS MEASURES OF DISEASE ACTIVITY IN NEOVASCULAR AGE-RELATED MACULAR DEGENERATION

PURPOSE

Retinal fluid and thickness are important anatomical features of disease activity in neovascular age-related macular degeneration, as evidenced by clinical trials that have used these features for inclusion criteria, retreatment criteria, and outcome measures of the efficacy of intravitreal injections of anti-vascular endothelial growth factor agents.

METHODS

A literature review of anatomical measures of disease activity was conducted.

RESULTS

Treatment goals for neovascular age-related macular degeneration include improving/maintaining vision by drying the retina, and several analyses have evaluated the relationship between visual function and anatomy. The change in retinal thickness has been found to correlate with the change in the visual acuity, and variation in retinal thickness may predict visual acuity outcomes. In addition, specific fluid compartments may have different prognostic values. For example, the presence of intraretinal fluid has been associated with poorer visual acuity, whereas the presence of subretinal fluid has been associated with better visual acuity. Retinal fluid and thickness are important for selecting dosing interval durations in clinical trials and clinical practice.

CONCLUSION

Retinal thickness and retinal fluid are common anatomical measures of disease activity in neovascular age-related macular degeneration. Further research is required to fully elucidate the relationship between anatomical features and visual outcomes in neovascular age-related macular degeneration.

RetFluidNet: Retinal Fluid Segmentation for SD-OCT Images Using Convolutional Neural Network

Age-related macular degeneration (AMD) is one of the leading causes of irreversible blindness and is characterized by fluid-related accumulations such as intra-retinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED). Spectral-domain optical coherence tomography (SD-OCT) is the primary modality used to diagnose AMD, yet it does not have algorithms that directly detect and quantify the fluid. This work presents an improved convolutional neural network (CNN)-based architecture called RetFluidNet to segment three types of fluid abnormalities from SD-OCT images. The model assimilates different skip-connect operations and atrous spatial pyramid pooling (ASPP) to integrate multi-scale contextual information; thus, achieving the best performance. This work also investigates between consequential and comparatively inconsequential hyperparameters and skip-connect techniques for fluid segmentation from the SD-OCT image to indicate the starting choice for future related researches. RetFluidNet was trained and tested on SD-OCT images from 124 patients and achieved an accuracy of 80.05%, 92.74%, and 95.53% for IRF, PED, and SRF, respectively. RetFluidNet showed significant improvement over competitive works to be clinically applicable in reasonable accuracy and time efficiency. RetFluidNet is a fully automated method that can support early detection and follow-up of AMD.

Lightweight Learning-Based Automatic Segmentation of Subretinal Blebs on Microscope-Integrated Optical Coherence Tomography Images

PURPOSE

Subretinal injections of therapeutics are commonly used to treat ocular diseases. Accurate dosing of therapeutics at target locations is crucial but difficult to achieve using subretinal injections due to leakage, and there is no method available to measure the volume of therapeutics successfully administered to the subretinal location during surgery. Here, we introduce the first automatic method for quantifying the volume of subretinal blebs, using porcine eyes injected with Ringer's lactate solution as samples.

DESIGN

Ex vivo animal study.

METHODS

Microscope-integrated optical coherence tomography was used to obtain 3D visualization of subretinal blebs in porcine eyes at Duke Eye Center. Two different injection phases were imaged and analyzed in 15 eyes (30 volumes), selected from a total of 37 eyes. The inclusion/exclusion criteria were set independently from the algorithm-development and testing team. A novel lightweight, deep learning-based algorithm was designed to segment subretinal bleb boundaries. A cross-validation method was used to avoid selection bias. An ensemble-classifier strategy was applied to generate final results for the test dataset.

RESULTS

The algorithm performs notably better than 4 other state-of-the-art deep learning-based segmentation methods, achieving an F1 score of 93.86 ± 1.17% and 96.90 ± 0.59% on the independent test data for entry and full blebs, respectively.

CONCLUSION

The proposed algorithm accurately segmented the volumetric boundaries of Ringer's lactate solution delivered into the subretinal space of porcine eyes with robust performance and real-time speed. This is the first step for future applications in computer-guided delivery of therapeutics into the subretinal space in human subjects.

Model-to-Data Approach for Deep Learning in Optical Coherence Tomography Intraretinal Fluid Segmentation

Importance

Amid an explosion of interest in deep learning in medicine, including within ophthalmology, concerns regarding data privacy, security, and sharing are of increasing importance. A model-to-data approach, in which the model itself is transferred rather than data, can circumvent many of these challenges but has not been previously demonstrated in ophthalmology.

Objective

To determine whether a model-to-data deep learning approach (ie, validation of the algorithm without any data transfer) can be applied in ophthalmology.

Design, Setting, and Participants

This single-center cross-sectional study included patients with active exudative age-related macular degeneration undergoing optical coherence tomography (OCT) at the New England Eye Center from August 1, 2018, to February 28, 2019. Data were primarily analyzed from March 1 to June 20, 2019.

Main Outcomes and Measures

Training of the deep learning model, using a model-to-data approach, in recognizing intraretinal fluid (IRF) on OCT B-scans.

Results

The model was trained (learning curve Dice coefficient, >80%) using 400 OCT B-scans from 128 participants (69 female [54%] and 59 male [46%]; mean [SD] age, 77.5 [9.1] years). In comparing the model with manual human grading of IRF pockets, no statistically significant difference in Dice coefficients or intersection over union scores was found (P > .05).

Conclusions and Relevance

A model-to-data approach to deep learning applied in ophthalmology avoided many of the traditional hurdles in large-scale deep learning, including data sharing, security, and privacy concerns. Although the clinical relevance of these results is limited at this time, this proof-of-concept study suggests that such a paradigm should be further examined in larger-scale, multicenter deep learning studies.

Effect of Intravitreal Ranibizumab on Intraretinal Hard Exudates in Eyes with Diabetic Macular Edema

PURPOSE

To study the effect of monthly 0.3-mg intravitreal ranibizumab injections on intraretinal hard exudates (HEs) and correlate the effect with macular thickness in eyes with diabetic macular edema (DME).

DESIGN

Randomized, controlled trial.

METHODS

Twenty-four eyes of 24 subjects with DME were included in this institutional review board-approved, prospective longitudinal study. Subjects were randomly assigned to receive monthly intravitreal ranibizumab injections either until macular edema resolved or until both macular edema and HEs resolved. All subjects underwent spectral-domain optical coherence tomography imaging (Cirrus OCT), at baseline and monthly for 12 months. Mean HE area and mean macular thickness at baseline and month 12 were compared using a paired t test and correlated with Pearson analysis.

RESULTS

The average age of the 24 subjects was 65 (±8.55) years. There was a significant decrease (P = .001) in mean HE area from baseline (0.48 ± 0.43 mm²) to month 12 (0.17 ± 0.19 mm²). There was a significant increase (P < .001) in best-corrected visual acuity from a letter score of 63.38 (±7.92) at baseline to 76.38 (±8.93) at month 12. There was no additional vision benefit for continuing to treat with monthly injections for persistent HE in the absence of macular thickening.

CONCLUSION

In eyes with DME, monthly intravitreal ranibizumab injections resulted in significant reduction in intraretinal HEs that paralleled reductions in macular thickness and volume. Baseline foveal HEs were associated with worse vision outcomes, and continued treatment for persistent HEs in the absence of edema did not result in better vision.

Submacular perfluorocarbon liquid: long-term follow-up

PURPOSE

To report the long-term follow-up of patients with inadvertent retained submacular perfluorocarbon liquid (PFCL) bubbles after vitrectomy surgery.

BACKGROUND

PFCL has unique chemical and physical features which facilitate displacement of subretinal fluid from central toward periphery allowing intraoperative reattachment of the mobile retina and manipulation of the anterior part of the detached retina without performing drainage retinotomy. Despite these advantages, PFCL droplets may be inadvertently retained in the submacular space.

METHODS

A retrospective case series study was conducted. We reviewed 1224 consecutive patients who had undergone primary pars plana vitrectomy at one vitreoretinal center operated by single surgeon (Hammouda Ghoraba). Chart review was performed to identify patients and procedural factors that could predispose to retained submacular PFCL with anatomical and functional outcomes with follow-up at least 18 months.

RESULTS

The retina was completely attached in 97% of eyes where PFCL was used after one or two vitrectomy surgeries. Submacular PFCL was found in 12 (0.98%) eyes. There was no statistical difference between perfluorodecalin and perfluoro-n-octane as regards retention rate. The possible risk factors associated with submacular retention of PFCL phenomenon are the presence of post-equatorial retinal breaks in six patients (50%) and large retinal break (2-3 clock hours) in three patients (25%).

CONCLUSIONS

Submacular PFC migration occurs in nearly 1% of cases when it is used. It occurred in cases of RRD with post-equatorial or large breaks. Using perfluoro-n-octane or perfluorodecalin had no statistical difference. Different behaviors of submacular PFC droplets occurred. Some cases of small submacular PFC droplets maintained fair VA for a long period.

Factors Predictive of Subretinal Fluid Resolution in Coats Disease: Analysis of 177 Eyes in 177 Patients at a Single Center

PURPOSE

The aim of this study was to investigate factors predictive of subretinal fluid (SRF) resolution in Coats disease.

DESIGN

Retrospective cohort study.

METHODS

Institutional review board-approved review of patients diagnosed with Coats disease demonstrating SRF (stage 3-5) at a single center from November 1973 to July 2018 with comparison of eyes that had resolution of SRF to those in which SRF persisted.

RESULTS

There were 177 cases (154 males, 87%) of Coats disease diagnosed at a mean age of 8 years. After a mean follow-up of 62 months, SRF resolved in 110 (62%) and persisted in 67 (38%) eyes. Comparison (resolved SRF vs persistent SRF) revealed classification as stage 3A [63 (57%) vs 20 (29%)], stage 3B [47 (43%) vs 40 (60%)], or stage 4 [0 (0%) vs 7 (11%)] (P < 0.001). Eyes with resolved SRF presented with fewer clock hours of telangiectasia (mean: 5 vs 7 clock hours, P < 0.001), light bulb aneurysms (mean: 5 vs 7 clock hours, P < 0.001), exudation (mean: 7 vs 10 clock hours, P < 0.001), and extent of SRF (mean: 7 vs 10 clock hours, P < 0.001). Factors predictive of SRF resolution included absence of iris neovascularization on fluorescein angiography [odds ratio 0.05 (95% confidence interval 0.01-0.60), P = 0.02], and less elevated SRF by ultrasonography [odds ratio 0.84 (95% confidence interval 0.76-0.95), P = 0.004). For every 1-mm decrease in SRF, likelihood of SRF resolution increased by 16%.

CONCLUSIONS

Resolution of SRF was achieved in the majority of eyes (62%) with stage 3 to 5 Coats disease. Predictors of SRF resolution included lack of neovascularization on fluorescein angiography and less elevation of SRF by ultrasonography.

Three-dimensional analysis of morphologic changes and visual outcomes in diabetic macular edema

PURPOSE

To investigate the association of retinal fluid volume with the visual and anatomic outcomes of patients with diabetic macular edema (DME) after treatment with bevacizumab.

STUDY DESIGN

Retrospective observational study.

METHODS

We retrospectively analyzed 65 eyes of 58 DME patients treated with bevacizumab. The volumes of the inner intraretinal fluid (IRF) in the inner nuclear layer (INL), outer IRF in the outer plexiform layer (OPL)/outer nuclear layer (ONL), and subretinal fluid (SRF) were calculated. The correlations between the baseline fluid volumes and best-corrected visual acuity (BCVA), area of disorganization of the retinal inner layers (DRIL), disrupted external limiting membrane (ELM), and disrupted ellipsoid zone (EZ) at 12 months after treatment were assessed.

RESULTS

The baseline volume of the inner IRF correlated with poor BCVA at the final visit (r = 0.52, P < .001) whereas the baseline volume of the outer IRF and SRF volume did not show a significant correlation with BCVA at the final visit (P = .07 and P = .61, respectively). The improvement in BCVA correlated with the reduction in the baseline volume of the inner IRF and outer IRF (r = 0.25, P = .04 and r = 0.36, P = .003), but not with the SRF volume (P = .59). The baseline volume of the inner IRF correlated positively with the area of DRIL and the disrupted ELM at the final visit (r = 0.56, P < .001 and r = 0.25, P = .04, respectively). Such relationship remained in each quadrant of the macula (P < .005 for all quadrants).

CONCLUSION

The baseline volume of the inner IRF correlated with a poor visual outcome, an increased DRIL area, and a more disrupted ELM area in patients with DME after treatment with bevacizumab.

Statistically Evaluating Detection Capabilities of Sub Retinal Fluid Detection Methodology in Optical Coherence Tomographic Images

Significance of therapeutic imaging in medical image based diagnostics and treatment, and there in practice implementation in image investigation is undebatable. The cyst (fluid accumulation) development just above the Retinal Pigment Epithelium (RPE) Layer is known as Sub-Retinal Fluid (SRF). This distinct anatomical localization of SRF, differs it from normal cysts. Progressive increase of SRF, adversely affects the visual acuity. Improvement in visual acuity by reduction in SRF is a desired outcome of currently available treatments. In this paper, we assess and evaluate the detection capabilities of the Sub Retinal Fluid (SRF) detection technique through the statistical analytical methodologies. Receiver Operating Characteristic (ROC) curve analysis, projects the system's ability to discriminate between the images with and without pathology (in this research pathology is SRF). Overall Sensitivity and Specificity of the SRF detection technique is 96.2% and 92.9% respectively.

Retinal volume change is a reliable OCT biomarker for disease activity in neovascular AMD

PURPOSE

Current algorithms for automated computer interpretation of optical coherence tomography (OCT) imaging of patients suffering from neovascular age-related macular degeneration (AMD) mostly rely on fluid detection. However, fluid detection itself and correct interpretation of the fluid currently limits diagnostic accuracy. We therefore performed a detailed analysis of the requirements that would have to be met for fluid detection approaches. We further investigated if monitoring retinal volume would be a viable alternative to detect disease activity.

METHODS

Retrospective analysis and manual grading of 764 OCT volume scans of 44 patients with exudative AMD treated with intravitreal anti-VEGF injections at a pro-re-nata (PRN) treatment regimen for at least 24 months.

RESULTS

Detection of subretinal fluid (SRF) or intraretinal fluid (IRF) alone is not sufficient for disease detection. A combination of SRF and IRF can detect disease activity with a sensitivity of 98.6% and a specificity of 82%. With further characterization of IRF into exudative and degenerative cysts, specificity can be increased to 100%. However, correct characterization is currently not achieved by published fluid detection approaches. Change of macular retinal volume (MRV) can depict disease activity with sensitivity of 88.4% and specificity of 89.6%. Combination with the detection of SRF can further improve diagnostic accuracy to a specificity of 93.3% and sensitivity of 93.9% without relying on IRF or IRF characterization.

CONCLUSION

Fluid detection without further characterization is not sufficient for AMD monitoring. Either further distinction between exudative and degenerative cysts is necessary, or other activity markers have to be taken into account. MRV offers good potential to fill this diagnostic gap and might become an important monitoring marker.

Fundus Autofluorescence Patterns of Submacular Fluid Resolution Following Repair of Macula-Involving Rhegmatogenous Retinal Detachments

BACKGROUND AND OBJECTIVE

Submacular fluid (SMF) can persist for months to years following rhegmatogenous retinal detachment (RD) repair. The authors' objective was to describe fundus autofluorescence (FAF) and optical coherence tomography (OCT) findings associated with the absorption of persistent submacular fluid (SMF) following RD repair.

PATIENTS AND METHODS

Retrospective review of clinical data and FAF and OCT imaging from sequential postoperative visits in a cohort of patients with persistent SMF following RD repair.

RESULTS

In 11 of 13 eyes with persistent SMF, patches of decreased FAF signal corresponded to SMF on OCT. In eight eyes, there was a hypo- to hyperautofluorescence transition at the time of SMF resolution. These areas of increased FAF signal correlated with inner segment/outer segment (IS/OS) junction loss on OCT.

CONCLUSION

FAF imaging can be informative when following SMF after RD repair; a hypo- to hyper-FAF signal transition correlates with SMF resolution and photoreceptor loss. [Ophthalmic Surg Lasers Imaging Retina. 2016;47:1020-1029.].

Automatic Subretinal Fluid Segmentation of Retinal SD-OCT Images With Neurosensory Retinal Detachment Guided by Enface Fundus Imaging

OBJECTIVE

Accurate segmentation of neurosensory retinal detachment (NRD) associated subretinal fluid in spectral domain optical coherence tomography (SD-OCT) is vital for the assessment of central serous chorioretinopathy (CSC). A novel two-stage segmentation algorithm was proposed, guided by Enface fundus imaging.

METHODS

In the first stage, Enface fundus image was segmented using thickness map prior to detecting the fluid-associated abnormalities with diffuse boundaries. In the second stage, the locations of the abnormalities were used to restrict the spatial extent of the fluid region, and a fuzzy level set method with a spatial smoothness constraint was applied to subretinal fluid segmentation in the SD-OCT scans.

RESULTS

Experimental results from 31 retinal SD-OCT volumes with CSC demonstrate that our method can achieve a true positive volume fraction (TPVF), false positive volume fraction (FPVF), and positive predicative value (PPV) of 94.3%, 0.97%, and 93.6%, respectively, for NRD regions. Our approach can also discriminate NRD-associated subretinal fluid from subretinal pigment epithelium fluid associated with pigment epithelial detachment with a TPVF, FPVF, and PPV of 93.8%, 0.40%, and 90.5%, respectively.

CONCLUSION

We report a fully automatic method for the segmentation of subretinal fluid.

SIGNIFICANCE

Our method shows the potential to improve clinical therapy for CSC.

The use of CEUS in the diagnosis of retinal/choroidal detachment and associated intraocular masses - preliminary investigation in patients with equivocal findings at conventional ultrasound

PURPOSE

To investigate whether contrast-enhanced ultrasound (CEUS) may help to diagnose retinal/choroidal detachment and may help to differentiate intraocular lumps in cases with equivocal features on conventional grayscale and Doppler modes.

MATERIALS AND METHODS

The institutional review board approved this retrospective study. The need for informed consent was waived. A computerized data search was performed in the database of our institution for patients with vitreous hemorrhage who underwent CEUS of the eye to assess retinal/choroidal detachment and/or associated masses. This process yielded a total of 31 patients (18 men, 13 women, age range: 39 - 88 years) in whom CEUS was performed because the findings on conventional grayscale and Doppler modes were equivocal. CEUS was performed using low acoustic power contrast-specific modes. A 2.4 - 4.8  mL bolus of SonoVue was injected, followed by a saline flush. All examinations were digitally recorded for retrospective analysis. Confirmation of CEUS findings was obtained at surgery (n = 20) or with binocular indirect fundoscopy performed after clearance of the ocular media (n = 11). Two readers with different levels of ultrasound experience independently reviewed the imaging features. A five-degree scale ranging from definitely absent (score 1) to definitely present (score 5) was used to assess the presence or absence of retinal/choroidal detachment on conventional ultrasound modes alone and with the addition of CEUS. ROC curve analysis was performed to assess the diagnostic accuracy of both methods. The inter-reader agreement was also evaluated. In patients with associated intraocular lumps, conventional Doppler modes and CEUS were used to differentiate non-tumor masses from tumor masses.

RESULTS

According to the reference standard, 13 patients had retinal detachment, 4 had choroidal detachment, and 3 had both retinal and choroidal detachment. There were 8 associated intraocular lumps (4 subretinal hemorrhages, 3 malignant melanomas, 1 metastasis). The inter-reader agreement was good (K = 0.644) and very good (K = 0.833) for conventional modes and CEUS, respectively. The diagnostic performance of CEUS was high for both readers (area ± standard error under the ROC curve: 0.966 ± 0.031 and 0.900 ± 0.055 for readers 1 and 2, respectively). There were 2 false-positive results and 1 false-negative result in patients with proliferative diabetic retinopathy. CEUS was effective in differentiating subretinal hemorrhage from hypovascular tumors.

CONCLUSION

CEUS can be used as a problem-solving technique when conventional ultrasound modes are not diagnostic for retinal/choroidal detachment and when intraocular lumps cannot be characterized as tumor or non-tumor masses on conventional modes. The evaluation of patients with proliferative diabetic retinopathy, however, may be problematic.

[Influence of genetic mutations on clinical presentation of subretinal neovascularization. Report 1: The impact of CFH and IL-8 genes polymorphism]

Genetic analysis was performed in patients with subretinal neovascularization (CNV). The results showed significant association of CFH (compliment factor H) gene polymorphism with increase (rs1061170, rs514943 and rs380390) or decrease (rs529825, rs7524776, rs1831281, rs2274700, rs1576340, rs12144939, rs7540032) of CNV development risk. The incidence of IL-8 gene mutation was significantly (p = 0.008) higher in patients after chorioretinitis. Apparently -125 > A polymorphism in patients with chorioretinitis increases risk of CNV development, thus promoting raise of proangiogenic factors concentration in eyes with inflammatory background. The clinical presentation in patients with AMD and myopic disease associated with (-125) A mutation of promoter region of IL-8 gene was similar to that of patients with chorioretinitis. The features are the following: focal pattern, no drusen and RPE detachment, predominantly classic form of CNV (without occult pattern), formation of well-organized newly developed vessels.