DAA‐UNet: A Dense Connectivity and Atrous Spatial Pyramid Pooling Attention UNet Model for Retinal Optical Coherence Tomography Fluid Segmentation

To evaluate the response to primary bevacizumab treatment of eyes with age-related macular degeneration (AMD) and choroidal neovascularization (CNV) with a large pigment epithelial detachment (PED) component and to compare the increase in visual acuity and reabsorption of retinal fluid in PED eyes with eyes with CNV in AMD with a minimal to no PED component. We reviewed 43 consecutive eyes with CNV and AMD on primary bevacizumab therapy. There were 13 eyes with a large PED component in AMD with CNV and 30 eyes with a minimal to no PED in CNV. Only patients with no previous treatment for AMD and those started on purely intravitreal bevacizumab treatment were taken in the study. Pigment epithelial detachment size, time to PED collapse, and retinal or subretinal fluid resolution were determined as was Early Treatment Diabetic Retinopathy Study vision. Time to resolution of intraretinal and subretinal fluid was compared between the PED group and the non-PED group using survival analysis. In AMD with CNV eyes having a large PED component, sub- and intraretinal fluid initially resolved faster than the sub-PED fluid (P = 0.03). The subretinal pigment epithelial fluid itself was highly resistant. Visual acuity improvement was similar in both groups. Despite monthly intravitreal bevacizumab injections for neovascular AMD patients with a large component PED, the majority had minimal to no response of the PED. Sub- and intraretinal fluid response was faster in neovascular AMD without large PEDs, but after 7 months, vision change and reabsorption of intra- and subretinal fluid were similar in the two groups. Sub- and intraretinal fluid response did not appear to be related to PED size. Bevacizumab was very effective in reducing more of the sub- and intraretinal fluid than the PED fluid in AMD with CNV.

Predictors of AMD Treatment Response

Intravitreal triamcinolone acetonide (IVTA) has increasingly been applied as treatment for various intraocular neovascular and oedematous diseases. Comparing the various diseases with respect to effect and side-effects of the treatment, the best response in terms of gain in visual acuity (VA) has been achieved for intraretinal oedematous diseases such as diffuse diabetic macular oedema, branch retinal vein occlusion, central retinal vein occlusion and pseudophakic cystoid macular oedema. In eyes with various types of non-infectious uveitis, including acute or chronic sympathetic ophthalmia and Adamantiadis-Behcet's disease, VA increased and the degree of intraocular inflammation decreased. Some studies have suggested that intravitreal triamcinolone may be useful as angiostatic therapy in eyes with iris neovascularization and proliferative ischaemic retinopathies. Intravitreal triamcinolone may possibly be helpful as adjunct therapy for exudative age-related macular degeneration (AMD), particularly in combination with photodynamic therapy. In eyes with chronic, therapy-resistant ocular hypotony, intravitreal triamcinolone can induce an increase in intraocular pressure (IOP) and may stabilize the eye. The complications of intravitreal triamcinolone therapy include: secondary ocular hypertension in about 40% of the eyes injected; medically uncontrollable high IOP leading to antiglaucomatous surgery in about 1-2% of the eyes; posterior subcapsular cataract and nuclear cataract leading to cataract surgery in about 15-20% of elderly patients within 1 year of injection; postoperative infectious endophthalmitis occurring at a rate of about one per 1000; non-infectious endophthalmitis, perhaps due to a reaction to the solvent agent, and pseudo-endophthalmitis with triamcinolone acetonide crystals appearing in the anterior chamber. Intravitreal triamcinolone injection can be combined with other intraocular surgeries, including cataract surgery, particularly in eyes with iris neovascularization. Cataract surgery performed some months after the injection does not show a markedly elevated complication rate. The injection may be repeated if the resultant benefits decrease after the initial IVTA injection. In non-vitrectomized eyes, the duration of the effect and side-effects of a single intravitreal injection of triamcinolone is about 6-9 months for a dosage of about 20 mg, and about 2-4 months for a dosage of 4 mg. So far, it has remained unclear whether the solvent agent should be removed, and if so, how.

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

Resident Perspectives

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.

Optical coherence tomography is a non‐invasive imaging technique that provides micrometer‐resolution images of retinal structures. These images can assist in identifying changes under the retina's surface, such as edema. This study proposes a novel deep learning model AR U‐Net++ for segmenting retinal layers and fluids. The four retinal layers ILM (Internal Limiting Membrane), IPL (Inner Plexiform Layer), RPE (Retinal Pigment Epithelium), BM (Bruch Membrane), and IRF (Intra Retinal Fluid), SRF (Sub Retinal Fluid), and PED (Pigment Epithelial Detachment) are segmented using AR U‐Net++. The proposed architecture AR U‐Net++ achieves better accuracy (99.67%), mean IoU (0.84), and dice coefficient (0.94) than the existing models of U‐Net, AR U‐Net, and AR W‐Net. The novelty of the suggested model AR U‐Net++ is to identify the exact location and depth of the retinal fluid in between the retinal layers and generating reports that aids the clinicians in the diagnosis of Age related Macular Degeneration.

Ranibizumab Combined With Low-Dose Sorafenib for Exudative Age-Related Macular Degeneration

Comparing Retinal Thickness Measurements Using Automated Fast Macular Thickness Map versus Six-Radial Line Scans with Manual Measurements

Introduction: In this retrospective cohort study, we wanted to evaluate the performance and analyze the insights of an artificial intelligence (AI) algorithm in detecting retinal fluid in spectral-domain OCT volume scans from a large cohort of patients with neovascular age-related macular degeneration (AMD) and diabetic macular edema (DME). Methods: A total of 3,981 OCT volumes from 374 patients with AMD and 11,501 OCT volumes from 811 patients with DME were acquired with Heidelberg-Spectralis OCT device (Heidelberg Engineering Inc., Heidelberg, Germany) between 2013 and 2021. Each OCT volume was annotated for the presence or absence of intraretinal fluid (IRF) and subretinal fluid (SRF) by masked reading center graders (ground truth). The performance of an already published AI algorithm to detect IRF and SRF separately, and a combined fluid detector (IRF and/or SRF) of the same OCT volumes was evaluated. An analysis of the sources of disagreement between annotation and prediction and their relationship to central retinal thickness was performed. We computed the mean areas under the curves (AUC) and under the precision-recall curves (AP), accuracy, sensitivity, specificity, and precision. Results: The AUC for IRF was 0.92 and 0.98, for SRF 0.98 and 0.99, in the AMD and DME cohort, respectively. The AP for IRF was 0.89 and 1.00, for SRF 0.97 and 0.93, in the AMD and DME cohort, respectively. The accuracy, specificity, and sensitivity for IRF were 0.87, 0.88, 0.84, and 0.93, 0.95, 0.93, and for SRF 0.93, 0.93, 0.93, and 0.95, 0.95, 0.95 in the AMD and DME cohort, respectively. For detecting any fluid, the AUC was 0.95 and 0.98, and the accuracy, specificity, and sensitivity were 0.89, 0.93, and 0.90 and 0.95, 0.88, and 0.93, in the AMD and DME cohort, respectively. False positives were present when retinal shadow artifacts and strong retinal deformation were present. False negatives were due to small hyporeflective areas in combination with poor image quality. The combined detector correctly predicted more OCT volumes than the single detectors for IRF and SRF, 89.0% versus 81.6% in the AMD and 93.1% versus 88.6% in the DME cohort. Discussion/Conclusion: The AI-based fluid detector achieves high performance for retinal fluid detection in a very large dataset dedicated to AMD and DME. Combining single detectors provides better fluid detection accuracy than considering the single detectors separately. The observed independence of the single detectors ensures that the detectors learned features particular to IRF and SRF.

Purpose: To evaluate the effectiveness of Razumab^® (world's first biosimilar ranibizumab; Intas Pharmaceuticals Ltd., India) in Indian patients with wet age-related macular degeneration (wet AMD), diabetic macular edema (DME) and retinal vein occlusion (RVO). Methods: RE-ENACT, a retrospective, multicenter study, analyzed pooled data of patients with wet AMD, DME, and RVO. Patients who had received ≥3 injections of Razumab^® between January and August 2016, were included. Endpoints were: improvement in best corrected visual acuity (BCVA, measured by logMAR/Snellen's chart), decrease in central macular thickness (CMT, measured by Spectral Domain Optical Coherence Tomography), and proportion of patients with intraretinal fluid (IRF) and subretinal fluid (SRF) at Weeks 4, 8 and 12. Results: Of 561 patients included, 348 (62.04%) were men. Mean ± SE BCVA improved from baseline (0.75 ± 0.01) to Week 4 (0.72 ± 0.01, p = 0.0318), attained significance at Week 8 (0.59 ± 0.01, p ＜ 0.0001), which was maintained at Week 12 (0.49 ± 0.01, p ＜ 0.0001). Mean ± SE CMT significantly (p ＜ 0.0001) decreased from baseline (418.47 ± 4.78μm) to Weeks 4 (407.35 ± 4.65μm), 8 (342.10 ± 3.66μm), and 12 (301.17 ± 2.82μm). Proportion of patients with IRF and SRF significantly (p ＜ 0.0001) decreased from baseline to Weeks 4, 8 and 12 (67.02% vs. 48.48%, 42.60%, and 34.22%, respectively for IRF; and 72.37% vs. 48.48%, 37.97%, 31.37%, respectively for SRF). No new safety concerns with biosimilar ranibizumab were observed. Conclusions: Razumab^® is effective in reducing macular thickness and improving visual acuity in patients with wet agerelated macular degeneration, diabetic macular edema, and retinal vein occlusion in routine clinical practice. Razumab^® demonstrated considerable effectiveness with no new safety concerns.

Diagnosis and monitoring of retina diseases related to pathologies such as accumulated fluid can be performed using optical coherence tomography (OCT). OCT acquires a series of 2D slices (Bscans). This work presents a fully-automated method based on graph shortest path algorithms and convolutional neural network (CNN) to segment and detect three types of fluid including sub-retinal fluid (SRF), intra-retinal fluid (IRF) and pigment epithelium detachment (PED) in OCT Bscans of subjects with age-related macular degeneration (AMD) and retinal vein occlusion (RVO) or diabetic retinopathy. The proposed method achieves an average dice coefficient of 76.44%, 92.25% and 82.14% in Cirrus, Spectralis and Topcon datasets, respectively. The effectiveness of the proposed methods was also demonstrated in segmenting fluid in OCT images from the 2017 Retouch challenge.

PurposeTo explore the relationship between retinal fluid status and best-corrected visual acuity (BCVA) in patients treated with intravitreal aflibercept (IVT-AFL) treat-and-extend (T&E) in the ALTAIR study.MethodsOutcomes were investigated according to overall fluid status at week 16 (predefined) and the relationship between any fluid, intraretinal fluid (IRF), subretinal fluid (SRF), or pigment epithelial detachment with BCVA at baseline, and weeks 16, 52, and 96 (post-hoc). The analyses involved treatment-naïve patients (N = 246) with exudative age-related macular degeneration (AMD), aged ≥ 50 years with BCVA of 73–25 Early Treatment Diabetic Retinopathy Study letters, who participated in the ALTAIR study.ResultsThe mean (standard deviation) change in BCVA from baseline to week 52 was + 10.6 (10.9) and + 6.5 (16.0) letters in patients without and with fluid at week 16, respectively; and to week 96 was + 9.1 (14.3) and + 4.3 (16.1) letters in patients without and with fluid at week 16, respectively. The last injection interval was 16 weeks in 33.6% and 2.0% (week 52), and 62.9% and 17.6% (week 96) of patients without or with fluid at week 16, respectively. At baseline, 35.7% of patients had IRF and 85.2% of patients had SRF, which decreased to 11.8% (IRF) and 31.7% (SRF) of patients, 8.5% (IRF) and 18.7% (SRF), and 6.5% (IRF) and 20.7% (SRF) at weeks 16, 52, and 96, respectively. Presence of IRF at all timepoints was associated with poorer BCVA than if IRF was absent, while the presence of SRF was not associated with poorer BCVA compared with the absence of SRF.ConclusionIVT-AFL T&E dosing was effective at clearing fluid regardless of fluid type in ~ 80% of patients with exudative AMD. Patients without fluid at week 16 had numerically better BCVA than those with fluid at week 16. Over 60% of patients without fluid at week 16 achieved the maximum treatment interval of 16 weeks by study end, compared with < 20% of patients with fluid at week 16. IRF (weeks 16, 52, 96), although evident in a small number of patients, was associated with poorer BCVA, whereas SRF was not.Trial registrationClinicalTrials.gov Identifier: NCT02305238

Loss-balanced parallel decoding network for retinal fluid segmentation in OCT

To measure the concentration of cytokines in the aqueous humour of eyes with exudative age-related macular degeneration (AMD). The clinical interventional study included a study group of 18 patients with exudative AMD and a control group of 20 patients undergoing routine cataract surgery. Age did not vary significantly (p = 0.36) between study group (80.8 ± 6.4 years) and control group (77.0 ± 9.9 years), nor did gender (p = 0.75). During the interventions, aqueous humour samples were obtained, in which the concentration of cytokines was measured using a solid-phase chemiluminescence immunoassay. Macular thickness was measured by optical coherence tomography (OCT). In the study group as compared to the control group, significantly higher concentrations were measured for epithelial growth factor (EGF) (p = 0.017), human growth factor (HGF) (p= 0.048), intercellular adhesion molecule-1 (ICAM1) (p = 0.028), interleukin 12p40 (IL12p40) (p = 0.009), interleukin 1a2 (IL1a2) (p = 0.01), interleukin 3 (IL3) (p = 0.02), interleukin 6 (IL6) (p = 0.006), interleukin 8 (IL8) (p = 0.02), monocyte chemoattractant protein-1 (MCP-1) (p = 0.048), monokine induced by interferon gamma (MIG) (p = 0.016), matrix metalloproteinase 9 (MMP9) (p = 0.004) and plasminogen activator inhibitor 1 (PAI1) (p = 0.006). Macular thickness was significantly associated with the concentrations of EGF (p = 0.001), HGF (p = 0.02), ICAM1 (p = 0.001), interleukin 12p40 (p = 0.006), IL 1a2 (p = 0.002), MIG (p = 0.001), MMP9 (p < 0.001) and PAI1 (p = 0.01). Interleukin 6 and MCP-1 showed significant associations with the height of retinal pigment epithelium detachment. Numerous cytokines are associated with the presence and the amount of exudative AMD.