Search

The purpose of the study was to detect Hard Exudates (HE) and classify Disorganization of Retinal Inner Layers (DRIL) implementing a Deep Learning (DL) system on optical coherence tomography (OCT) images of eyes with diabetic macular edema (DME). We collected a dataset composed of 442 OCT images on which we annotated 6847 HE and the presence of DRIL. A complex operational pipeline was defined to implement data cleaning and image transformations, and train two DL models. The state-of-the-art neural network architectures (Yolov7, ConvNeXt, RegNetX) and advanced techniques were exploited to aggregate the results (Ensemble learning, Edge detection) and obtain a final model. The DL approach reached good performance in detecting HE and classifying DRIL. Regarding HE detection the model got an AP@0.5 score equal to 34.4% with Precision of 48.7% and Recall of 43.1%; while for DRIL classification an Accuracy of 91.1% with Sensitivity and Specificity both of 91.1% and AUC and AUPR values equal to 91% were obtained. The P-value was lower than 0.05 and the Kappa coefficient was 0.82. The DL models proved to be able to identify HE and DRIL in eyes with DME with a very good accuracy and all the metrics calculated confirmed the system performance. Our DL approach demonstrated to be a good candidate as a supporting tool for ophthalmologists in OCT images analysis. [ABSTRACT FROM AUTHOR]

Background: To report a case of cataract surgery in unintentional Ozurdex (Allergan, Inc., Irvine, California, USA) injection into the lens. Case presentation: A 82-years old man reporting decreased visual acuity in his right eye came to our Ophthalmology service. Due to the clinical history, and on the basis of ophthalmoscopic and imaging examinations diabetic macular edema was diagnosed. Thus, intravitreal dexamethasone implant was scheduled and therefore performed. The following day Ozurdex appeared to be located into the lens. After careful evaluation and strict follow up examinations, due to the risks associated with the presence of the implant into the lens, phacoemulsification with Ozurdex removal and intraocular lens (IOL) implantation was scheduled and performed. Conclusions: In this case report we reported the surgical management of accidental into-the lens dexamethasone implant carefully taking into account the dexamethasone pharmacokinetic. [ABSTRACT FROM AUTHOR]

Introduction: The purpose of this study was to investigate long-term outcomes of intravitreal injections (IVI) of antivascular endothelial growth factor (VEGF) in neovascular age-related macular degeneration (nAMD) with type 3 macular neovascularization (MNV). Methods: This retrospective study included 19 eyes of 17 patients with nAMD and type 3 MNV treated with anti-VEGF IVI with a loading dose and a PRN regimen. Best corrected visual acuity (BCVA), central macular thickness (CMT), presence of macular intraretinal fluid (IRF) and subretinal fluid (SRF), flow area (FA), subfoveal choroidal thickness (CT), and macular atrophy (MA) were assessed at baseline (T0) and during follow-up (T1, post-loading phase; T2, 1 year; T3, 2 years; T4 >2 years). The correlations between MA at the last follow-up and standard deviation (SD) values of CMT and CT during follow-up were assessed. The influence of the number of injections on the change in MA over time was also analyzed. MA differences at T4 were assessed for pseudodrusen presence. Results: BCVA improved significantly during follow-up (p = 0.013) particularly increasing from baseline to post-loading phase and then did not modify significantly thereafter. CMT significantly reduced from T0 to T1 and remained stable during follow-up (p = <0.001). MNV flow area showed a trend toward an increase in the post-loading phase that was not statistically significant (p = 0.082) and CT decreased significantly during follow-up (p < 0.001). MA changed significantly during follow-up (p < 0.001) with a significant increase from T0 to T3 and from T0 to T4 (p < 0.010). A Cochran-Armitage test for trend showed a significant reduction (p = 0.001) of macular IRF and SRF during follow-up. MA at T4 showed a significant positive correlation with SD (standard deviation) values of CMT (p = 0.040) and CT (p = 0.020). Indeed, the number of injections did not influence the change over time of MA (p = 0.709). MA at T4 was not statistically significantly different between patients with pseudodrusen at baseline (p = 0.497). Conclusions: Intravitreal anti-VEGF injections with PRN regimen in MNV type 3 showed functional and anatomical benefits. Variations of retinal thickness and choroidal thickness during treatment were related to MA modification over time. [ABSTRACT FROM AUTHOR]