Deep learning for automatic prediction of early activation of treatment naïve non-exudative MNVs in AMD.

BACKGROUND: Around 30% of non-exudative macular neovascularizations(NE-MNVs) exudate within 2 years from diagnosis in patients with age-related macular degeneration(AMD).The aim of the study is to develop a deep learning classifier based on optical coherence tomography(OCT) and OCT angiography(OCTA) to identify NE-MNVs at risk of exudation. METHODS: AMD patients showing OCTA and fluorescein angiography (FA) documented NE-MNV with a 2-years minimum imaging follow-up were retrospectively selected. Patients showing OCT B-scan-documented MNV exudation within the first 2 years formed the EX-GROUP while the others formed QU-GROUP.ResNet-101, Inception-ResNet-v2 and DenseNet-201 were independently trained on OCTA and OCT B-scan images. Combinations of the 6 models were evaluated with major and soft voting techniques. RESULTS: Eighty-nine (89) eyes of 89 patients with a follow-up of 5.7 ± 1.5 years were recruited(35 EX GROUP and 54 QU GROUP). Inception-ResNet-v2 was the best performing among the 3 single convolutional neural networks(CNNs).The major voting model resulting from the association of the 3 different CNNs resulted in improvement of performance both for OCTA and OCT B-scan (both significantly higher than human graders' performance). Soft voting model resulting from the combination of OCTA and OCT B-scan based major voting models showed a testing accuracy of 94.4%. Peripheral arcades and large vessels on OCTA enface imaging were more prevalent in QU GROUP. CONCLUSIONS: Artificial intelligence shows high performances in identifications of NE-MNVs at risk for exudation within the first 2 years of follow up, allowing better customization of follow up timing and avoiding treatment delay. Better results are obtained with the combination of OCTA and OCT B-scan image analysis.

Long-term predictors of anti-VEGF treatment response in patients with neovascularization secondary to CSCR: a longitudinal study.

PURPOSE: To identify the baseline predictors of anti-VEGF treatment response at 3 years in patients affected by choroidal neovascularization (CNV) secondary to central serous chorioretinopathy (CSCR). METHODS: In this retrospective longitudinal study, medical records of patients diagnosed with CNV secondary to CSCR and treated using anti-VEGF injections between April 2015 and May 2020 were reviewed. The potential qualitative and quantitative predictors of treatment response were identified or measured based on the multimodal imaging examination available for each patient at the baseline, including structural OCT, fluorescein angiography (FA), indocyanine green angiography (ICGA), and OCT-angiography (OCT-A). Univariate and multivariate analyses were performed. RESULTS: Twenty-nine eyes from 29 patients affected by CNV complicating CSCR were included in the study. At the end of the 3-year follow-up, the mean BCVA was 20/50 Snellen equivalent (0.38 ± 0.36 LogMAR), and no significant difference with baseline BCVA (0.37 ± 0.29 LogMAR) was found (p = 0.9). Twenty out of 29 eyes (69%) had active lesions at the end of the follow-up. At multivariate analysis, none of the included features was independently associated with the 3-year BCVA outcome. Pigment epithelium detachment (PED) height (ß = 0.017, p = 0.028) and outer limiting membrane (OLM) preservation at the fovea (ß = -5.637, p = 0.026) were independently associated with the CNV activity at 3 years. CONCLUSION: PED height and OLM obliteration at the fovea might be considered baseline predictors of lesion activity at 3-year follow-up in patients with CNV secondary to CSCR treated with anti-VEGF therapy.

Novel IMPG2 variant causing adult macular vitelliform dystrophy: A case report.

INTRODUCTION: Adult-onset vitelliform macular dystrophy (AVMD) is an inherited maculopathy characterized by metamorphopsias and decrease in visual acuity occurring between the fourth and the sixth decade. It is characterized by an 'egg yolk' macular lesion eventually evolving towards foveal atrophy and fibrosis. It is usually an autosomal dominant inherited disorder with variable penetrance, mainly related to variants in BEST1, PRPH2, IMPG1, and IMPG2 genes. CASE DESCRIPTION: A 47-year-old woman complaining of "wavy" vision was referred to our clinic. Her past medical history and reported family history did not reveal any ocular disease. Complete ophthalmological evaluation was performed. Funduscopic examination and multimodal imaging revealed a round vitelliform lesion in both eyes, leading to a diagnosis of AVMD. Genetic analysis revealed a novel, likely pathogenetic, heterozygous c.478G > T (p.Glu160Ter), (NM_016247) variant in the IMPG2 gene. DISCUSSION: Our patient exhibits a novel pathogenetic variant in a gene associated with AVMD. Heterozygous variants in the IMPG2 gene have been reported in multiple individuals with vitelliform macular dystrophy, with an autosomal dominant mode of inheritance. Genetic screening is essential to characterize patients, to predict vision loss in patients with a positive family history and to characterize eligible patients for new potential emerging therapies. Genotype-phenotype correlation studies are needed to have a clearer picture of pathogenetic mechanisms. Our study characterizes the phenotype related to a novel IMPG2 pathogenic variant through multimodal imaging.

Subretinal Pseudocysts: A Comprehensive Analysis of this Novel OCT Finding.

INTRODUCTION: In current clinical practice, several optical coherence tomography (OCT) biomarkers have been proposed for the assessment of severity and prognosis of different retinal diseases. Subretinal pseudocysts are subretinal cystoid spaces with hyperreflective borders and only a few single cases have been reported thus far. The aim of the study was to characterize and investigate this novel OCT finding, exploring its clinical outcome. METHODS: Patients were evaluated retrospectively across different centers. The inclusion criterion was the presence of subretinal cystoid space on OCT scans, regardless of concurrent retinal diseases. Baseline examination was set as the first time the subretinal pseudocyst was identified by OCT. Medical and ophthalmological histories were collected at baseline. OCT and OCT-angiography were performed at baseline and at each follow-up examination. RESULTS: Twenty-eight eyes were included in the study and 31 subretinal pseudocysts were characterized. Out of 28 eyes, 16 were diagnosed with neovascular age-related macular degeneration (AMD), 7 with central serous chorioretinopathy, 4 with diabetic retinopathy, and 1 with angioid streaks. Subretinal and intraretinal fluid were present in 25 and 13 eyes, respectively. Mean distance of the subretinal pseudocyst from the fovea was 686 µm. The diameter of the pseudocyst was positively associated with the height of the subretinal fluid (r = 0.46; p = 0.018) and central macular thickness (r = 0.612; p = 0.001). At follow-up, subretinal pseudocysts disappeared in most of the reimaged eyes (16 out of 17). Of these, two patients presented retinal atrophy at baseline examination and eight patients (47%) developed retinal atrophy at follow-up. Conversely, seven eyes (41%) did not develop retinal atrophy. CONCLUSION: Subretinal pseudocysts are precarious OCT findings, usually disclosed in a context of subretinal fluid, and are probably transient alterations within the photoreceptor outer segments and retinal pigment epithelium (RPE) layer. Despite their nature, subretinal pseudocysts have been associated with photoreceptor loss and incomplete RPE definition.

Deep learning to distinguish Best vitelliform macular dystrophy (BVMD) from adult-onset vitelliform macular degeneration (AVMD).

Initial stages of Best vitelliform macular dystrophy (BVMD) and adult vitelliform macular dystrophy (AVMD) harbor similar blue autofluorescence (BAF) and optical coherence tomography (OCT) features. Nevertheless, BVMD is characterized by a worse final stage visual acuity (VA) and an earlier onset of critical VA loss. Currently, differential diagnosis requires an invasive and time-consuming process including genetic testing, electrooculography (EOG), full field electroretinogram (ERG), and visual field testing. The aim of our study was to automatically classify OCT and BAF images from stage II BVMD and AVMD eyes using a deep learning algorithm and to identify an image processing method to facilitate human-based clinical diagnosis based on non-invasive tests like BAF and OCT without the use of machine-learning technology. After the application of a customized image processing method, OCT images were characterized by a dark appearance of the vitelliform deposit in the case of BVMD and a lighter inhomogeneous appearance in the case of AVMD. By contrast, a customized method for processing of BAF images revealed that BVMD and AVMD were characterized respectively by the presence or absence of a hypo-autofluorescent region of retina encircling the central hyperautofluorescent foveal lesion. The human-based evaluation of both BAF and OCT images showed significantly higher correspondence to ground truth reference when performed on processed images. The deep learning classifiers based on BAF and OCT images showed around 90% accuracy of classification with both processed and unprocessed images, which was significantly higher than human performance on both processed and unprocessed images. The ability to differentiate between the two entities without recurring to invasive and expensive tests may offer a valuable clinical tool in the management of the two diseases.

Quality of life assessment in amyloid transthyretin (ATTR) amyloidosis.

BACKGROUND: Amyloid transthyretin (ATTR) amyloidosis is caused by the systemic deposition of transthyretin molecules, either normal (wild-type ATTR, ATTRwt) or mutated (variant ATTR, ATTRv). ATTR amyloidosis is a disease with a severe impact on patients' quality of life (QoL). Nonetheless, limited attention has been paid to QoL so far, and no specific tools for QoL assessment in ATTR amyloidosis currently exist. QoL can be evaluated through patient-reported outcome measures (PROMs), which are completed by patients, or through scales, which are compiled by clinicians. The scales investigate QoL either directly or indirectly, i.e., by assessing the degree of functional impairment and limitations imposed by the disease. DESIGN: Search for the measures of QoL evaluated in phase 2 and phase 3 clinical trials on ATTR amyloidosis. RESULTS: Clinical trials on ATTR amyloidosis have used measures of general health status, such as the Short Form 36 Health Survey (SF-36), or tools developed in other disease settings such as the Kansas City Cardiomyopathy Questionnaire (KCCQ) or adaptations of other scales such as the modified Neuropathy Impairment Score +7 (mNIS+7). CONCLUSIONS: Scales or PROMs for ATTR amyloidosis would be useful to better characterize newly diagnosed patients and to assess disease progression and response to treatment. The ongoing ITALY (Impact of Transthyretin Amyloidosis on Life qualitY) study aims to develop and validate 2 PROMs encompassing the whole phenotypic spectrum of ATTRwt and ATTRv amyloidosis, that might be helpful for patient management and may serve as surrogate endpoints for clinical trials.