Search

Abstract Background Diabetic macular edema (DME) is the main cause of visual loss in individuals with diabetic retinopathy (DR). This study aims to investigate the effects of central macular intraretinal cysts on the underlying outer retinal layer (ORL) in patients with diabetic macular edema (DME). Methods In this retrospective and cross-sectional study, diabetic patients with or without DR were categorized into three groups: without DME (group 1), with DME but without any cyst featuring a plateau in the lower region (group 2), and patients with cyst featuring an inferior cyst plateau (group 3), defined as a flat conformation at its posterior aspect. Variables such as central macular intraretinal cyst height, inferior cyst plateau, and ORL thickness were measured, and ellipsoid zone (EZ) disruption was assessed via Spectral-domain optical coherence tomography (SD-OCT) and compared between groups. Correlations between OCT-measured variables and best-corrected visual acuity (BCVA) were investigated. Results A total of 164 eyes were included: 48 in group 1, 47 in group 2 and 69 in group 3. Compared with Groups 1 and 2, Group 3 presented a greater intraretinal cyst height (p

Abstract Credit scoring models are critical for financial institutions to assess borrower risk and maintain profitability. Although machine learning models have improved credit scoring accuracy, imbalanced class distributions remain a major challenge. The widely used Synthetic Minority Oversampling TEchnique (SMOTE) struggles with high-dimensional, non-linear data and may introduce noise through class overlap. Generative Adversarial Networks (GANs) have emerged as an alternative, offering the ability to model complex data distributions. Conditional Wasserstein GANs (cWGANs) have shown promise in handling both numerical and categorical features in credit scoring datasets. However, research on extracting latent features from non-linear data and improving model explainability remains limited. To address these challenges, this paper introduces the Non-parametric Oversampling Technique for Explainable credit scoring (NOTE). The NOTE offers a unified approach that integrates a Non-parametric Stacked Autoencoder (NSA) for capturing non-linear latent features, cWGAN for oversampling the minority class, and a classification process designed to enhance explainability. The experimental results demonstrate that NOTE surpasses state-of-the-art oversampling techniques by improving classification accuracy and model stability, particularly in non-linear and imbalanced credit scoring datasets, while also enhancing the explainability of the results.

Abstract Purpose To verify the correlation between the full-macular and the ganglion cell complex (GCC) thickness measurements and retinal sensitivity (RS) assessed by microperimetry (MP) 6 months after surgical peeling for idiopathic epiretinal membrane (ERM). Methods Forty-three were submitted to pars-plana posterior vitrectomy (PPV) with concomitant peeling of internal limiting membrane (ILM) for idiopathic ERM treatment. Best-corrected visual acuity (BCVA) and 3D volumetric high-definition optical coherence tomography (OCT) imaging were preoperatively acquired. Six months after the surgery, BCVA, OCT imaging, and RS measured by MP were assessed. For the OCT parameters, we analyzed both the full-macular and the ganglion cell layer complex (GCC) thicknesses. The MP parameters tested were 44 points covering 20 central degrees (6 mm), with direct correspondence with the nine sectors of the OCT-ETDRS map. This approach enables the direct topographic correlation between the structure and functional measurements. The OCT and MP exam measurements were also performed in 43 eyes of age-matched healthy controls. Correlations between BCVA, RS, and OCT parameters were examined. Results All patients exhibited a substantial improvement in visual acuity following surgery. The RS parameters were significantly lower in patients compared to the controls. The full-macular thickness measurements were thicker than controls preoperatively and significantly reduced postoperatively; however, remaining significantly higher than controls, in the 4 inner sectors, at the fovea and for the average macular thickness. Preoperative GCC measurements were higher than those in controls. There was a significant reduction in GCC thickness in all sectors postoperatively, especially in the outer sectors, as well as in the average macular thickness. A positive correlation was found between full-macular and GCC thickness and RS postoperatively in several sectors. Conclusions Our results demonstrate that ERM peeling can improve visual acuity in the postoperative period. However, RS may not fully restore, remaining significantly lower when compared to the controls. Both full-macular and GCC thickness measurements were reduced 6 months after surgery. However, significant thinning of the GCC thickness was observed when compared to the normal control eyes, indicating the occurrence of some degree of ganglion cell layer atrophy. We have demonstrated significant correlations among various OCT thickness parameters, particularly for GCC measurements. We believe that GCC integrity may play an important role in visual function after ERM surgery, and that MP may help better understand the correlations between structural and functional findings following ERM surgery.

Abstract Purpose To assess the relationship between macular choroidal thickness (CT) measurements and retinal sensitivity (RS) in eyes with myopia and different stages of myopic maculopathy. Methods A masked, cross-sectional, and consecutive study involving patients with emmetropia/myopia (control group) and high myopia (HM) eyes. Automated choroidal thickness (CT) and manual outer retinal layer (ORL) thickness were acquired using swept-source optical coherence tomography, while retinal sensitivity (RS) assessed by microperimetry (MP3) in all regions of the macular Early Treatment Diabetic Retinopathy Study (ETDRS) grid. Comparisons were made between groups, and correlations were performed among these measurements, demographic and ocular parameters and myopic maculopathy classification. Results A total of 37 (74 eyes) patients were included in the study. The mean age was 39 ± 13 years, and 28 patients (76%) were female. HM eyes exhibited inferior best-corrected visual acuity and a more advanced myopic maculopathy classification compared to the control group. The mean macular CT were 255 and 179 μm in the control and HM eyes (P

Abstract Background To verify the correlation between retinal sensitivity (RS) assessed by the microperimetry (MP) and optical coherence tomography (OCT) parameters measured in eyes submitted to pars-plana vitrectomy (PPV) for idiopathic epiretinal membrane (ERM) treatment. Methods 43 patients underwent PPV. Best-corrected visual acuity (BCVA) and OCT imaging were acquired preoperatively and 6 months after surgery. The RS values were recorded 6 months after the surgery. Total macular thickness (TMT) measurements and OCT-evaluated structural findings were also analyzed. The MP examination tested 44 points, with direct topographic correspondence with the OCT-ETDRS map. Correlations between BCVA, RS, and OCT parameters were assessed. Results TMT measurements in patients were significantly thicker preoperatively and reduced after surgery. All patients demonstrated BCVA improvements after surgery. The RS parameters after surgery were significantly lower in patients. For OCT structural analyses, patients with lower RS at the fovea correlated with the preexisting disorganization of retinal inner layers (DRIL). In addition, lower RS values were associated with DRIL, outer retinal changes (ORC), and intraretinal microcysts after surgery. Conclusions The RS values after surgery were significantly lower when compared to controls. The DRIL presence before and after surgery, and microcysts and ORC after surgery were related to worse visual outcomes.

Abstract Purpose Tuberous Sclerosis (TS) is a rare, multisystem genetic disease caused by mutations in the TSC1 and TSC2 genes, leading to abnormalities in cell differentiation and proliferation. This study aimed to evaluate the neural integrity of individuals with TS by using Optical Coherence Tomography (OCT) to examine the peripapillary retinal nerve fiber layer (RNFL) thickness and the macular thickness in patients with TS and to compare with healthy controls. Methods Peripapillary and macular OCT scans (Optopol Revo NX SD OCT) were performed on 41 eyes from 22 TS patients, divided into two groups based on the presence of retinal hamartomas, and compared to 20 eyes from a control group. The average peripapillary RNFL thickness was measured for each quadrant. The macular total thickness and ganglion cell layer (GCL) + inner plexiform layer (IPL) thickness were measured based on the Early Treatment Diabetic Retinopathy Study (ETDRS) map. All measurements were then compared between the groups and controls. Results The TS group showed significantly reduced RNFL thickness and macular thickness when compared to the control group. Specifically, patients with retinal hamartomas exhibited an even more pronounced thinning of both RNFL and macular thickness. Conclusions These findings suggest that TS patients undergo significant changes in retinal neurodevelopment and experience axonal loss. This finding may have significant prognostic utility regarding central nervous system degeneration in TS, particularly among patients with retinal hamartomas. OCT may serve as a valuable tool for assessing axonal structural abnormalities in TS patients. Trial Registration Number: Not applicable.

Web data extraction is concerned, among other things, with routine data accessing and downloading from continuously-updated dynamic Web pages. There is a relevant trade-off between the rate at which the external Web sites are accessed and the computational burden on the accessing client. We address the problem by proposing a predictive model, typical of the Operating Systems literature, of the rate-of-update of each Web source. The presented model has been implemented into a new version of the Dynamo project: a middleware that assists in generating informative RSS feeds out of traditional HTML Web sites. To be effective, i.e., make RSS feeds be timely and informative and to be scalable, Dynamo needs a careful tuning and customization of its polling policies, which are described in detail.

All-trans retinoic acid (ATRA) is a vitamin A derivative which can increase intracranial pressure, causing visual loss and papilledema. Those patients should be treated similarly to others patients with idiopathic intracranial hypertension. We described a case of a 32-year-old woman presenting with severe visual loss and intracranial hypertension induced by ATRA for acute promyelocytic leukemia, which was treated clinically and with optic nerve sheath fenestration. Patients receiving ATRA therapy should be monitored to neurological and ophthalmic signs and symptoms of intracranial hypertension.

Abstract Background The present case aims to describe a previously healthy man who presented multiple attacks of transient monocular visual loss after Pfizer-BioNTech COVID-19 vaccination and to discuss the possible mechanisms related to occurrence of this condition. Case presentation We report a case of multiple attacks of transient monocular visual loss in a previously healthy middle-aged man two weeks after Pfizer-BioNTech COVID-19 vaccination. TVL attacks were described as sudden and painless complete visual loss, lasting about one minute, followed by a full recovery. He presented several non-simultaneous attacks in both eyes, 16 in the right eye, and 2 in the left eye on the same day, fifteen days after receiving the second dose of the Pfizer-BioNTech COVID-19 vaccine. The brain’s magnetic resonance angiography, echocardiogram, and doppler ultrasound imaging of the carotid and vertebral arteries were non-revealing. The complete blood exam revealed a slightly elevated C-reactive protein test. We assessed fundus examination during the transient visual loss attack and revealed diffuse vascular narrowing for both arterial and venous branches, notably in the emergence of the optic disc in right eye. In addition, the circumpapillary optical coherence tomography angiography (OCTA) vessel density map was reduced. Oral verapamil hydrochloride 60 mg twice daily was initiated, and the attacks of transient visual loss improved after two days. Conclusions To date, and the best of our knowledge, this is the first case report of multiple transient monocular visual loss attacks due to retinal vasospasm in a previously healthy middle-aged man documented by fundus retinography and OCTA. We discuss in this article the possible association of retinal vasospasm and Pfizer-BioNTech COVID-19 vaccination, probably related to vaccine-induced inflammation.

ABSTRACT Structural imaging of the brain is the most widely used diagnostic tool for investigating neurodegenerative diseases. More advanced structural imaging techniques have been applied to early or prodromic phases, but they are expensive and not widely available. Therefore, it is highly desirable to search for noninvasive, easily accessible, low-cost clinical biomarkers suitable for large-scale population screening, in order to focus on making diagnoses at the earliest stages of the disease. In this scenario, imaging studies focusing on the structures of the retina have increasingly been used for evaluating neurodegenerative diseases. The retina shares embryological, histological, biochemical, microvascular and neurotransmitter similarities with the cerebral cortex, thus making it a uniquely promising biomarker for neurodegenerative diseases. Optical coherence tomography is a modern noninvasive imaging technique that provides high-resolution two-dimensional cross-sectional images and quantitative reproducible three-dimensional volumetric measurements of the optic nerve head and retina. This technology is widely used in ophthalmology practice for diagnosing and following up several eye diseases, such as glaucoma, diabetic retinopathy and age-related macular degeneration. Its clinical impact on neurodegenerative diseases has raised enormous interest over recent years, as several clinical studies have demonstrated that these diseases give rise to reduced thickness of the inner retinal nerve fiber layer, mainly composed of retinal ganglion cells and their axons. In this review, we aimed to address the clinical utility of optical coherence tomography for diagnosing and evaluating different neurodegenerative diseases, to show the potential of this noninvasive and easily accessible method.

Prediction and quantification of future volatility and returns play an important role in financial modeling, both in portfolio optimisation and risk management. Natural language processing today allows one to process news and social media comments to detect signals of investors' confidence. We have explored the relationship between sentiment extracted from financial news and tweets and FTSE100 movements. We investigated the strength of the correlation between sentiment measures on a given day and market volatility and returns observed the next day. We found that there is evidence of correlation between sentiment and stock market movements. Moreover, the sentiment captured from news headlines could be used as a signal to predict market returns; we also found that the same does not apply for volatility. However, for the sentiment found in Twitter comments we obtained, in a surprising finding, a correlation coefficient of –0.7 (p < 0.05), which indicates a strong negative correlation between negative sentiment captured from the tweets on a given day and the volatility observed the next day. It is important to keep in mind that stock volatility rises greatly when the market collapses but not symmetrically so when it goes up (the so-called leverage effect). We developed an accurate classifier for the prediction of market volatility in response to the arrival of new information by deploying topic modeling, based on Latent Dirichlet Allocation, in order to extract feature vectors from a collection of tweets and financial news. The obtained features were used as additional input to the classifier. Thanks to the combination of sentiment and topic modeling even on modest (essentially personal) architecture our classifier achieved a directional prediction accuracy for volatility of 63%.

Abstract Background To draw comparisons between spectral domain optic coherence tomography (SD-OCT) features of subretinal silicon oil (SO), perfluoro-n-octane (PFO) or C3F8 gas. Methods Cases diagnosed with retained subretinal vitreous substitutes (VS) were retrospectively selected. Demographic data were collected and OCT features were analyzed. Results In the 13 cases with subretinal PFO, hyper-reflectivity under the bubble was noted in 8 eyes (61.5%); choroidal shadow at the borders of the bubble in 11 eyes (84.6%); hyper-reflective halo around the bubble in 5 eyes (38.4%) and a hyper-reflective apical dot in 8 eyes (61.5%).The two cases with multiple PFO bubbles had complete septum dividing the bubbles. The one case with subretinal SO had hyper reflectivity under the bubble; no choroidal shadow at the edge of the bubble; hyper-reflective halo was noted around the bubble and the apical hyper-reflective dot was present; there was no complete septum dividing multiple bubbles. The single case with subretinal C3F8 had some bubbles with totally round base, incomplete septum, hyper reflectivity under the bubble, choroidal shadow at the edge of the bubble, a hyper-reflective halo and an apical dot. Conclusion Different subretinal VS share similar SD-OCT characteristics. Round base bubbles are only observed with subretinal C3F8 gas, while incomplete septum are related to retained subretinal SO or gas.