Search

PurposeThe purpose of this study was to evaluate the ability to align two types of retinal images taken on different platforms; color fundus (CF) photographs and infrared scanning laser ophthalmoscope (IR SLO) images using mathematical warping and artificial intelligence (AI).MethodsWe collected 109 matched pairs of CF and IR SLO images. An AI algorithm utilizing two separate networks was developed. A style transfer network (STN) was used to segment vessel structures. A registration network was used to align the segmented images to each. Neither network used a ground truth dataset. A conventional image warping algorithm was used as a control. Software displayed image pairs as a 5 × 5 checkerboard grid composed of alternating subimages. This technique permitted vessel alignment determination by human observers and 5 masked graders evaluated alignment by the AI and conventional warping in 25 fields for each image.ResultsOur new AI method was superior to conventional warping at generating vessel alignment as judged by masked human graders (P < 0.0001). The average number of good/excellent matches increased from 90.5% to 94.4% with AI method.ConclusionsAI permitted a more accurate overlay of CF and IR SLO images than conventional mathematical warping. This is a first step toward developing an AI that could allow overlay of all types of fundus images by utilizing vascular landmarks.Translational relevanceThe ability to align and overlay imaging data from multiple instruments and manufacturers will permit better analysis of this complex data helping understand disease and predict treatment.

AimTo show the characteristics and outcomes of patients with neovascular age-related macular degeneration (nAMD) who had extended remission (ER) while on a pro re nata (PRN) treatment protocol.MethodsThis was a retrospective case-control study of a consecutive series of patients with nAMD treated with a PRN antivascular endothelial growth factor (anti-VEGF) drug regimen. ER was defined as the absence of haemorrhage, intraretinal/subretinal fluid on optical coherence tomography and leakage on fluorescein angiography for 52 weeks after cessation of anti-VEGF therapy. Matching patients with nAMD who did not achieve ER were included as control group. Cox regression analysis was fitted to identify predictors of time to achieve ER and time to recurrence. A logistic regression analysis of baseline characteristics was used to identify predictors of achieving ER.ResultsOf 830 eyes treated with anti-VEGF monotherapy, 77 (9.2%) eyes achieved ER during a median follow-up of 236 weeks (range 70-525 weeks). Cox regression analysis showed that ER was achieved earlier in eyes with isolated intraretinal fluid (HR, 2.05; 95% CI 1.929 to 4.520; p=0.045) at presentation. Logistic regression analysis showed that type 3 choroidal neovascularisation (OR, 0.090; 95% CI 0.021 to 0.382; p=0.001), thinner choroid (OR, 0.993; 95% CI 0.988 to 0.998; p=0.004) and absence of macular atrophy (OR, 0.233; 95% CI 0.065 to 0.839; p=0.026) at baseline increased the likelihood of achieving ER.ConclusionER is achievable in 9.2% of patients under PRN therapy for nAMD. At presentation with nAMD, anatomical features on retinal imaging may predict the likelihood of achieving ER and a shorter time to achieve ER.

PurposeTo assess early retinal microvascular and functional changes in diabetic patients without clinical evidence of diabetic retinopathy with optical coherence tomography angiography and central visual analyzer.MethodsThis was an observational case-control study of diabetic patients without diabetic retinopathy and nondiabetic controls. Patients underwent optical coherence tomography angiography imaging and visual acuity testing using the central visual analyzer. The foveal avascular zone area and the capillary density in the superficial and deep capillary plexuses were measured manually by a masked grader.ResultsSixty eyes from 35 diabetic patients were included in the study group, and 45 eyes from 31 nondiabetic patients were included in the control group. The foveal avascular zone area was not significantly different between the diabetic group and controls (both P > 0.05). The mean capillary density in the deep capillary plexus was significantly lower in diabetic eyes compared with control eyes (P = 0.04). The mean visual acuity in all central visual analyzer modules was significantly decreased in diabetic patients compared with controls (all P < 0.05).ConclusionOptical coherence tomography angiography was able to detect retinal microvascular changes in the deep capillary plexus, and the central visual analyzer showed signs of decreased visual acuity under conditions simulating suboptimal contrast and glare in diabetic patients without diabetic retinopathy.

PurposeTo determine the proportion of treatment-experienced eyes with exudative age-related macular degeneration successfully treated with every-4-week aflibercept that can be kept dry on fixed every-8-week aflibercept injections (maintenance).MethodsIn this retrospective chart review, we evaluated our cohort of patients treated with a treatment paradigm for CNV in AMD. Initially, patients were treated with bevacizumab or ranibizumab and switched to every-4-week aflibercept when therapeutic responses were not durable or were suboptimal. Maintenance every-8-week therapy was initiated when the retina was completely dry on every-4-week aflibercept therapy. The primary outcome measure was recurrence of exudation on optical coherence tomography (OCT) during maintenance.ResultsThirty-six eyes of 31 consecutive patients with median age of 79 years (range, 65-89) were included. Maintenance was started after a median of 34 (range, 8-88) injections. Recurrence was observed in 20 eyes (55%). Of these, 11 eyes (31%) reactivated at 8 weeks. Median time to failure of maintenance schedule was 40 weeks by Kaplan-Meier analysis. Baseline demographic and anatomic characteristics were not associated with failure of maintenance schedule.ConclusionIn treatment-experienced eyes that respond completely to every-4-week aflibercept, maintenance therapy with every-8-week injections can only temporarily maintain anatomic success with the majority of eyes developing recurring activity. This regimen fails early in one third of eyes and has a median effective duration of 40 weeks. Aflibercept appears to be inadequate to maintain control of exudation in most eyes in at least half of eyes undergoing long-term therapy.

Simple Summary: Subunit vaccines based on peptides as minimal immunogenic epitopes/antigens can induce highly specific immune responses without triggering adverse reactions. Here, Salmonella enterica serovar Typhimurium peptides were identified in the immunopeptidome of porcine macrophages genotyped for different class II molecules that were stimulated with the bacteria. Immunopeptidomics and mass spectrometry approaches were used for peptide identification, along with two predictors for immunoinformatics analysis. Differences between individuals were observed concerning length, elution preferences, and elution predictions. Thirty-one bacterial peptides were identified; outer membrane protein-A and chaperonin GroEL protein had the highest number of peptides. Since proteins having the highest number of peptides have been suggested as candidate antigens for vaccines, all the bacterial peptides identified can be considered promising candidate epitopes upon determining their immunogenicity. Salmonellosis is a zoonotic infection that has a major impact on human health; consuming contaminated pork products is the main source of such infection. Vaccination responses to classic vaccines have been unsatisfactory; that is why peptide subunit-based vaccines represent an excellent alternative. Immunopeptidomics was used in this study as a novel approach for identifying antigens coupled to major histocompatibility complex class II molecules. Three homozygous individuals having three different haplotypes (Lr-0.23, Lr-0.12, and Lr-0.21) were thus selected as donors; peripheral blood macrophages were then obtained and stimulated with Salmonella typhimurium (MOI 1:40). Although similarities were observed regarding peptide length distribution, elution patterns varied between individuals; in total, 1990 unique peptides were identified as follows: 372 for Pig 1 (Lr-0.23), 438 for Pig 2 (Lr.0.12) and 1180 for Pig 3 (Lr.0.21). Thirty-one S. typhimurium unique peptides were identified; most of the identified peptides belonged to outer membrane protein A and chaperonin GroEL. Notably, 87% of the identified bacterial peptides were predicted in silico to be elution ligands. These results encourage further in vivo studies to assess the immunogenicity of the identified peptides, as well as their usefulness as possible protective vaccine candidates. [ABSTRACT FROM AUTHOR]

BackgroundHierarchical cluster analysis (HCA) is a widely used classificatory technique in many areas of scientific knowledge. Applications usually yield a dendrogram from an HCA run over a given data set, using a grouping algorithm and a similarity measure. However, even when such parameters are fixed, ties in proximity (i.e. two equidistant clusters from a third one) may produce several different dendrograms, having different possible clustering patterns (different classifications). This situation is usually disregarded and conclusions are based on a single result, leading to questions concerning the permanence of clusters in all the resulting dendrograms; this happens, for example, when using HCA for grouping molecular descriptors to select that less similar ones in QSAR studies.ResultsRepresenting dendrograms in graph theoretical terms allowed us to introduce four measures of cluster frequency in a canonical way, and use them to calculate cluster frequencies over the set of all possible dendrograms, taking all ties in proximity into account. A toy example of well separated clusters was used, as well as a set of 1666 molecular descriptors calculated for a group of molecules having hepatotoxic activity to show how our functions may be used for studying the effect of ties in HCA analysis. Such functions were not restricted to the tie case; the possibility of using them to derive cluster stability measurements on arbitrary sets of dendrograms having the same leaves is discussed, e.g. dendrograms from variations of HCA parameters. It was found that ties occurred frequently, some yielding tens of thousands of dendrograms, even for small data sets.ConclusionsOur approach was able to detect trends in clustering patterns by offering a simple way of measuring their frequency, which is often very low. This would imply, that inferences and models based on descriptor classifications (e.g. QSAR) are likely to be biased, thereby requiring an assessment of their reliability. Moreover, any classification of molecular descriptors is likely to be far from unique. Our results highlight the need for evaluating the effect of ties on clustering patterns before classification results can be used accurately.Graphical abstractFour cluster contrast functions identifying statistically sound clusters within dendrograms considering ties in proximity.