Artificial intelligence for detecting keratoconus.

BACKGROUND: Keratoconus remains difficult to diagnose, especially in the early stages. It is a progressive disorder of the cornea that starts at a young age. Diagnosis is based on clinical examination and corneal imaging; though in the early stages, when there are no clinical signs, diagnosis depends on the interpretation of corneal imaging (e.g. topography and tomography) by trained cornea specialists. Using artificial intelligence (AI) to analyse the corneal images and detect cases of keratoconus could help prevent visual acuity loss and even corneal transplantation. However, a missed diagnosis in people seeking refractive surgery could lead to weakening of the cornea and keratoconus-like ectasia. There is a need for a reliable overview of the accuracy of AI for detecting keratoconus and the applicability of this automated method to the clinical setting. OBJECTIVES: To assess the diagnostic accuracy of artificial intelligence (AI) algorithms for detecting keratoconus in people presenting with refractive errors, especially those whose vision can no longer be fully corrected with glasses, those seeking corneal refractive surgery, and those suspected of having keratoconus. AI could help ophthalmologists, optometrists, and other eye care professionals to make decisions on referral to cornea specialists. Secondary objectives To assess the following potential causes of heterogeneity in diagnostic performance across studies. • Different AI algorithms (e.g. neural networks, decision trees, support vector machines) • Index test methodology (preprocessing techniques, core AI method, and postprocessing techniques) • Sources of input to train algorithms (topography and tomography images from Placido disc system, Scheimpflug system, slit-scanning system, or optical coherence tomography (OCT); number of training and testing cases/images; label/endpoint variable used for training) • Study setting • Study design • Ethnicity, or geographic area as its proxy • Different index test positivity criteria provided by the topography or tomography device • Reference standard, topography or tomography, one or two cornea specialists • Definition of keratoconus • Mean age of participants • Recruitment of participants • Severity of keratoconus (clinically manifest or subclinical) SEARCH METHODS: We searched CENTRAL (which contains the Cochrane Eyes and Vision Trials Register), Ovid MEDLINE, Ovid Embase, OpenGrey, the ISRCTN registry, ClinicalTrials.gov, and the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP). There were no date or language restrictions in the electronic searches for trials. We last searched the electronic databases on 29 November 2022. SELECTION CRITERIA: We included cross-sectional and diagnostic case-control studies that investigated AI for the diagnosis of keratoconus using topography, tomography, or both. We included studies that diagnosed manifest keratoconus, subclinical keratoconus, or both. The reference standard was the interpretation of topography or tomography images by at least two cornea specialists. DATA COLLECTION AND ANALYSIS: Two review authors independently extracted the study data and assessed the quality of studies using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. When an article contained multiple AI algorithms, we selected the algorithm with the highest Youden's index. We assessed the certainty of evidence using the GRADE approach. MAIN RESULTS: We included 63 studies, published between 1994 and 2022, that developed and investigated the accuracy of AI for the diagnosis of keratoconus. There were three different units of analysis in the studies: eyes, participants, and images. Forty-four studies analysed 23,771 eyes, four studies analysed 3843 participants, and 15 studies analysed 38,832 images. Fifty-four articles evaluated the detection of manifest keratoconus, defined as a cornea that showed any clinical sign of keratoconus. The accuracy of AI seems almost perfect, with a summary sensitivity of 98.6% (95% confidence interval (CI) 97.6% to 99.1%) and a summary specificity of 98.3% (95% CI 97.4% to 98.9%). However, accuracy varied across studies and the certainty of the evidence was low. Twenty-eight articles evaluated the detection of subclinical keratoconus, although the definition of subclinical varied. We grouped subclinical keratoconus, forme fruste, and very asymmetrical eyes together. The tests showed good accuracy, with a summary sensitivity of 90.0% (95% CI 84.5% to 93.8%) and a summary specificity of 95.5% (95% CI 91.9% to 97.5%). However, the certainty of the evidence was very low for sensitivity and low for specificity. In both groups, we graded most studies at high risk of bias, with high applicability concerns, in the domain of patient selection, since most were case-control studies. Moreover, we graded the certainty of evidence as low to very low due to selection bias, inconsistency, and imprecision. We could not explain the heterogeneity between the studies. The sensitivity analyses based on study design, AI algorithm, imaging technique (topography versus tomography), and data source (parameters versus images) showed no differences in the results. AUTHORS' CONCLUSIONS: AI appears to be a promising triage tool in ophthalmologic practice for diagnosing keratoconus. Test accuracy was very high for manifest keratoconus and slightly lower for subclinical keratoconus, indicating a higher chance of missing a diagnosis in people without clinical signs. This could lead to progression of keratoconus or an erroneous indication for refractive surgery, which would worsen the disease. We are unable to draw clear and reliable conclusions due to the high risk of bias, the unexplained heterogeneity of the results, and high applicability concerns, all of which reduced our confidence in the evidence. Greater standardization in future research would increase the quality of studies and improve comparability between studies.

A 4-week high-AGE diet does not impair glucose metabolism and vascular function in obese individuals.

BACKGROUNDAccumulation of advanced glycation endproducts (AGEs) may contribute to the pathophysiology of type 2 diabetes and its vascular complications. AGEs are widely present in food, but whether restricting AGE intake improves risk factors for type 2 diabetes and vascular dysfunction is controversial.METHODSAbdominally obese but otherwise healthy individuals were randomly assigned to a specifically designed 4-week diet low or high in AGEs in a double-blind, parallel design. Insulin sensitivity, secretion, and clearance were assessed by a combined hyperinsulinemic-euglycemic and hyperglycemic clamp. Micro- and macrovascular function, inflammation, and lipid profiles were assessed by state-of-the-art in vivo measurements and biomarkers. Specific urinary and plasma AGEs Nε-(carboxymethyl)lysine (CML), Nε-(1-carboxyethyl)lysine (CEL), and Nδ-(5-hydro-5-methyl-4-imidazolon-2-yl)-ornithine (MG-H1) were assessed by mass spectrometry.RESULTSIn 73 individuals (22 males, mean ± SD age and BMI 52 ± 14 years, 30.6 ± 4.0 kg/m2), intake of CML, CEL, and MG-H1 differed 2.7-, 5.3-, and 3.7-fold between the low- and high-AGE diets, leading to corresponding changes of these AGEs in urine and plasma. Despite this, there was no difference in insulin sensitivity, secretion, or clearance; micro- and macrovascular function; overall inflammation; or lipid profile between the low and high dietary AGE groups (for all treatment effects, P > 0.05).CONCLUSIONThis comprehensive RCT demonstrates very limited biological consequences of a 4-week diet low or high in AGEs in abdominally obese individuals.TRIAL REGISTRATIONClinicaltrials.gov, NCT03866343; trialregister.nl, NTR7594.FUNDINGDiabetesfonds and ZonMw.

Association between retinal vascular measures and brain white matter lesions in schizophrenia.

OBJECTIVE: Recent studies have examined retinal vascular abnormalities in schizophrenia as retinal vascular imaging is a non-invasive proxy to cerebral microvasculature. However, relation between retinal vascular abnormalities and brain structure is not well examined in schizophrenia. Hence in this study, for the first time, we examined the relationship between retinal vascular measures and brain white matter lesions in schizophrenia. We examined brain white matter lesions as they are considered a predictive marker for future adverse cerebrovascular event. METHODS: We acquired retinal vascular images of both eyes using a non-mydriatic camera and calculated retinal vascular diameter, tortuosity, trajectory and fractal dimension using validated methods. All patients underwent Magnetic Resonance Imaging of bran and we computed white matter hypo-intensities using Freesurfer software. We performed a linear regression analysis to examine the relationship between white matter hypo-intensities and retinal vascular measures controlling for age, sex, fasting blood sugar, creatinine, whole-brain volume, and antipsychotic dose. RESULTS: The regression model was significant in Schizophrenia patients (R=0.983;R2 =0.966;-F=10.849;p = 0.008) but not in healthy volunteers (R=0.828;R2 =0.686;F=0.182; p = 0.963). Among the retinal vascular measures, arterial tortuosity (ß = 0.963;p-0.002), tortuosity (ß = -1.002;p = 0.001) and fractal dimension (ß = -0.688;p = 0.014) were significant predictors of white matter lesions. DISCUSSION: The current study's findings support the conclusion that retinal vascular fractal dimension and tortuosity are associated with changes in cerebral white matter and may be considered proxy markers for cerebral microvasculature in schizophrenia. Considering the relationship between white matter lesions and stroke, these observations could have important clinical implications to screen schizophrenia patients for risk of adverse cerebrovascular event.

Overflow phenomenon in serum lutein after supplementation: a systematic review supported with SNPs analyses.

Lutein, a type of carotenoids, is found to delay the onset and progression of age-related macular degeneration (AMD). Several lutein supplementation studies showed that after an initial increase, lutein serum levels demonstrated a subsequent decrease despite continuous supplementation. In this systematic literature review, this obscure phenomenon was tried to be explained. The subsequent drop in lutein levels was postulated due to down-regulation of lutein receptors scavenger receptor class B type I (SR-BI) in the gastrointestinal tract, upregulation of lutein degrading enzyme ß-carotene dioxygenase (BCDO2), or perhaps a combination of both. Some single nucleotides polymorphisms (SNPs) that could have influence on the occurrence of this phenomenon. To date, an exact scientific explanation for this phenomenon has not been established. Further research is needed to investigate this phenomenon in depth to reach an irrefutable explanation, giving that lutein is proven to be effective in delaying the onset and progression of AMD and its metabolism in the human body becomes of equal importance.

Reference values for anterior chamber morphometrics with swept-source optical coherence tomography in a Caucasian population.

PURPOSE: The aim of this study is to generate reference values for anterior chamber morphometrics to facilitate the interpretation of anterior chamber images acquired with swept-source optical coherence tomography (SS-OCT) in a Caucasian population. MATERIALS AND METHODS: Non-glaucomatous Caucasian subjects, who were newly referred to the outpatient department of the University Eye Clinic Maastricht UMC+, were asked to undergo anterior segment assessment with SS-OCT imaging prior to routine eye examination (including visual acuity and refractive error measurements, Goldmann applanation tonometry, slit lamp examination and funduscopy). RESULTS: A total of 647 Caucasian subjects, aged 40-80 years, were included. Mean age (± standard deviation) was 61.7±10.3 years, and 294 were male (45%). Mean refractive error was -0.63±3.17 D. OCT images and measurements were obtained in all cases. The mean temporal trabecular iris angle at 500 µm (TIA500) was 27.04°±12.04°, the mean nasal TIA500 was 27.43°±11.75°, the mean anterior chamber depth (ACD) was 2.75±0.39 mm and the mean lens vault (LV) was 0.44±0.30 mm. A smaller temporal TIA500, shallower ACD and higher LV were found in female and hyperopic subjects. CONCLUSION: Reference values for anterior chamber morphometrics in Caucasians were generated, quantified and presented. These values offer tools for the interpretation of anterior segment images of Caucasians and differentiation between presumed mechanisms of angle closure.

The influence of consuming an egg or an egg-yolk buttermilk drink for 12 wk on serum lipids, inflammation, and liver function markers in human volunteers.

OBJECTIVES: Dietary cholesterol elevates serum total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) concentrations modestly. There are indications that the cholesterol-raising effect depends on the food matrix, that is, sphingolipids and lactic acid bacteria are suggested to influence cholesterol metabolism. Traditional buttermilk is rich in both sphingolipids and lactic acid bacteria. Therefore, the aim of this study was to evaluate whether effects on cholesterol metabolism depend on food matrix (e.g., cholesterol provided as egg [yolk] or incorporated into traditionally prepared buttermilk drink). METHODS: Participants (N = 97) took part in a 12-wk intervention study. The controls (n = 20) continued their regular egg consumption of one to two eggs a week. The other two groups consumed either one extra egg per day (n = 57) or a buttermilk drink containing one egg yolk (n = 20). Blood was sampled at day 1 and at the end of the experimental period (day 90) to analyze serum lipids, lipoproteins, and markers reflecting cholesterol metabolism, low-grade systemic inflammation, endothelial activity, and liver function. RESULTS: Serum TC and LDL-C concentrations increased significantly by respectively 0.63 mmol/L (P < 0.05) and 0.59 mmol/L (P < 0.05) in women consuming one additional egg per day compared with controls. There were no effects on markers for inflammation, endothelial activity, or liver function. The increase in serum TC and LDL-C concentration was no longer significant in women consuming the same egg yolk incorporated in a buttermilk drink (0.33 mmol/L [P = 0.66] and 0.31 mmol/L [P = 0.55], respectively). CONCLUSION: Daily egg consumption for 12 wk increases serum TC and LDL-C concentrations in women but not markers for inflammation, endothelial activity, and liver function. Interestingly, the rise in serum LDL-C concentrations is less pronounced when egg yolk is incorporated into a buttermilk drink, indeed suggesting that fractions in the buttermilk might influence dietary cholesterol absorption.