Development and External Validation of Machine Learning Models for Diabetic Microvascular Complications: Cross-Sectional Study With Metabolites.

BACKGROUND: Diabetic kidney disease (DKD) and diabetic retinopathy (DR) are major diabetic microvascular complications, contributing significantly to morbidity, disability, and mortality worldwide. The kidney and the eye, having similar microvascular structures and physiological and pathogenic features, may experience similar metabolic changes in diabetes. OBJECTIVE: This study aimed to use machine learning (ML) methods integrated with metabolic data to identify biomarkers associated with DKD and DR in a multiethnic Asian population with diabetes, as well as to improve the performance of DKD and DR detection models beyond traditional risk factors. METHODS: We used ML algorithms (logistic regression [LR] with Least Absolute Shrinkage and Selection Operator and gradient-boosting decision tree) to analyze 2772 adults with diabetes from the Singapore Epidemiology of Eye Diseases study, a population-based cross-sectional study conducted in Singapore (2004-2011). From 220 circulating metabolites and 19 risk factors, we selected the most important variables associated with DKD (defined as an estimated glomerular filtration rate <60 mL/min/1.73 m2) and DR (defined as an Early Treatment Diabetic Retinopathy Study severity level ≥20). DKD and DR detection models were developed based on the variable selection results and externally validated on a sample of 5843 participants with diabetes from the UK biobank (2007-2010). Machine-learned model performance (area under the receiver operating characteristic curve [AUC] with 95% CI, sensitivity, and specificity) was compared to that of traditional LR adjusted for age, sex, diabetes duration, hemoglobin A1c, systolic blood pressure, and BMI. RESULTS: Singapore Epidemiology of Eye Diseases participants had a median age of 61.7 (IQR 53.5-69.4) years, with 49.1% (1361/2772) being women, 20.2% (555/2753) having DKD, and 25.4% (685/2693) having DR. UK biobank participants had a median age of 61.0 (IQR 55.0-65.0) years, with 35.8% (2090/5843) being women, 6.7% (374/5570) having DKD, and 6.1% (355/5843) having DR. The ML algorithms identified diabetes duration, insulin usage, age, and tyrosine as the most important factors of both DKD and DR. DKD was additionally associated with cardiovascular disease history, antihypertensive medication use, and 3 metabolites (lactate, citrate, and cholesterol esters to total lipids ratio in intermediate-density lipoprotein), while DR was additionally associated with hemoglobin A1c, blood glucose, pulse pressure, and alanine. Machine-learned models for DKD and DR detection outperformed traditional LR models in both internal (AUC 0.838 vs 0.743 for DKD and 0.790 vs 0.764 for DR) and external validation (AUC 0.791 vs 0.691 for DKD and 0.778 vs 0.760 for DR). CONCLUSIONS: This study highlighted diabetes duration, insulin usage, age, and circulating tyrosine as important factors in detecting DKD and DR. The integration of ML with biomedical big data enables biomarker discovery and improves disease detection beyond traditional risk factors.

Blood biomarkers improve the prediction of prevalent and incident severe chronic kidney disease.

BACKGROUND: The prevalence of chronic kidney disease (CKD) is high. Identification of cases with CKD or at high risk of developing it is important to tailor early interventions. The objective of this study was to identify blood metabolites associated with prevalent and incident severe CKD, and to quantify the corresponding improvement in CKD detection and prediction. METHODS: Data from four cohorts were analyzed: Singapore Epidemiology of Eye Diseases (SEED) (n = 8802), Copenhagen Chronic Kidney Disease (CPH) (n = 916), Singapore Diabetic Nephropathy (n = 714), and UK Biobank (UKBB) (n = 103,051). Prevalent CKD (stages 3-5) was defined as estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m2; incident severe CKD as CKD-related mortality or kidney failure occurring within 10 years. We used multivariable regressions to identify, among 146 blood metabolites, those associated with CKD, and quantify the corresponding increase in performance. RESULTS: Chronic kidney disease prevalence (stages 3-5) and severe incidence were 11.4% and 2.2% in SEED, and 2.3% and 0.2% in UKBB. Firstly, phenylalanine (Odds Ratio [OR] 1-SD increase = 1.83 [1.73, 1.93]), tyrosine (OR = 0.75 [0.71, 0.79]), docosahexaenoic acid (OR = 0.90 [0.85, 0.95]), citrate (OR = 1.41 [1.34, 1.47]) and triglycerides in medium high density lipoprotein (OR = 1.07 [1.02, 1.13]) were associated with prevalent stages 3-5 CKD. Mendelian randomization analyses suggested causal relationships. Adding these metabolites beyond traditional risk factors increased the area under the curve (AUC) by 3% and the sensitivity by 7%. Secondly, lactate (HR = 1.33 [1.08, 1.64]) and tyrosine (HR = 0.74 [0.58, 0.95]) were associated with incident severe CKD among individuals with eGFR < 90 mL/min/1.73 m2 at baseline. These metabolites increased the c-index by 2% and sensitivity by 5% when added to traditional risk factors. CONCLUSION: The performance improvements of CKD detection and prediction achieved by adding metabolites to traditional risk factors are modest and further research is necessary to fully understand the clinical implications of these findings.

Ocular microvascular complications in diabetic retinopathy: insights from machine learning.

INTRODUCTION: Diabetic retinopathy (DR) is a leading cause of preventable blindness among working-age adults, primarily driven by ocular microvascular complications from chronic hyperglycemia. Comprehending the complex relationship between microvascular changes in the eye and disease progression poses challenges, traditional methods assuming linear or logistical relationships may not adequately capture the intricate interactions between these changes and disease advances. Hence, the aim of this study was to evaluate the microvascular involvement of diabetes mellitus (DM) and non-proliferative DR with the implementation of non-parametric machine learning methods. RESEARCH DESIGN AND METHODS: We conducted a retrospective cohort study that included optical coherence tomography angiography (OCTA) images collected from a healthy group (196 eyes), a DM no DR group (120 eyes), a mild DR group (71 eyes), and a moderate DR group (66 eyes). We implemented a non-parametric machine learning method for four classification tasks that used parameters extracted from the OCTA images as predictors: DM no DR versus healthy, mild DR versus DM no DR, moderate DR versus mild DR, and any DR versus no DR. SHapley Additive exPlanations values were used to determine the importance of these parameters in the classification. RESULTS: We found large choriocapillaris flow deficits were the most important for healthy versus DM no DR, and became less important in eyes with mild or moderate DR. The superficial microvasculature was important for the healthy versus DM no DR and mild DR versus moderate DR tasks, but not for the DM no DR versus mild DR task-the stage when deep microvasculature plays an important role. Foveal avascular zone metric was in general less affected, but its involvement increased with worsening DR. CONCLUSIONS: The findings from this study provide valuable insights into the microvascular involvement of DM and DR, facilitating the development of early detection methods and intervention strategies.

Cardiovascular disease and thinning of retinal nerve fiber layer in a multi-ethnic Asian population: the Singapore epidemiology of eye diseases study.

Introduction: Our study aimed to examine the relationship between cardiovascular diseases (CVD) with peripapillary retinal fiber layer (RNFL) and macular ganglion cell-inner plexiform layer (GCIPL) thickness profiles in a large multi-ethnic Asian population study. Methods: 6,024 Asian subjects were analyzed in this study. All participants underwent standardized examinations, including spectral domain OCT imaging (Cirrus HD-OCT; Carl Zeiss Meditec). In total, 9,188 eyes were included for peripapillary RNFL analysis (2,417 Malays; 3,240 Indians; 3,531 Chinese), and 9,270 eyes (2,449 Malays, 3,271 Indians, 3,550 Chinese) for GCIPL analysis. History of CVD was defined as a self-reported clinical history of stroke, myocardial infarction, or angina. Multivariable linear regression models with generalized estimating equations were performed, adjusting for age, gender, ethnicity, diabetes, hypertension, hyperlipidaemia, chronic kidney disease, body mass index, current smoking status, and intraocular pressure. Results: We observed a significant association between CVD history and thinner average RNFL (ß = -1.63; 95% CI, -2.70 to -0.56; p = 0.003). This association was consistent for superior (ß = -1.79, 95% CI, -3.48 to -0.10; p = 0.038) and inferior RNFL quadrant (ß = -2.14, 95% CI, -3.96 to -0.32; p = 0.021). Of the CVD types, myocardial infarction particularly showed significant association with average (ß = -1.75, 95% CI, -3.08 to -0.42; p = 0.010), superior (ß = -2.22, 95% CI, -4.36 to -0.09; p = 0.041) and inferior (ß = -2.42, 95% CI, -4.64 to -0.20; p = 0.033) RNFL thinning. Among ethnic groups, the association between CVD and average RNFL was particularly prominent in Indian eyes (ß = -1.92, 95% CI, -3.52 to -0.33; p = 0.018). CVD was not significantly associated with average GCIPL thickness, albeit a consistent negative direction of association was observed (ß = -0.22, 95% CI, -1.15 to 0.71; p = 0.641). Discussion: In this large multi-ethnic Asian population study, we observed significant association between CVD history and RNFL thinning. This finding further validates the impact of impaired systemic circulation on RNFL thickness.

Prediction of diabetic kidney disease risk using machine learning models: A population-based cohort study of Asian adults.

Background: Machine learning (ML) techniques improve disease prediction by identifying the most relevant features in multidimensional data. We compared the accuracy of ML algorithms for predicting incident diabetic kidney disease (DKD). Methods: We utilized longitudinal data from 1365 Chinese, Malay, and Indian participants aged 40-80 y with diabetes but free of DKD who participated in the baseline and 6-year follow-up visit of the Singapore Epidemiology of Eye Diseases Study (2004-2017). Incident DKD (11.9%) was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2 with at least 25% decrease in eGFR at follow-up from baseline. A total of 339 features, including participant characteristics, retinal imaging, and genetic and blood metabolites, were used as predictors. Performances of several ML models were compared to each other and to logistic regression (LR) model based on established features of DKD (age, sex, ethnicity, duration of diabetes, systolic blood pressure, HbA1c, and body mass index) using area under the receiver operating characteristic curve (AUC). Results: ML model Elastic Net (EN) had the best AUC (95% CI) of 0.851 (0.847-0.856), which was 7.0% relatively higher than by LR 0.795 (0.790-0.801). Sensitivity and specificity of EN were 88.2 and 65.9% vs. 73.0 and 72.8% by LR. The top 15 predictors included age, ethnicity, antidiabetic medication, hypertension, diabetic retinopathy, systolic blood pressure, HbA1c, eGFR, and metabolites related to lipids, lipoproteins, fatty acids, and ketone bodies. Conclusions: Our results showed that ML, together with feature selection, improves prediction accuracy of DKD risk in an asymptomatic stable population and identifies novel risk factors, including metabolites. Funding: This study was supported by the National Medical Research Council, NMRC/OFLCG/001/2017 and NMRC/HCSAINV/MOH-001019-00. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Risk prediction models for type 2 diabetes using either fasting plasma glucose or HbA1c in Chinese, Malay, and Indians: Results from three multi-ethnic Singapore cohorts.

AIMS: To assess three well-established type 2 diabetes (T2D) risk prediction models based on fasting plasma glucose (FPG) in Chinese, Malays, and Indians, and to develop simplified risk models based on either FPG or HbA1c. METHODS: We used a prospective multiethnic Singapore cohort to evaluate the established models and develop simplified models. 6,217 participants without T2D at baseline were included, with an average follow-up duration of 8.3 years. The simplified risk models were validated in two independent multiethnic Singapore cohorts (N = 12,720). RESULTS: The established risk models had moderate-to-good discrimination (area under the receiver operating characteristic curves, AUCs 0.762 - 0.828) but a lack of fit (P-values < 0.05). Simplified risk models that included fewer predictors (age, BMI, systolic blood pressure, triglycerides, and HbA1c or FPG) showed good discrimination in all cohorts (AUCs ≥ 0.810), and sufficiently captured differences between the ethnic groups. While recalibration improved fit the simplified models in validation cohorts, there remained evidence of miscalibration in Chinese (p ≤ 0.012). CONCLUSIONS: Simplified risk models including HbA1c or FPG had good discrimination in predicting incidence of T2D in three major Asian ethnic groups. Risk functions with HbA1c performed as well as those with FPG.

Validation of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from UK Biobank.

BACKGROUND: Currently in the United Kingdom, cardiovascular disease (CVD) risk assessment is based on the QRISK3 score, in which 10% 10-year CVD risk indicates clinical intervention. However, this benchmark has limited efficacy in clinical practice and the need for a more simple, non-invasive risk stratification tool is necessary. Retinal photography is becoming increasingly acceptable as a non-invasive imaging tool for CVD. Previously, we developed a novel CVD risk stratification system based on retinal photographs predicting future CVD risk. This study aims to further validate our biomarker, Reti-CVD, (1) to detect risk group of ≥ 10% in 10-year CVD risk and (2) enhance risk assessment in individuals with QRISK3 of 7.5-10% (termed as borderline-QRISK3 group) using the UK Biobank. METHODS: Reti-CVD scores were calculated and stratified into three risk groups based on optimized cut-off values from the UK Biobank. We used Cox proportional-hazards models to evaluate the ability of Reti-CVD to predict CVD events in the general population. C-statistics was used to assess the prognostic value of adding Reti-CVD to QRISK3 in borderline-QRISK3 group and three vulnerable subgroups. RESULTS: Among 48,260 participants with no history of CVD, 6.3% had CVD events during the 11-year follow-up. Reti-CVD was associated with an increased risk of CVD (adjusted hazard ratio [HR] 1.41; 95% confidence interval [CI], 1.30-1.52) with a 13.1% (95% CI, 11.7-14.6%) 10-year CVD risk in Reti-CVD-high-risk group. The 10-year CVD risk of the borderline-QRISK3 group was greater than 10% in Reti-CVD-high-risk group (11.5% in non-statin cohort [n = 45,473], 11.5% in stage 1 hypertension cohort [n = 11,966], and 14.2% in middle-aged cohort [n = 38,941]). C statistics increased by 0.014 (0.010-0.017) in non-statin cohort, 0.013 (0.007-0.019) in stage 1 hypertension cohort, and 0.023 (0.018-0.029) in middle-aged cohort for CVD event prediction after adding Reti-CVD to QRISK3. CONCLUSIONS: Reti-CVD has the potential to identify individuals with ≥ 10% 10-year CVD risk who are likely to benefit from earlier preventative CVD interventions. For borderline-QRISK3 individuals with 10-year CVD risk between 7.5 and 10%, Reti-CVD could be used as a risk enhancer tool to help improve discernment accuracy, especially in adult groups that may be pre-disposed to CVD.

Deep Learning System Outperforms Clinicians in Identifying Optic Disc Abnormalities.

BACKGROUND: The examination of the optic nerve head (optic disc) is mandatory in patients with headache, hypertension, or any neurological symptoms, yet it is rarely or poorly performed in general clinics. We recently developed a brain and optic nerve study with artificial intelligence-deep learning system (BONSAI-DLS) capable of accurately detecting optic disc abnormalities including papilledema (swelling due to elevated intracranial pressure) on digital fundus photographs with a comparable classification performance to expert neuro-ophthalmologists, but its performance compared to first-line clinicians remains unknown. METHODS: In this international, cross-sectional multicenter study, the DLS, trained on 14,341 fundus photographs, was tested on a retrospectively collected convenience sample of 800 photographs (400 normal optic discs, 201 papilledema and 199 other abnormalities) from 454 patients with a robust ground truth diagnosis provided by the referring expert neuro-ophthalmologists. The areas under the receiver-operating-characteristic curves were calculated for the BONSAI-DLS. Error rates, accuracy, sensitivity, and specificity of the algorithm were compared with those of 30 clinicians with or without ophthalmic training (6 general ophthalmologists, 6 optometrists, 6 neurologists, 6 internists, 6 emergency department [ED] physicians) who graded the same testing set of images. RESULTS: With an error rate of 15.3%, the DLS outperformed all clinicians (average error rates 24.4%, 24.8%, 38.2%, 44.8%, 47.9% for general ophthalmologists, optometrists, neurologists, internists and ED physicians, respectively) in the overall classification of optic disc appearance. The DLS displayed significantly higher accuracies than 100%, 86.7% and 93.3% of clinicians (n = 30) for the classification of papilledema, normal, and other disc abnormalities, respectively. CONCLUSIONS: The performance of the BONSAI-DLS to classify optic discs on fundus photographs was superior to that of clinicians with or without ophthalmic training. A trained DLS may offer valuable diagnostic aid to clinicians from various clinical settings for the screening of optic disc abnormalities harboring potentially sight- or life-threatening neurological conditions.

Relationship between dry eye symptoms and quality of life: associations and mediation analysis.

PURPOSE: (1) To determine the independent association of dry eye symptoms with health-related quality of life (HRQoL) in the Singapore population and (2) to further investigate which factors mediate this association. METHODS: In this cross-sectional study, 7707 participants were included. The presence of dry eye symptoms was defined as experiencing at least one out of the six symptoms either 'often' or 'all the time'. The EuroQoL-5 dimensions (EQ-5D) utility instrument (raw scores converted to UK time trade-off (TTO) values) was used to assess generic HRQoL and the overall score from the Visual Functioning Questionnaire for visual functioning. The association between dry eye symptoms and EQ-5D was investigated using multivariable linear regression, adjusting for demographic and socioeconomic information, comorbidities, systemic and ocular examinations results. Mediation analysis was used to determine whether certain factors mediated this association. RESULTS: After adjusting for relevant factors, those with dry eye symptoms had significantly lower HRQoL (difference in EQ-5D TTO: -0.062 (95% CI -0.073 to -0.050)), with the inability to open eyes affected the most (-0.101 (95% CI -0.161 to -0.042)), followed by a sandy sensation (-0.089 (95% CI -0.121 to -0.058)), a burning sensation (-0.070 (95% CI -0.105 to -0.036)), red eyes (-0.059 (95% CI -0.082 to -0.036)), a dry sensation (-0.058 (95% CI -0.072 to -0.044)) and crusting of eyelids (-0.040 (95% CI -0.071 to -0.008)). Visual functioning and the presence of recent falls accounted for 8.63% (4.98%-14.5%) and 2.93% (0.04%-5.68%) of the indirect relationship between dry eye and HRQoL, respectively. CONCLUSION: Dry eye symptoms were independently associated with poor HRQoL. Moreover, this was partly mediated by reduced visual functioning and experiencing recent falls. Our results suggest that efforts to reduce severity of dry eye symptoms are essential to optimise patients' overall functioning and well-being.

Handheld chromatic pupillometry can accurately and rapidly reveal functional loss in glaucoma.

BACKGROUND/AIMS: Early detection and treatment of glaucoma can delay vision loss. In this study, we evaluate the performance of handheld chromatic pupillometry (HCP) for the objective and rapid detection of functional loss in glaucoma. METHODS: In this clinic-based, prospective study, we enrolled 149 patients (median (IQR) years: 68.5 (13.6) years) with confirmed glaucoma and 173 healthy controls (55.2 (26.7) years). Changes in pupil size in response to 9 s of exponentially increasing blue (469 nm) and red (640 nm) light-stimuli were assessed monocularly using a custom-built handheld pupillometer. Pupillometric features were extracted from individual traces and compared between groups. Features with the highest classification potential, selected using a gradient boosting machine technique, were incorporated into a generalised linear model for glaucoma classification. Receiver operating characteristic curve analyses (ROC) were used to compare the performance of HCP, optical coherence tomography (OCT) and Humphrey Visual Field (HVF). RESULTS: Pupillary light responses were altered in glaucoma compared with controls. For glaucoma classification, HCP yielded an area under the ROC curve (AUC) of 0.94 (95% CI 0.91 to 0.96), a sensitivity of 87.9% and specificity of 88.4%. The classification performance of HCP in early-moderate glaucoma (visual field mean deviation (VFMD) > -12 dB; AUC=0.91 (95% CI 0.87 to 0.95)) was similar to HVF (AUC=0.91) and reduced compared with OCT (AUC=0.97; p=0.01). For severe glaucoma (VFMD ≤ -12 dB), HCP had an excellent classification performance (AUC=0.98, 95% CI 0.97 to 1) that was similar to HVF and OCT. CONCLUSION: HCP allows for an accurate, objective and rapid detection of functional loss in glaucomatous eyes of different severities.

Combining retinal and choroidal microvascular metrics improves discriminative power for diabetic retinopathy.

PURPOSE: To use optical coherence tomography angiography (OCTA) parameters from both the retinal and choroidal microvasculature to detect the presence and severity of diabetic retinopathy (DR). METHOD: This is a cross-sectional case-control study. OCTA parameters from retinal vasculature, fovea avascular zone (FAZ) and choriocapillaris were evaluated from 3×3 mm2 fovea-centred scans. Areas under the receiver operating characteristic (ROC) curve were used to compare the discriminative power on the presence of diabetes mellitus (DM), the presence of DR and need for referral: group 1 (no DM vs DM no DR), group 2 (no DR vs any DR) and group 3 (non-proliferative DR (NPDR) vs proliferative DR (PDR)). RESULTS: 35 eyes from 27 participants with no DM and 132 eyes from 75 with DM were included. DR severity was classified into three groups: no DR group (62 eyes), NPDR (51 eyes), PDR (19 eyes). All retinal vascular parameters, FAZ parameters and choriocapillaris parameters were strongly altered with DR stages (p<0.01), except for the deep plexus FAZ area (p=0.619). Choriocapillaris parameters allowed to better discriminate between no DM versus DM no DR group compared with retinal parameters (areas under the ROC curve=0.954 vs 0.821, p=0.006). A classification model including retinal and choroidal microvasculature significantly improved the discrimination between DR and no DR compared with each parameter separately (p=0.029). CONCLUSIONS: Evaluating OCTA parameters from both the retinal and choroidal microvasculature in 3×3 mm scans improves the discrimination of DM and early DR.

Machine learning identifying peripheral circulating metabolites associated with intraocular pressure alterations.

AIMS: To identify blood metabolite markers associated with intraocular pressure (IOP) in a population-based cross-sectional study. METHODS: This study was conducted in a multiethnic Asian population (Chinese, n=2805; Indians, n=3045; Malays, n=3041 aged 40-80 years) in Singapore. All subjects underwent standardised systemic and ocular examinations, and biosamples were collected. Selected metabolites (n=228) in either serum or plasma were analysed and quantified using nuclear magnetic resonance spectroscopy. Least absolute shrinkage and selection operator regression was used for metabolites selection. Multivariable linear regression was used to evaluate the relationship between metabolites and IOP in each of the three ethnic groups, followed by a meta-analysis combining the three cohorts. RESULTS: Six metabolites, including albumin, glucose, lactate, glutamine, ratio of saturated fatty acids to total fatty acids (SFAFA) and cholesterol esters in very large high-density lipoprotein (HDL), were significantly associated with IOP in all three cohorts. Higher levels of albumin (per SD, beta=0.24, p=0.002), lactate (per SD, beta=0.27, p=0.008), glucose (per SD, beta=0.11, p=0.010) and cholesterol esters in very large HDL (per SD, beta=0.47, p=0.006), along with lower levels of glutamine (per SD, beta=0.17, p<0.001) and SFAFA (per SD, beta=0.21, p=0.008) were associated with higher IOP levels. CONCLUSION: We identify several novel blood metabolites associated with IOP. These findings may provide insight into the physiological and pathological processes underlying IOP control.

Relationships between Lipid-Related Metabolites and Age-Related Macular Degeneration Vary with Complement Genotype.

Objective: Lipid dysregulation and complement system (CS) activation are 2 important pathophysiology pathways for age-related macular degeneration (AMD). We hypothesized that the relationship between lipids and AMD may also differ according to CS genotype profile. Thus, the objective was to investigate the relationships between lipid-related metabolites and AMD according to CS genotypes. Design: Population-based cross-sectional study. Participants: A total of 6947 participants from Singapore Epidemiology of Eye Diseases study with complete relevant data were included. Methods: We investigated a total of 32 blood lipid-related metabolites from nuclear magnetic resonance metabolomics data including lipoproteins and their subclasses, cholesterols, glycerides, and phospholipids, as well as 4 CS single nucleotide polymorphisms (SNPs): rs10922109 (complement factor H), rs10033900 (complement factor I), rs116503776 (C2-CFB-SKIV2L), and rs2230199 (C3). We first investigated the associations between AMD and the 32 lipid-related metabolites using multivariable logistic regression models. Then, to investigate whether the effect of lipid-related metabolites on AMD differ according to the CS SNPs, we tested the possible interactions between the CS SNPs and the lipid-related metabolites. Main Outcome Measures: Age-related macular degeneration was defined using the Wisconsin grading system. Results: Among the 6947 participants, the prevalence of AMD was 6.1%, and the mean age was 58.3 years. First, higher levels of cholesterol in high-density lipoprotein (HDL) and medium and large HDL particles were associated with an increased risk of AMD, and higher levels of serum total triglycerides (TG) and several very-low-density lipoprotein subclass particles were associated with a decreased risk of AMD. Second, these lipids had significant interaction effects on AMD with 2 CS SNPs: rs2230199 and rs116503776 (after correction for multiple testing). For rs2230199, in individuals without risk allele, higher total cholesterol in HDL2 was associated with an increased AMD risk (odds ratio [OR] per standard deviation increase, 1.20; 95% confidence interval (CI), 1.06-1.37; P = 0.005), whereas, in individuals with at least 1 risk allele, higher levels of these particles were associated with a decreased AMD risk (OR, 0.69; 95% CI, 0.45-1.05; P = 0.079). Conversely, for rs116503776, in individuals without risk allele, higher serum total TG were associated with a decreased AMD risk (OR, 0.84; 95% CI, 0.74-0.95; P = 0.005), whereas, in individuals with 2 risk alleles, higher levels of these particles were associated with an increased risk of AMD (OR, 2.3, 95% CI, 0.99-5.39, P = 0.054). Conclusions: Lipid-related metabolites exhibit opposite directions of effects on AMD according to CS genotypes. This indicates that lipid metabolism and CS may have synergistic interplay in the AMD pathogenesis.

Retinal vascular profile in predicting incident cardiometabolic diseases among individuals with diabetes.

OBJECTIVE: To determine the longitudinal associations between retinal vascular profile (RVP) and four major cardiometabolic diseases; and to quantify the predictive improvements when adding RVP beyond traditional risk factors in individuals with diabetes. METHODS: Subjects were enrolled from the Singapore Epidemiology of Eye Disease (SEED) study, a multi-ethnic population-based cohort. Four incident cardiometabolic diseases, calculated over a ~ 6-year period, were considered: cardiovascular disease (CVD), hypertension (HTN), diabetic kidney disease (DKD), and hyperlipidemia (HLD). The RVP-vessel tortuosity, branching angle, branching coefficient, fractal dimension, vessel caliber, and DR status-was characterized at baseline using a computer-assisted program. Traditional risk factors at baseline included age, gender, ethnicity, smoking, blood pressure (BP), HbA1c, estimated glomerular filtration rate (eGFR), or cholesterol. The improvements in predictive performance when adding RVP (compared with only traditional risk factors) was calculated using several metrics including area under the receiver operating characteristics curve (AUC) and net reclassification improvement (NRI). RESULTS: Among 1770 individuals with diabetes, incidences were 6.3% (n = 79/1259) for CVD, 48.7% (n = 166/341) for HTN, 14.6% (n = 175/1199) for DKD, and 59.4% (n = 336/566) for HLD. DR preceded the onset of CVD (RR 1.85[1.14;3.00]) and DKD (1.44 [1.06;1.96]). Narrower arteriolar caliber preceding the onset of HTN (0.84 [0.72;0.99]), and changes in arteriolar branching angle preceded the onset of CVD (0.78 [0.62;0.98]) and HTN (1.15 [1.03;1.29]). The largest predictive improvement was found for HTN with AUC increment of 3.4% (p = .027) and better reclassification of 11.4% of the cases and 4.6% of the controls (p = .008). CONCLUSION: We found that RVPs improved the prediction of HTN in individuals with diabetes, but add limited information for CVD, DKD, and HLD predictions.

Retinal photograph-based deep learning predicts biological age, and stratifies morbidity and mortality risk.

BACKGROUND: ageing is an important risk factor for a variety of human pathologies. Biological age (BA) may better capture ageing-related physiological changes compared with chronological age (CA). OBJECTIVE: we developed a deep learning (DL) algorithm to predict BA based on retinal photographs and evaluated the performance of our new ageing marker in the risk stratification of mortality and major morbidity in general populations. METHODS: we first trained a DL algorithm using 129,236 retinal photographs from 40,480 participants in the Korean Health Screening study to predict the probability of age being ≥65 years ('RetiAGE') and then evaluated the ability of RetiAGE to stratify the risk of mortality and major morbidity among 56,301 participants in the UK Biobank. Cox proportional hazards model was used to estimate the hazard ratios (HRs). RESULTS: in the UK Biobank, over a 10-year follow up, 2,236 (4.0%) died; of them, 636 (28.4%) were due to cardiovascular diseases (CVDs) and 1,276 (57.1%) due to cancers. Compared with the participants in the RetiAGE first quartile, those in the RetiAGE fourth quartile had a 67% higher risk of 10-year all-cause mortality (HR = 1.67 [1.42-1.95]), a 142% higher risk of CVD mortality (HR = 2.42 [1.69-3.48]) and a 60% higher risk of cancer mortality (HR = 1.60 [1.31-1.96]), independent of CA and established ageing phenotypic biomarkers. Likewise, compared with the first quartile group, the risk of CVD and cancer events in the fourth quartile group increased by 39% (HR = 1.39 [1.14-1.69]) and 18% (HR = 1.18 [1.10-1.26]), respectively. The best discrimination ability for RetiAGE alone was found for CVD mortality (c-index = 0.70, sensitivity = 0.76, specificity = 0.55). Furthermore, adding RetiAGE increased the discrimination ability of the model beyond CA and phenotypic biomarkers (increment in c-index between 1 and 2%). CONCLUSIONS: the DL-derived RetiAGE provides a novel, alternative approach to measure ageing.

Circulating Metabolic Biomarkers Are Consistently Associated With Type 2 Diabetes Risk in Asian and European Populations.

CONTEXT: While Asians have a higher risk of type 2 diabetes (T2D) than Europeans for a given body mass index (BMI), it remains unclear whether the same markers of metabolic pathways are associated with diabetes. OBJECTIVE: We evaluated associations between metabolic biomarkers and incidence of T2D in 3 major Asian ethnic groups (Chinese, Malay, and Indian) and a European population. METHODS: We analyzed data from adult males and females of 2 cohorts from Singapore (n = 6393) consisting of Chinese, Malays, and Indians and 3 cohorts of European-origin participants from Finland (n = 14 558). We used nuclear magnetic resonance to quantify 154 circulating metabolic biomarkers at baseline and performed logistic regression to assess associations with T2D risk adjusted for age, sex, BMI and glycemic markers. RESULTS: Of the 154 metabolic biomarkers, 59 were associated with higher risk of T2D in both Asians and Europeans (P < 0.0003, Bonferroni-corrected). These included branched chain and aromatic amino acids, the inflammatory marker glycoprotein acetyls, total fatty acids, monounsaturated fatty acids, apolipoprotein B, larger very low-density lipoprotein particle sizes, and triglycerides. In addition, 13 metabolites were associated with a lower T2D risk in both populations, including omega-6 polyunsaturated fatty acids and larger high-density lipoprotein particle sizes. Associations were consistent within the Asian ethnic groups (all Phet ≥ 0.05) and largely consistent for the Asian and European populations (Phet ≥ 0.05 for 128 of 154 metabolic biomarkers). CONCLUSION: Metabolic biomarkers across several biological pathways were consistently associated with T2D risk in Asians and Europeans.

High-Density Lipoprotein 3 Cholesterol and Primary Open-Angle Glaucoma: Metabolomics and Mendelian Randomization Analyses.

PURPOSE: We hypothesized that the effect of blood lipid-related metabolites on primary open-angle glaucoma (POAG) would differ according to specific lipoprotein particles and lipid sub-fractions. We investigated the associations of blood levels of lipoprotein particles and lipid sub-fractions with POAG. DESIGN: Cross-sectional study. PARTICIPANTS: Individuals recruited for the baseline visit of the population-based Singapore Epidemiology of Eye Disease study (n = 8503). METHODS: All participants underwent detailed standardized ocular and systemic examinations. A total of 130 blood lipid-related metabolites were quantified using a nuclear magnetic resonance metabolomics platform. The analyses were conducted in 2 stages. First, we investigated whether and which lipid-related metabolites were directly associated with POAG using regression analyses followed by Bayesian network modeling. Second, we investigated if any causal relationship exists between the identified lipid-related metabolites, if any, and POAG using 2-sample Mendelian randomization (MR) analysis. We performed genome-wide association studies (GWAS) on high-density lipoprotein (HDL) 3 cholesterol (after inverse normal transformation) and used the top variants associated with HLD3 cholesterol as instrumental variables (IVs) in the MR analysis. MAIN OUTCOME MEASURE: Primary open-angle glaucoma. RESULTS: Of the participants, 175 (2.1%) had POAG. First, a logistic regression model showed that total HDL3 cholesterol (negatively) and phospholipids in very large HDL (positively) were associated with POAG. Further analyses using a Bayesian network analysis showed that only total HDL3 cholesterol was directly associated with POAG (odds ratio [OR], 0.72 per 1 standard deviation increase in HDL3 cholesterol; 95% confidence interval [CI], 0.61-0.84), independently of age, gender, intraocular pressure (IOP), body mass index (BMI), education level, systolic blood pressure, axial length, and statin medication. Using 5 IVs identified from the GWAS and with the inverse variance weighted MR method, we found that higher levels of HDL3 cholesterol were associated with a decreased odds of POAG (OR, 0.91; 95% CI, 0.84-0.99, P = 0.021). Other MR methods, including weighted median, mode-based estimator, and contamination mixture methods, derived consistent OR estimates. None of the routine lipids (blood total, HDL, or low-density lipoprotein [LDL] cholesterol) were associated with POAG. CONCLUSIONS: Overall, these results suggest that the relationship between HDL3 cholesterol and POAG might be causal and specific, and that dysregulation of cholesterol transport may play a role in the pathogenesis of POAG.

Machine learning to determine relative contribution of modifiable and non-modifiable risk factors of major eye diseases.

AIMS: To use machine learning (ML) to determine the relative contributions of modifiable and non-modifiable clinical, metabolic, genetic, lifestyle and socioeconomic factors on the risk of major eye diseases. METHODS: We conducted analyses in a cross-sectional multi-ethnic population-based study (n=10 033 participants) and determined a range of modifiable and non-modifiable risk factors of common eye diseases, including diabetic retinopathy (DR), non-diabetic-related retinopathy (NDR); early and late age-related macular degeneration (AMD); nuclear, cortical and posterior subcapsular (PSC) cataract; and primary open-angle (POAG) and primary angle-closure glaucoma (PACG). Risk factors included individual characteristics, metabolic profiles, genetic background, lifestyle patterns and socioeconomic status (n~100 risk factors). We used gradient boosting machine to estimate the relative influence (RI) of each risk factor. RESULTS: Among the range of risk factors studied, the highest contributions were duration of diabetes for DR (RI=22.1%), and alcohol consumption for NDR (RI=6.4%). For early and late AMD, genetic background (RI~20%) and age (RI~15%) contributed the most. Axial length was the main risk factor of PSC (RI=30.8%). For PACG, socioeconomic factor (mainly educational level) had the highest influence (20%). POAG was the disease with the highest contribution of modifiable risk factors (cumulative RI~35%), followed by PACG (cumulative RI ~30%), retinopathy (cumulative RI between 20% and 30%) and late AMD (cumulative RI ~20%). CONCLUSION: This study illustrates the utility of ML in identifying factors with the highest contributions. Risk factors possibly amenable to interventions were intraocular pressure (IOP) and Body Mass Index (BMI) for glaucoma, alcohol consumption for NDR and levels of HbA1c for DR.

Novel Serum and Urinary Metabolites Associated with Diabetic Retinopathy in Three Asian Cohorts.

Diabetic retinopathy (DR) is a microvascular complication of diabetes mellitus, a metabolic disorder, but understanding of its pathophysiology remains incomplete. Meta-analysis of three population-based cross-sectional studies (2004-11) representing three major Asian ethnic groups (aged 40-80 years: Chinese, 592; Malays, 1052; Indians, 1320) was performed. A panel of 228 serum/plasma metabolites and 54 urinary metabolites were quantified using nuclear magnetic resonance (NMR) spectroscopy. Main outcomes were defined as any DR, moderate/above DR, and vision-threatening DR assessed from retinal photographs. The relationship between metabolites and DR outcomes was assessed using multivariate logistic regression models, and metabolites significant after Bonferroni correction were meta-analyzed. Among serum/plasma metabolites, lower levels of tyrosine and cholesterol esters to total lipids ratio in IDL and higher levels of creatinine were positively associated with all three outcomes of DR (all p < 0.005). Among urinary metabolites, lower levels of citrate, ethanolamine, formate, and hypoxanthine were positively associated with all three DR outcomes (all p < 0.005). Higher levels of serum/plasma 3-hydroxybutyrate and lower levels of urinary 3-hydroxyisobutyrate were associated with VTDR. Comprehensive metabolic profiling in three large Asian cohorts with DR demonstrated alterations in serum/plasma and urinary metabolites mostly related to amino acids, lipoprotein subclasses, kidney function, and glycolysis.