A Prism Vote method for individualized risk prediction of traits in genotype data of Multi-population.

Multi-population cohorts offer unprecedented opportunities for profiling disease risk in large samples, however, heterogeneous risk effects underlying complex traits across populations make integrative prediction challenging. In this study, we propose a novel Bayesian probability framework, the Prism Vote (PV), to construct risk predictions in heterogeneous genetic data. The PV views the trait of an individual as a composite risk from subpopulations, in which stratum-specific predictors can be formed in data of more homogeneous genetic structure. Since each individual is described by a composition of subpopulation memberships, the framework enables individualized risk characterization. Simulations demonstrated that the PV framework applied with alternative prediction methods significantly improved prediction accuracy in mixed and admixed populations. The advantage of PV enlarges as genetic heterogeneity and sample size increase. In two real genome-wide association data consists of multiple populations, we showed that the framework considerably enhanced prediction accuracy of the linear mixed model in five-group cross validations. The proposed method offers a new aspect to analyze individual's disease risk and improve accuracy for predicting complex traits in genotype data.

Evaluating the impact of sugar-sweetened beverages tax on overweight, obesity, and type 2 diabetes in an affluent Asian setting: A willingness-to-pay survey and simulation analysis.

BACKGROUND: The potential health effects of taxing sugar-sweetened beverages (SSBs) has been insufficiently examined in Asian contexts. This study aimed to assess the impact of SSB taxation on the prevalence of obesity/overweight and type 2 diabetes mellitus (T2DM) in Hong Kong using a willingness-to-pay (WTP) survey and simulation analysis. METHODS: A random telephone survey was conducted with 1000 adults from May to June 2020. We used a contingent valuation approach to assess individuals' WTP for SSBs under four tax payment scenarios (5%, 10%, 40%, and 50% of the current market price). Based on the WTP, a simulation analysis was conducted to project changes in SSB purchase and associated reductions in the prevalence of obesity/overweight and T2DM over a 10-year simulation period. FINDINGS: When 5% and 10% taxation rates were introduced, approximately one-third of the population were unwilling to maintain their SSB purchase. Our simulation demonstrated a gradual decline in the prevalence of obesity/overweight and diabetes with a more pronounced decrease when higher taxation rates were introduced. 10% taxation resulted in a mean reduction of 1532.7 cases of overweight/obesity per 100 thousand population at the sixth year, while T2DM prevalence decreased by 267.1 (0.3%). CONCLUSIONS: This study underscores the effects of an SSB tax on purchase behaviors and health outcomes in an affluent Asia setting, with a more pronounced influence on adult population. These findings are expected to inform policymakers in making decisions regarding an effective and equitable tax rate on SSBs.

Long-term effectiveness of elderly health care voucher scheme strategies: a system dynamics simulation analysis.

BACKGROUND: The elderly healthcare voucher (EHCV) scheme is expected to lead to an increase in the number of elderly people selecting private primary healthcare services and reduce reliance on the public sector in Hong Kong. However, studies thus far have reported that this scheme has not received satisfactory responses. In this study, we examined changes in the ratio of visits between public and private doctors in primary care (to measure reliance on the public sector) for different strategic scenarios in the EHCV scheme. METHODS: Based on comments from an expert panel, a system dynamics model was formulated to simulate the impact of various enhanced strategies in the scheme: increasing voucher amounts, lowering the age eligibility, and designating vouchers for chronic conditions follow-up. Data and statistics for the model calibration were collected from various sources. RESULTS: The simulation results show that the current EHCV scheme is unable to reduce the utilization of public healthcare services, as well as the ratio of visits between public and private primary care among the local aging population. When comparing three different tested scenarios, even if the increase in the annual voucher amount could be maintained at the current pace or the age eligibility can be lowered to include those aged 60 years, the impact on shifts from public-to-private utilization were insignificant. The public-to-private ratio could only be marginally reduced from 0.74 to 0.64 in the first several years. Nevertheless, introducing a chronic disease-oriented voucher could result in a significant drop of 0.50 in the public-to-private ratio during the early implementation phase. However, the effect could not be maintained for an extended period. CONCLUSIONS: Our findings will assist officials in improving the design of the EHCV scheme, within the wider context of promoting primary care among the elderly. We suggest that an additional chronic disease-oriented voucher can serve as an alternative strategy. The scheme must be redesigned to address more specific objectives or provide a separate voucher that promotes under-utilized healthcare services (e.g., preventive care), instead of services designed for unspecified reasons, which may lead to concerns regarding exploitation.

Epidemiological changes in measles infections in southern China between 2009 and 2016: a retrospective database analysis.

BACKGROUND: The incidence rate of measles in China reached a nadir in 2012 after 2 supplementary immunization activities (SIAs) were undertaken in 2009 and 2010. However, the disease began re-emerging in 2013, with a high prevalence rate observed in 2013-2014 in the southern province of Guangdong. In this study, we assessed the changes that occurred in measles epidemiology during 2009-2016, particularly between 2009 and 2011 (when the influence of the SIAs were in full effect) and between 2012 and 2016 (when this influence subsided). METHODS: Data from 22,362 patients with measles diagnosed between 2009 and 2016, and whose diagnoses were confirmed clinically and/or with laboratory testing, were extracted from the National Infectious Disease Monitoring Information System. Descriptive analyses were performed, and changes in epidemiological characteristics between 2009 and 2011 and 2012-2016 were compared. RESULTS: There was a substantial surge in 0-8-month-old patients after 2012; the incidence rate increased from 4.0 per 100,000 population in 2011 (10.3% of the total) to 280 per 100,000 population in 2013 (32.8% of the total). Patients aged 0-6 years represented 73.4% of the total increase between 2011 and 2013. Compared with 2009-2011, adults aged ≥25 years accounted for a higher proportion of patients in 2013 and after (p < 0.01), and were highest in 2016 (31% of the patient total). CONCLUSION: Despite the remarkable results achieved by SIAs in terms of providing herd immunity, the 2013 resurgence of measles revealed insufficient immunization coverage among children. Therefore routine immunization programs should be strengthened, and supplementary vaccinations targeting adults should also be contemplated.

Effects of Kinesio tape on lower limb muscle strength, hop test, and vertical jump performances: a meta-analysis.

BACKGROUND: To date, published systematic reviews concerning the effects of Kinesio Taping (KT) on muscle strength have not analysed facilitatory and inhibitory applications separately. As a result, their results could be substantially affected by clinical heterogeneity. This meta-analysis was conducted to determine the effectiveness of using a facilitatory application of KT for lower limb muscle strength and functional performance (distance in a single-leg hop and vertical jump height) in individuals without disabilities and in those with musculoskeletal conditions (muscle fatigue, chronic musculoskeletal diseases, and post-operative orthopaedic conditions). METHODS: Searches were conducted on six major electronic databases. Randomised controlled trials that used facilitatory KT were included. Standardised mean differences (SMDs) were calculated and random-effects models were used for analysis. RESULTS: Thirty-seven randomised controlled trials were included. KT was superior to controls for improving lower limb muscle strength in individuals with muscle fatigue (short-term effect, pooled SMD = 0.53, 95% CI = 0.09 to 0.96; long-term effect, pooled SMD = 0.61, 95% CI = 0.12 to 1.11) and in individuals with chronic musculoskeletal diseases (pooled SMD = 1.24, 95% CI = 0.33 to 2.16) with large effect sizes. The use of KT in populations without disabilities was not supported. There is insufficient evidence for the effect of KT on functional performance in individuals with musculoskeletal conditions. CONCLUSIONS: Contrary to prior research, the existing evidence shows that KT can improve lower limb muscle strength in individuals with muscle fatigue and chronic musculoskeletal diseases. The effect sizes produced in this meta-analysis show that KT may be superior to some existing treatments for these conditions. In addition, this study suggests that practitioners may wish to avoid the use of KT in individuals without disabilities. TRIAL REGISTRATION: PROSPERO registration number CRD42017075490 , registered on 21 November 2017.

A Zoom-Focus algorithm (ZFA) to locate the optimal testing region for rare variant association tests.

MOTIVATION: Increasing amounts of whole exome or genome sequencing data present the challenge of analysing rare variants with extremely small minor allele frequencies. Various statistical tests have been proposed, which are specifically configured to increase power for rare variants by conducting the test within a certain bin, such as a gene or a pathway. However, a gene may contain from several to thousands of markers, and not all of them are related to the phenotype. Combining functional and non-functional variants in an arbitrary genomic region could impair the testing power. RESULTS: We propose a Zoom-Focus algorithm (ZFA) to locate the optimal testing region within a given genomic region. It can be applied as a wrapper function in existing rare variant association tests to increase testing power. The algorithm consists of two steps. In the first step, Zooming, a given genomic region is partitioned by an order of two, and the best partition is located. In the second step, Focusing, the boundaries of the zoomed region are refined. Simulation studies showed that ZFA substantially increased the statistical power of rare variants' tests, including the SKAT, SKAT-O, burden test and the W-test. The algorithm was applied on real exome sequencing data of hypertensive disorder, and identified biologically relevant genetic markers to metabolic disorders that were undetectable by a gene-based method. The proposed algorithm is an efficient and powerful tool to enhance the power of association study for whole exome or genome sequencing data. AVAILABILITY AND IMPLEMENTATION: The ZFA software is available at: http://www2.ccrb.cuhk.edu.hk/statgene/software.html. CONTACT: maggiew@cuhk.edu.hk or bzee@cuhk.edu.hk. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Incorporating methylation genome information improves prediction accuracy for drug treatment responses.

BACKGROUND: An accumulation of evidence has revealed the important role of epigenetic factors in explaining the etiopathogenesis of human diseases. Several empirical studies have successfully incorporated methylation data into models for disease prediction. However, it is still a challenge to integrate different types of omics data into prediction models, and the contribution of methylation information to prediction remains to be fully clarified. RESULTS: A stratified drug-response prediction model was built based on an artificial neural network to predict the change in the circulating triglyceride level after fenofibrate intervention. Associated single-nucleotide polymorphisms (SNPs), methylation of selected cytosine-phosphate-guanine (CpG) sites, age, sex, and smoking status, were included as predictors. The model with selected SNPs achieved a mean 5-fold cross-validation prediction error rate of 43.65%. After adding methylation information into the model, the error rate dropped to 41.92%. The combination of significant SNPs, CpG sites, age, sex, and smoking status, achieved the lowest prediction error rate of 41.54%. CONCLUSIONS: Compared to using SNP data only, adding methylation data in prediction models slightly improved the error rate; further prediction error reduction is achieved by a combination of genome, methylation genome, and environmental factors.

A fast and powerful W-test for pairwise epistasis testing.

Epistasis plays an essential role in the development of complex diseases. Interaction methods face common challenge of seeking a balance between persistent power, model complexity, computation efficiency, and validity of identified bio-markers. We introduce a novel W-test to identify pairwise epistasis effect, which measures the distributional difference between cases and controls through a combined log odds ratio. The test is model-free, fast, and inherits a Chi-squared distribution with data adaptive degrees of freedom. No permutation is needed to obtain the P-values. Simulation studies demonstrated that the W-test is more powerful in low frequency variants environment than alternative methods, which are the Chi-squared test, logistic regression and multifactor-dimensionality reduction (MDR). In two independent real bipolar disorder genome-wide associations (GWAS) datasets, the W-test identified significant interactions pairs that can be replicated, including SLIT3-CENPN, SLIT3-TMEM132D, CNTNAP2-NDST4 and CNTCAP2-RTN4R The genes in the pairs play central roles in neurotransmission and synapse formation. A majority of the identified loci are undiscoverable by main effect and are low frequency variants. The proposed method offers a powerful alternative tool for mapping the genetic puzzle underlying complex disorders.

Stratified polygenic risk prediction model with application to CAGI bipolar disorder sequencing data.

Genetic data consists of a wide range of marker types, including common, low-frequency, and rare variants. Multiple genetic markers and their interactions play central roles in the heritability of complex disease. In this study, we propose an algorithm that uses a stratified variable selection design by genetic architectures and interaction effects, achieved by a dataset-adaptive W-test. The polygenic sets in all strata were integrated to form a classification rule. The algorithm was applied to the Critical Assessment of Genome Interpretation 4 bipolar challenge sequencing data. The prediction accuracy was 60% using genetic markers on an independent test set. We found that epistasis among common genetic variants contributed most substantially to prediction precision. However, the sample size was not large enough to draw conclusions for the lack of predictability of low-frequency variants and their epistasis.

A W-test collapsing method for rare-variant association testing in exome sequencing data.

Advancement in sequencing technology enables the study of association between complex disorder phenotypes and single-nucleotide polymorphisms with rare mutations. However, the rare genetic variant has extremely small variance and impairs testing power of traditional statistical methods. We introduce a W-test collapsing method to evaluate rare-variant association by measuring the distributional differences between cases and controls through combined log of odds ratio within a genomic region. The method is model-free and inherits chi-squared distribution with degrees of freedom estimated from bootstrapped samples of the data, and allows for fast and accurate P-value calculation without the need of permutations. The proposed method is compared with the Weighted-Sum Statistic and Sequence Kernel Association Test on simulation datasets, and showed good performances and significantly faster computing speed. In the application of real next-generation sequencing dataset of hypertensive disorder, it identified genes of interesting biological functions associated to metabolism disorder and inflammation, including the MACROD1, NLRP7, AGK, PAK6, and APBB1. The proposed method offers an efficient and effective way for testing rare genetic variants in whole exome sequencing datasets.

A statistical method utilizing information of imported cases to estimate the transmissibility for an influenza pandemic.

BACKGROUND: In a new influenza pandemic, travel data such as arrival times of cases seeded by the originating country can be regarded as a combination of the epidemic size and the mobility networks of infections connecting the originating country with other regions. It can be a complete and timely source for estimating the basic reproduction number (R 0 ), a key indicator of disease transmissibility. METHOD: In this study, we developed a likelihood-based method using arrival times of infected cases in different countries to estimate R 0 for influenza pandemics. A simulation was conducted to assess the performance of the proposed method. We further applied the method to the outbreak of the influenza pandemic A/H1N1 in Mexico. RESULTS: In the numerical application, the estimated R 0 was equal to 1.69 with a 95% confidence interval (1.65, 1.73). For the simulation results, the estimations were robust to the decline of travel rate and other parameter assumptions. Nevertheless, the estimates were moderately sensitive to the assumption of infectious duration. Generally, the findings were in line with other relevant studies. CONCLUSIONS: Our approach as well as the estimate is potential to assist officials in planning control and prevention measures. Improved coordination to streamline or even centralize surveillance of imported cases among countries will thus be beneficial to public health.

Prospective Association between Diabetic Retinopathy and Cardiovascular Disease-A Systematic Review and Meta-analysis of Cohort Studies.

BACKGROUND: Diabetic retinopathy (DR) is linked to increased risk of cardiovascular (CV) disease. However, the effect size of the association was not consistent. In this study, we performed a systematic review and meta-analysis of available cohort studies to determine the association between DR and CV disease, and to investigate the factors that influence the association. METHODS: Terms related to DR and CV disease were searched from MEDLINE and EMBASE database. High-quality articles (Newcastle-Ottawa scales above 6) conducted in cohort studies reporting the association between DR and CV disease were identified. Study-specific estimates were pooled using random effects with inverse variance meta-analysis. Subgroup analysis was performed according to diabetes types. Heterogeneity of included studies was assessed using the I(2) test. The cause of the heterogeneity was examined using metaregression analyses. RESULTS: A total of 13 studies representing 17,611 patients without CV disease at baseline were included. At follow-up, there were 1457 CV disease-related incidences. Overall, DR was associated with increased risk of CV disease (relative risk [RR]: 2.42, 95% confidence interval [CI]: 1.77-3.31) in diabetes. Specifically, the RR was 3.59 (95% CI: 1.79-7.20) for type 1 diabetes and 1.81 (95% CI: 1.47-2.23) for type 2 diabetes. Significant heterogeneity was found in studies with type 1 diabetes. Metaregression analysis showed that baseline systolic blood pressure was a key factor leading to the heterogeneity. CONCLUSION: In conclusion, DR is significantly associated with CV disease incidence and CV disease-related mortality in diabetes. Patients with DR may need more intensive management to control future CV disease attacks.

The diminishing socioeconomic disparity in obesity in a Chinese population with rapid economic development: analysis of serial cross-sectional health survey data 2002-2010.

BACKGROUND: Social changes and economic development are associated with obesity epidemic. This study is to investigate the trends of socio-economic disparity in obesity from 2002 to 2010 in a Chinese population experiencing the world's fastest economic development. METHODS: Four standardized surveys were conducted in a population of 85 million residents in Guangdong, China between 2002 and 2010. Multistage random cluster sampling was used to recruit representative samples. Information on socio-economic status (SES), proxied by education, occupation and residential area, was collected by face-to-face interviews. The weight, height and waist circumference of the participants were also measured. RESULTS: Women with low education had an increased BMI of 0.85 kg/m(2), while women with high education had a decreased BMI of 0.16 kg/m(2) (p = 0.032 for interaction test). Similar trends were observed by using occupation and residential area as the SES indicators. Analysis in men yielded similar patterns. Waist circumference increased from 73.7 to 78.4 cm, and the increasing trends of statistical significance (p < 0.01) were observed in both genders and across all SES levels, with the magnitudes of increase in low SES levels being more pronounced. The impact of gradient in food attainment and occupational physical activity across the SES levels may wear off with further economic development, while factors such as health awareness, diet pattern and leisure activity may become increasingly important in driving the disparity. CONCLUSION: The impact of gradient in food attainment and occupational physical activity across the SES levels may wear off with further economic development, while factors such as health awareness, diet pattern and leisure activity may become increasingly important in driving the disparity. Our findings suggest that health education should focus on the disadvantaged populations on health awareness for adopting healthier diet pattern and increasing physical activity.

Projection of dengue fever transmissibility under climate change in South and Southeast Asian countries.

Vector-borne infectious disease such as dengue fever (DF) has spread rapidly due to more suitable living environments. Considering the limited studies investigating the disease spread under climate change in South and Southeast Asia, this study aimed to project the DF transmission potential in 30 locations across four South and Southeast Asian countries. In this study, weekly DF incidence data, daily mean temperature, and rainfall data in 30 locations in Singapore, Sri Lanka, Malaysia, and Thailand from 2012 to 2020 were collected. The effects of temperature and rainfall on the time-varying reproduction number (Rt) of DF transmission were examined using generalized additive models. Projections of location-specific Rt from 2030s to 2090s were determined using projected temperature and rainfall under three Shared Socioeconomic Pathways (SSP126, SSP245, and SSP585), and the peak DF transmissibility and epidemic duration in the future were estimated. According to the results, the projected changes in the peak Rt and epidemic duration varied across locations, and the most significant change was observed under middle-to-high greenhouse gas emission scenarios. Under SSP585, the country-specific peak Rt was projected to decrease from 1.63 (95% confidence interval: 1.39-1.91), 2.60 (1.89-3.57), and 1.41 (1.22-1.64) in 2030s to 1.22 (0.98-1.51), 2.09 (1.26-3.47), and 1.37 (0.83-2.27) in 2090s in Singapore, Thailand, and Malaysia, respectively. Yet, the peak Rt in Sri Lanka changed slightly from 2030s to 2090s under SSP585. The epidemic duration in Singapore and Malaysia was projected to decline under SSP585. In conclusion, the change of peak DF transmission potential and disease outbreak duration would vary across locations, particularly under middle-to-high greenhouse gas emission scenarios. Interventions should be considered to slow down global warming as well as the potential increase in DF transmissibility in some locations of South and Southeast Asia.

Predictive evolutionary modelling for influenza virus by site-based dynamics of mutations.

Influenza virus continuously evolves to escape human adaptive immunity and generates seasonal epidemics. Therefore, influenza vaccine strains need to be updated annually for the upcoming flu season to ensure vaccine effectiveness. We develop a computational approach, beth-1, to forecast virus evolution and select representative virus for influenza vaccine. The method involves modelling site-wise mutation fitness. Informed by virus genome and population sero-positivity, we calibrate transition time of mutations and project the fitness landscape to future time, based on which beth-1 selects the optimal vaccine strain. In season-to-season prediction in historical data for the influenza A pH1N1 and H3N2 viruses, beth-1 demonstrates superior genetic matching compared to existing approaches. In prospective validations, the model shows superior or non-inferior genetic matching and neutralization against circulating virus in mice immunization experiments compared to the current vaccine. The method offers a promising and ready-to-use tool to facilitate vaccine strain selection for the influenza virus through capturing heterogeneous evolutionary dynamics over genome space-time and linking molecular variants to population immune response.

Impact of climate change on dengue fever epidemics in South and Southeast Asian settings: A modelling study.

The potential for dengue fever epidemic due to climate change remains uncertain in tropical areas. This study aims to assess the impact of climate change on dengue fever transmission in four South and Southeast Asian settings. We collected weekly data of dengue fever incidence, daily mean temperature and rainfall from 2012 to 2020 in Singapore, Colombo, Selangor, and Chiang Mai. Projections for temperature and rainfall were drawn for three Shared Socioeconomic Pathways (SSP126, SSP245, and SSP585) scenarios. Using a disease transmission model, we projected the dengue fever epidemics until 2090s and determined the changes in annual peak incidence, peak time, epidemic size, and outbreak duration. A total of 684,639 dengue fever cases were reported in the four locations between 2012 and 2020. The projected change in dengue fever transmission would be most significant under the SSP585 scenario. In comparison to the 2030s, the peak incidence would rise by 1.29 times in Singapore, 2.25 times in Colombo, 1.36 times in Selangor, and >10 times in Chiang Mai in the 2090s under SSP585. Additionally, the peak time was projected to be earlier in Singapore, Colombo, and Selangor, but be later in Chiang Mai under the SSP585 scenario. Even in a milder emission scenario of SSP126, the epidemic size was projected to increase by 5.94%, 10.81%, 12.95%, and 69.60% from the 2030s-2090s in Singapore, Colombo, Selangor, and Chiang Mai, respectively. The outbreak durations in the four settings were projected to be prolonged over this century under SSP126 and SSP245, while a slight decrease is expected in 2090s under SSP585. The results indicate that climate change is expected to increase the risk of dengue fever transmission in tropical areas of South and Southeast Asia. Limiting greenhouse gas emissions could be crucial in reducing the transmission of dengue fever in the future.

Spatiotemporal relationship between temperature and non-accidental mortality: Assessing effect modification by socioeconomic status.

BACKGROUND: Most published studies have assessed the overall health impact of temperature by using one-station or multiple-station averaged meteorological and air quality data. Concern has arisen about whether the temperature health impact is homogeneous across the whole territory geographically, since green space and socioeconomic factors may modify the impact. OBJECTIVE: This study aims at investigating how small-area mortality is modified by local temperature and other meteorological, air quality, green space, and socioeconomic factors of small geographic units in a subtropical urban setting. METHODS: Data on meteorological, air pollutants, and non-accidental mortality count in Hong Kong during 2006-2016 were obtained. Combined with green space and socioeconomic data, spatiotemporal analysis using Generalized Additive Mixed Models was conducted to examine the temperature-mortality relationship, adjusted for seasonality, long-term trend, other meteorological factors, pollutants, socioeconomic characteristics and green space. RESULTS: Socioeconomic status was found to modify the temporal temperature-mortality relationship. A J-shape association was identified for most areas in Hong Kong, where a sharp increase of mortality was observed when daily minimum temperature dropped lower than the turning point. However, for people living in the most affluent areas, after the initial increase there was a decrease of mortality for colder days. Besides, when comparing the two spatiotemporal models (i.e. using nearby or central temperature monitoring station), while leaving the other predictors unchanged, this study showed that there was little difference in the overall model performances. CONCLUSION: This study indicated that the daily fluctuation of mortality was associated with daily temperature, while the spatial variation of mortality within this city could be explained by the geographical distribution of green space and socioeconomic factors. Since people living in affluent areas were found to be more tolerant of cold temperatures, it would be more efficient to tailor cold temperature health education and warning information for socioeconomically deprived communities.

Health Disparity Resulting from the Effect of Built Environment on Temperature-Related Mortality in a Subtropical Urban Setting.

Whereas previous studies have assessed the overall health impact of temperature in Hong Kong, the aim of this study was to investigate whether the health impact is modified by local temperature of small geographic units, which may be related to the diverse socioeconomic characteristics of these units. The effects of local temperature on non-accidental and cause-specific mortality were analyzed using Bayesian spatial models at a small-area level, adjusting for potential confounders, i.e., area-level air pollutants, socioeconomic status, and green space, as well as spatial dependency. We found that a 10% increase in green space density was associated with an estimated 4.80% decrease in non-accidental mortality risk and a 5.75% decrease in cardiovascular disease mortality risk in Hong Kong, whereas variation in local annual temperature did not significantly contribute to mortality. We also found that the spatial variation of mortality within this city could be explained by the geographic distribution of green space and socioeconomic factors rather than local temperature or air pollution. The findings and methodology of this study may help to further understanding and investigation of social and structural determinants of health disparities, particularly place-based built environment across class-based small geographic units in a city, taking into account the intersection of multiple factors from individual to population levels.

Risk Assessment of CHD Using Retinal Images with Machine Learning Approaches for People with Cardiometabolic Disorders.

BACKGROUND: Coronary heart disease (CHD) is the leading cause of death worldwide, constituting a growing health and social burden. People with cardiometabolic disorders are more likely to develop CHD. Retinal image analysis is a novel and noninvasive method to assess microvascular function. We aim to investigate whether retinal images can be used for CHD risk estimation for people with cardiometabolic disorders. METHODS: We have conducted a case-control study at Shenzhen Traditional Chinese Medicine Hospital, where 188 CHD patients and 128 controls with cardiometabolic disorders were recruited. Retinal images were captured within two weeks of admission. The retinal characteristics were estimated by the automatic retinal imaging analysis (ARIA) algorithm. Risk estimation models were established for CHD patients using machine learning approaches. We divided CHD patients into a diabetes group and a non-diabetes group for sensitivity analysis. A ten-fold cross-validation method was used to validate the results. RESULTS: The sensitivity and specificity were 81.3% and 88.3%, respectively, with an accuracy of 85.4% for CHD risk estimation. The risk estimation model for CHD with diabetes performed better than the model for CHD without diabetes. CONCLUSIONS: The ARIA algorithm can be used as a risk assessment tool for CHD for people with cardiometabolic disorders.