Absolute Rank Research Articles

Newly estimated disability weights for 196 health states in Hubei Province, China

BackgroundThe disability weight (DW) reflects the severity of non-fatal outcomes and is an important parameter in calculating the burden of disease. However, the universality of the global, national, or subnational DWs remains controversial. This study aims to measure DWs specific to Hubei Province of China using non-parametric regression to anchor the DWs.MethodsPaired comparison (PC) data collected from a web-based survey in Hubei Province targeting the general population were used to estimate the DWs of 196 health states. Specifically, PC data from 33,925 respondents were analyzed by probit regression analysis, and the results were then anchored to 0–1 scale using non-parametric regression based on the DWs from Global Burden of Disease (GBD) 2013. The absolute DW values and rankings were compared to those in the Chinese disability weight measurement study, GBD 2013, and Japan.ResultsThe DWs for 196 health states ranged from 0.003 for mild distance vision impairment to 0.663 for severe heroin and opioid dependence in Hubei Province, China. Quite a lot mental disorders, such as moderate/severe episode of major depressive disorder, were considered more severe than the terminal phase with/without medication among Hubei residents. DW rankings of the health states are relatively stable in Hubei Province irrespective of the anchoring method used. A very small proportion (4 of 196, 2%) of DW rankings changed by 10 or more positions in China when compared with our results, but approximately 61% in GBD 2013 and 59% in Japan. Among the top 25 health states in this study, 9 of 11 health states categorized as mental, behavioral, and substance use disorders resulted in a lower ranking in GBD 2013, and all 6 states in Japan also showed a lower ranking, whereas China shared a similar ranking.ConclusionsThe burden of mental disorders among Hubei residents, especially moderate or severe major depressive disorder, deserves further attention. When using different anchoring methods, DW rankings were maintained relatively stable rather than the absolute values in Hubei. Substantial differences of DW rankings between our results and that in China, GBD 2013, and Japan draw attention to the need for deriving local disability for disease burden calculation.

Determining method of optimal process parameters using Taguchi method and technique for order preference by similarity to ideal solution in plastic injection molding

Many plastic injection molding (PIM) process optimization problems are multi-objective optimization problems that optimize multiple mechanical properties. This paper proposed a determining method of optimal process parameters and their effect ranks using Taguchi method and TOPSIS in the PIM. TOPSIS was used to convert multiple mechanical properties into single comprehensive response. It was selected as a reasonable multi-attribute decision-making (MADM) method from among some well-known MADM methods based on mean rank correlation coefficient and mean absolute rank deviation. Taguchi method was used to design experiment and find the optimal process parameters. The proposed method was applied to determine the optimal values of the process parameters: melt temperature (MT), packing pressure (PP), cooling time (CT), and injection pressure (IP) for improving tensile strength, elasticity module, flexural modulus, and impact strength in the PIM with Acrylonitrile-Butadiene-Styrene compound as plastic materials and AISI 1020 as mold materials. The optimal values of the process parameters were MT of 280°C, PP of 28 MPa, CT of 22 s, and IP of 50 MPa, and their effect ranking was IP (51.941%), PP (32.280%), CT (9.045%), and MT (6.734%). The method could be widely applied to not only the PIM process optimization but also various manufacturing process optimization problems.

Assessing the limitations of relief-based algorithms in detecting higher-order interactions

BackgroundEpistasis, the interaction between genetic loci where the effect of one locus is influenced by one or more other loci, plays a crucial role in the genetic architecture of complex traits. However, as the number of loci considered increases, the investigation of epistasis becomes exponentially more complex, making the selection of key features vital for effective downstream analyses. Relief-Based Algorithms (RBAs) are often employed for this purpose due to their reputation as “interaction-sensitive” algorithms and uniquely non-exhaustive approach. However, the limitations of RBAs in detecting interactions, particularly those involving multiple loci, have not been thoroughly defined. This study seeks to address this gap by evaluating the efficiency of RBAs in detecting higher-order epistatic interactions. Motivated by previous findings that suggest some RBAs may rank predictive features involved in higher-order epistasis negatively, we explore the potential of absolute value ranking of RBA feature weights as an alternative approach for capturing complex interactions. In this study, we assess the performance of ReliefF, MultiSURF, and MultiSURFstar on simulated genetic datasets that model various patterns of genotype-phenotype associations, including 2-way to 5-way genetic interactions, and compare their performance to two control methods: a random shuffle and mutual information.ResultsOur findings indicate that while RBAs effectively identify lower-order (2 to 3-way) interactions, their capability to detect higher-order interactions is significantly limited, primarily by large feature count but also by signal noise. Specifically, we observe that RBAs are successful in detecting fully penetrant 4-way XOR interactions using an absolute value ranking approach, but this is restricted to datasets with only 20 total features.ConclusionsThese results highlight the inherent limitations of current RBAs and underscore the need for the development of Relief-based approaches with enhanced detection capabilities for the investigation of epistasis, particularly in datasets with large feature counts and complex higher-order interactions.

Assessing the Limitations of Relief-Based Algorithms in Detecting Higher-Order Interactions.

The investigation of epistasis becomes increasingly complex as more loci are considered due to the exponential expansion of possible interactions. Consequently, selecting key features that influence epistatic interactions is crucial for effective downstream analyses. Recognizing this challenge, this study investigates the efficiency of Relief-Based Algorithms (RBAs) in detecting higher-order epistatic interactions, which may be critical for understanding the genetic architecture of complex traits. RBAs are uniquely non-exhaustive, eliminating the need to construct features for every possible interaction and thus improving computational tractability. Motivated by previous research indicating that some RBAs rank predictive features involved in higher-order epistasis as highly negative, we explore the utility of absolute value ranking of RBA feature weights as an alternative method to capture complex interactions. We evaluate ReliefF, MultiSURF, and MultiSURFstar on simulated genetic datasets that model various patterns of genotype-phenotype associations, including 2-way to 5-way genetic interactions, and compare their performance to two control methods: a random shuffle and mutual information. Our findings indicate that while RBAs effectively identify lower-order (2 to 3-way) interactions, their capability to detect higher-order interactions is significantly limited, primarily by large feature count but also by signal noise. Specifically, we observe that RBAs are successful in detecting fully penetrant 4-way XOR interactions using an absolute value ranking approach, but this is restricted to datasets with a minimal number of total features. These results highlight the inherent limitations of current RBAs and underscore the need for enhanced detection capabilities for the investigation of epistasis, particularly in datasets with large feature counts and complex higher-order interactions.

Absolute income mobility obscures marginalized children’s disadvantages

Intergenerational mobility captures the distance between the socioeconomic positions of parents versus their adult children. Researchers measure this distance in absolute and relative units, such as absolute dollars and relative ranks. Absolute and relative mobility often diverge. For example, absolute mobility can rise while relative mobility declines. How should scholars and policymakers understand this divergence? We conclude that they should understand it as follows: absolute mobility is less reflective than relative mobility of marginalized children's socioeconomic disadvantages. We base this conclusion on analyses of survey, administrative, and simulated data on income mobility in the contemporary United States. We analyze multiple points of difference in mobility, which facilitates the recognition of several asymmetries. First, high-income children's experiences weigh more heavily in absolute-mobility trends than low-income children's experiences, particularly when economic growth is positive. Second, this asymmetry is more characteristic of absolute- than relative-mobility trends. Third, absolute-mobility differences across demographic groups are more prone than relative-mobility differences to obscure marginalized groups' socioeconomic disadvantages. These asymmetries have policy implications: We caution that focusing on absolute mobility as a policy target can divert attention away from society's most disadvantaged children.

Relationship Between National Residency Matching Program (NRMP) Rank Order and Otolaryngology Residency Performance.

The process of resident recruitment is costly, and our surgical residency program expends significant time on the resident selection process while balancing general duties and responsibilities. The aim of our study was to explore the relationship between otolaryngology-head and surgery (OHNS) residents' National Residency Matching Program (NRMP) rank-list position at our institution and their subsequent residency performance. Retrospective cohort study. Single site institution. We retrospectively reviewed 7 consecutive resident classes (2011-2017) at a single tertiary OHNS residency program. We reviewed each resident's absolute rank order in the NRMP matches. Measures of residency performance included overall faculty evaluation during postgraduate year 5 (PGY5), annual in-service examination scores (scaled score), and the number of manuscripts published in peer-reviewed journals. Correlations between NRMP rank order and subsequent residency performance were assessed using Spearman's rho correlation coefficients (ρ). Twenty-eight residents entered residency training between 2011 and 2017. The average rank position of the trainees during this study was 9.7 (range: 1-22). We found no significant correlation between rank order and faculty evaluation during PGY5 (ρ = 0.097, P = .625) or number of publications (ρ = -0.256, P = .189). Additionally, when assessing the association between rank order and annual Otolaryngology Training Examination-scaled scores, no statistically significant relationship was found between the 2 (P > .05). Our results showed that there were no significant correlations between OHNS rank order and various measures of success in residency training, which aligns with existing literature. Further investigation of this relationship should be conducted to ensure the applicability of our findings.

Comparative Evaluation of Crash Hotspot Identification Methods: Empirical Bayes vs. Potential for Safety Improvement Using Variants of Negative Binomial Models

The empirical Bayes (EB) method is widely acclaimed for crash hotspot identification (HSID), which integrates crash prediction model estimates and observed crash frequency to compute the expected crash frequency of a site. The traditional negative binomial (NB) models, often used to estimate crash predictive models, typically struggle with accounting for the unobserved heterogeneity in crash data. Complex extensions of the NB models are applied to overcome these shortcomings. These techniques also present new challenges, for instance, applying the EB procedures, especially for out-of-sample data. This study applies a random parameter negative binomial (RPNB) model within the EB framework for HSID using out-of-sample data, comparing its performance with a varying dispersion parameter NB model (VDPNB). The research also evaluates the potential for safety improvement (PSI) scores for both models and compares them with EB estimates using three generalised criteria: high crashes consistency test (HCCT), common sites consistency test (CSCT), and absolute rank differences test (ARDT). The results yield dual insights. Firstly, the study highlights associations between crash covariates and frequency, emphasising the significance of roadway geometric design characteristics (e.g., lane width, number of lanes, and parking type) and traffic volume. Some variables also influenced overdispersion parameters in the VDPNB model. In the RPNB model, annual average daily traffic (AADT) and lane width emerged as random parameters. Secondly, the HSID performance assessment revealed the superiority of the EB method over PSI. Notably, the RPNB model, compared to the VDPNB, demonstrates superior performance in EB estimates for HSID with out-of-sample data. This research recommends adopting the EB method with RPNB models for robust HSID.

The use of Google community mobility reports to model residential waste generation behaviors during and after the COVID-19 lockdown

This paper proposed an original analytical approach to quantify waste generation and recycling behaviors in Regina, a Canadian capital city, using Google's Community Mobility Reports (GCMR) mobility data and ANN modeling. Residential waste collection rate (RWCR) in Regina appears to have changed abruptly during the pandemic. Compared to the pre-COVID-19 lockdown period, the correlation between the percent changes in RWCR and GCMR mobility became increasingly stronger and statistically significant during and after COVID-19 lockdown. It appears that the government imposed public health measures improved the consistency of waste generation and recycling behaviors. All 16 models predicting RWCR performed satisfactorily, with R2 near or over 0.70. The univariate Model A along with multivariate Model B and Model G performed better, with MAE of 11.5 to 15.1 tonnes/day and RMSE of 14.5 to 19.7 tonnes/day. The prediction scores of Model A using a Simple Additive Weighting (SAW) based ranking system seems to be negatively influenced by the distribution of the training and testing sets. There is no distinctive trend in the SAW-based prediction scores among the univariate model and the multivariate models. Instead of absolute values, percent changes and SAW relative ranking are adopted, making the proposed approach appropriate for cross-jurisdictional studies.

POLITENESS STRATEGIES USED BY DONALD TRUMP AND JOE BIDEN IN THE U.S. 2020 FINAL PRESIDENTIAL DEBATE

This study aims at determining the types of politeness strategies used by and the sociological variables influencing the candidates, Donald Trump and Joe Biden, in the U.S. 2020 Final Presidential Debate. This study also used the theory of Brown and Levinson (1987). This study used the U.S. 2020 final presidential debate video as the source of data. The video was uploaded on YouTube on October 22, 2020. The data were the speaker’s utterances containing politeness strategies. The results of the study revealed that the two candidates used various politeness strategies in the debate. There were 42 data obtained from the statements of the candidates who used the politeness strategy. The most widely used strategy by the two candidates is positive politeness. Then, negative politeness, off record, and bald on record respectively. Meanwhile, the use of politeness strategy by the candidates is influenced by social distance (D) and absolute ranking (R) of impositions.

Model-Driven Analysis of ECG Using Reinforcement Learning.

Modeling is essential to better understand the generative mechanisms responsible for experimental observations gathered from complex systems. In this work, we are using such an approach to analyze the electrocardiogram (ECG). We present a systematic framework to decompose ECG signals into sums of overlapping lognormal components. We use reinforcement learning to train a deep neural network to estimate the modeling parameters from an ECG recorded in babies from 1 to 24 months of age. We demonstrate this model-driven approach by showing how the extracted parameters vary with age. From the 751,510 PQRST complexes modeled, 82.7% provided a signal-to-noise ratio that was sufficient for further analysis (>5 dB). After correction for multiple tests, 10 of the 24 modeling parameters exhibited statistical significance below the 0.01 threshold, with absolute Kendall rank correlation coefficients in the [0.27, 0.51] range. These results confirm that this model-driven approach can capture sensitive ECG parameters. Due to its physiological interpretability, this approach can provide a window into latent variables which are important for understanding the heart-beating process and its control by the autonomous nervous system.

Sensitivity of isoprene emission rate to ozone in greening trees is concurrently determined by isoprene synthesis capacity and stomatal conductance

The sensitivity of isoprene emission rate (ISOrate) to ozone (O3) in plant suggests potentially large changes in future isoprene emissions, which will have important consequences for atmospheric chemistry. However, the interspecific variation of ISOrate sensitivity to O3 and its key drivers remain largely unknown. In this study, four urban greening tree species were exposed to two O3 treatments (charcoal-filtered air, CF; and non-filtered ambient air plus 60 ppb extra O3, EO3) in open-top chambers for one growing season. We aimed to compare the interspecific variation in O3 inhibitory effect on ISOrate and explore its physiological mechanism. EO3 decreased the ISOrate by on average 42.5 % across species. According to absolute effect size ranking, the highest ISOrate sensitivity to EO3 was observed in Salix matsudana, followed by Sophora japonica and hybrid poplar clone ‘546’, while Quercus mongolica ISOrate was the least sensitive. Leaf anatomical structures differed in tree species but did not respond to EO3. Furthermore, the ISOrate sensitivity to O3 was driven by the concurrent effects of O3 on ISO synthesis ability (i.e., dimethylallyl diphosphate and isoprene synthase contents) and stomatal conductance. Overall, the mechanistic understanding grained from this study may promote the integrity of O3 effect into process-based ISO emission models.

Selection of optimal validation methods for quantitative structure–activity relationships and applicability domain

ABSTRACT This brief literature survey groups the (numerical) validation methods and emphasizes the contradictions and confusion considering bias, variance and predictive performance. A multicriteria decision-making analysis has been made using the sum of absolute ranking differences (SRD), illustrated with five case studies (seven examples). SRD was applied to compare external and cross-validation techniques, indicators of predictive performance, and to select optimal methods to determine the applicability domain (AD). The ordering of model validation methods was in accordance with the sayings of original authors, but they are contradictory within each other, suggesting that any variant of cross-validation can be superior or inferior to other variants depending on the algorithm, data structure and circumstances applied. A simple fivefold cross-validation proved to be superior to the Bayesian Information Criterion in the vast majority of situations. It is simply not sufficient to test a numerical validation method in one situation only, even if it is a well defined one. SRD as a preferable multicriteria decision-making algorithm is suitable for tailoring the techniques for validation, and for the optimal determination of the applicability domain according to the dataset in question.

An evaluation of conventional and deep learning‐based image‐matching methods on diverse datasets

AbstractImage matching plays an important role in photogrammetry, computer vision and remote sensing. Modern deep learning‐based methods have been proposed for image matching; however, whether they will surpass and take the place of the conventional handcrafted methods in the remote sensing field still remains unclear. A comprehensive evaluation on stereo remote sensing images is also lacking. This paper comprehensively evaluates the performance of conventional and deep learning‐based image‐matching methods by dividing the matching process into feature point extraction, description and similarity measure on various datasets, including images captured from close‐range indoor and outdoor scenarios, unmanned aerial vehicles (UAVs) and satellite platforms. Different combinations of the three steps are evaluated. The experimental results reveal that, first, the performance of the different combinations varies between individual datasets, and it is difficult to determine the best combination. Second, by using more comprehensive indicators on all of the datasets, that is, the average rank and absolute rank, the combination of scale‐invariant feature transform (SIFT), ContextDesc and the nearest neighbour distance ratio (NNDR), and also the original SIFT, achieve the best results, and are recommended for use in remote sensing. Third, the deep learning‐based Sub‐SuperPoint extractor obtains a good performance, and is second only to SIFT. The learning based ContextDesc descriptor is as effective as the SIFT descriptor, and the learning based SuperGlue matcher is not as stable as NNDR, but leads to a few top‐performing combinations. Finally, the handcrafted methods are generally faster than the deep learning‐based methods, but the efficiency of the latter is acceptable. We conclude that although a full deep learning‐based method/combination has not yet beaten the conventional methods, there is still much room for improvement with the deep learning‐based methods because large‐scale aerial and satellite training datasets remain to be constructed, and specific methods for remote sensing images remain to be developed. The performance of the different combinations of feature extractor, descriptor and similarity measure varies between individual datasets. The combination of SIFT, ContextDesc and NNDR, and also the original SIFT, achieve the best results when using more comprehensive indicators on all the datasets. For extractor, the learning based Sub‐SuperPoint is second only to SIFT; for descriptor, learning‐based ContextDesc is as effective as the SIFT descriptor; and for matcher, learning‐based SuperGlue is not as stable as NNDR.

Prediction of body weight from chest radiographs using deep learning with a convolutional neural network.

Accurate body weights are not necessarily available in routine clinical practice. This study aimed to investigate whether body weight can be predicted from chest radiographs using deep learning. Deep-learning models with a convolutional neural network (CNN) were trained and tested using chest radiographs from 85,849 patients. The CNN models were evaluated by calculating the mean absolute error (MAE) and Spearman's rank correlation coefficient (ρ). The MAEs of the CNN models were 2.63kg and 3.35kg for female and male patients, respectively. The predicted body weight was significantly correlated with the actual body weight (ρ = 0.917, p < 0.001 for females; ρ = 0.915, p < 0.001 for males). The body weight was predicted using chest radiographs by applying deep learning. Our method is potentially useful for radiation dose management, determination of the contrast medium dose, and estimation of the specific absorption rate in patients with unknown body weights.

Ranking of Web Pages in a Personalized Search

The basic idea behind a personalized web search is to deliver search results that are tailored to meet user needs, which is one of the growing concepts in web technologies. The personalized web search presented in this paper is based on exploiting the implicit feedbacks of user satisfaction during her web browsing history to construct a user profile storing the web pages the user is highly interested in. A weight is assigned to each page stored in the user’s profile; this weight reflects the user’s interest in this page. We name this weight the relative rank of the page, since it depends on the user issuing the query. Therefore, the ranking algorithm provided in this paper is based on the principle that; the rank assigned to a page is the addition of two rank values R_rank and A_rank. A_rank is an absolute rank, since it is fixed for all users issuing the same query, it only depends on the link structures of the web and on the keywords of the query. Thus, it could be calculated by the PageRank algorithm suggested by Brin and Page in 1998 and used by the google search engine. While, R_rank is the relative rank, it is calculated by the methods given in this paper which depends mainly on recording implicit measures of user satisfaction during her previous browsing history.

Tо Ecological Monitoring of Self- Propelled Field Sprinklers

An improved method for making a decision on choosing the best self-propelled sprayer from alternative options is proposed. It refers to new developments in various fields of environmental monitoring and analysis of field self-propelled sprinklers, but which substantiate the studied factors characterizing each unit. The factors are taken with real values of calculated parameters, for which the desirability function and weighty coefficients are developed using the method of pairwise comparison of calculated parameters of absolute and relative rank places and the percentile function. The dependence of the transfer of calculated parameters of the unit from real values to dimensionless ones for the y’ scale is clarified, according to which the desirability function is calculated, and then a generalized parameter characterizing the complex evaluation of the unit in respect to the weighting coefficient of each estimated parameter. The decision is made according to the maximum value of the parameters of integrated assessment of the sprinkler, in this task for the sprinkler TUMAN-1 (Russia), for which the value of the parameter of integrated assessment is obtained.

Abstract A032: Probabilistic statistical-based model for determining individual cancer risk for making personal recommendations to prevent cancer

Abstract About 42% of cancer cases in the US are linked to modifiable risk factors, and thus could be preventable, says a study from American Cancer Society [1]. This proportion can be represented as the percentage by which an individual's possibility of developing cancer can be reduced. Having data on the prevalence of various types of cancer, it is possible to rank risk factors, and give an individual a personalized list of recommendations sorted by importance in terms of preventing cancer. For most commonly cancers, a set of modifiable risk factors is known: for example, for colon cancer these are the following: excess body weight, physical inactivity, smoking, high consumption of red or processed meat, alcohol intake, low intake of calcium, fruits and vegetables and whole-grain fiber, and the proportion of cases arising from modifiable factors is 54.6% [1]. Since we consider this number to be the percentage by which the risk of disease in an individual can be reduced, in the absence of modifiable factors, the likelihood of this cancer in a person will decrease by 54.6%, respectively. The contribution of each of the factors separately is also known: so, for colon cancer, physical inactivity causes 16.3% cases, alcohol intake – 12.8%, etc. Let’s reduce risks in proportion to each other so that their sum is equal to 54.6%. The total degree of fulfillment of risk factors will be the mathematical expectation of all individual risks, that is the sum of the products of the impact of each risk by its degree of presence. To determine the absolute risk and rank the risks in order of priority of occurrence for person, we take statistics on the probability of developing cancer [2]. So, for the colon cancer the probability of disease for men in the interval birth-49yrs is 0.4%. According to our assumption, this value can be reduced by a maximum of 54.6% provided that all risk factors are excluded. Let’s take a situation when a person has 50% modifiable risk factors. Then the absolute (posterior) risk of disease occurrence, provided that risk factors are reduced, will be calculated with Bayes' theorem, where the numerator will be the product of the a priori disease probability (0.4%) by the disease probability provided that all risk factors are excluded (100%-54.6 %=45.4%), and the denominator will be the total probability of the disease: 100% minus the product of risk reduction value (54.6%) multiplied by the degree of its presence (for us 50%). It will be equal to 0.25%. It's easy to check the correctness of the formula by substituting boundary values: for example, for the degree of presence of risks equal to 100%, the calculated absolute risk will not change and will be 0.4%, whereas in the absence of all risk factors the absolute risk will decrease by 54.6% and will be 0.18%.

Using Resident-Sensitive Quality Measures Derived From Electronic Health Record Data to Assess Residents' Performance in Pediatric Emergency Medicine.

Traditional quality metrics do not adequately represent the clinical work done by residents and, thus, cannot be used to link residency training to health care quality. This study aimed to determine whether electronic health record (EHR) data can be used to meaningfully assess residents' clinical performance in pediatric emergency medicine using resident-sensitive quality measures (RSQMs). EHR data for asthma and bronchiolitis RSQMs from Cincinnati Children's Hospital Medical Center, a quaternary children's hospital, between July 1, 2017, and June 30, 2019, were analyzed by ranking residents based on composite scores calculated using raw, unadjusted, and case-mix adjusted latent score models, with lower percentiles indicating a lower quality of care and performance. Reliability and associations between the scores produced by the 3 scoring models were compared. Resident and patient characteristics associated with performance in the highest and lowest tertiles and changes in residents' rank after case-mix adjustments were also identified. 274 residents and 1,891 individual encounters of bronchiolitis patients aged 0-1 as well as 270 residents and 1,752 individual encounters of asthmatic patients aged 2-21 were included in the analysis. The minimum reliability requirement to create a composite score was met for asthma data (α = 0.77), but not bronchiolitis (α = 0.17). The asthma composite scores showed high correlations ( r = 0.90-0.99) between raw, latent, and adjusted composite scores. After case-mix adjustments, residents' absolute percentile rank shifted on average 10 percentiles. Residents who dropped by 10 or more percentiles were likely to be more junior, saw fewer patients, cared for less acute and younger patients, or had patients with a longer emergency department stay. For some clinical areas, it is possible to use EHR data, adjusted for patient complexity, to meaningfully assess residents' clinical performance and identify opportunities for quality improvement.

Inequities in ambient fine particulate matter: A spatiotemporal analysis in Canadian communities

BackgroundExposure to fine particulate matter (PM2.5) is associated with adverse health outcomes but communities are not randomly exposed to PM2.5. Previous cross-sectional environmental injustice analyses in Canada found disproportionately higher exposure to PM2.5 in low-income populations, visible minorities and immigrants. Beyond static surveillance, it is also important to evaluate how changes in PM2.5 exposure over time may differentially impact disadvantaged communities. We examine whether communities with different sociodemographic characteristics benefited equitably from the overall decreases in ambient concentrations of PM2.5 from 2001 to 2016 in Canada. MethodsWe derived census tract level estimates of average annual PM2.5 using validated satellite-based estimations of annual average PM2.5 concentration surfaces. We investigated how the spatial distribution of PM2.5 has evolved over 15 years (2001–2016) by comparing absolute values and rank percentiles of census tract level annual average PM2.5 concentrations in 2001 and 2016. Using decennial census data and multivariable linear regression, we determined if sociodemographic characteristics are associated with changes in exposure to PM2.5, accounting for geographic boundary changes between census periods. ResultsOverall, ambient PM2.5 concentrations decreased from 2001 (median of 9.1 μg/m3) to 2016 (median of 6.4 μg/m3), with varying provincial patterns. Across communities, ranked census tract specific PM2.5 in 2001 and in 2016 are highly correlated (Spearman's rho = 0.75). We found that, on average and accounting for provincial differences and baseline PM2.5, communities with greater density of aboriginal population, lower education, higher shelter-cost-to-income ratio, unemployment or lower income experienced smaller absolute decreases in PM2.5 from 2001 to 2016. ConclusionsIdentifying sociodemographic groups that benefit least from decreasing exposure to PM2.5 highlights the need to consider environmental injustice when designing or revising air pollution policies.

Perception of vibrato rate by professional singing voice teachers

This study sought to investigate how voice clinicians perceive vibrato rate alterations when presented with controlled, synthesized singing voice samples which vary in vibrato rate and vibrato extent. Thirty-four professional voice teachers completed a twelve-item demographic survey and performed a visual sort and rate task (VSR). For the VSR task, each participant listened to twenty synthesized samples and sorted them from slowest vibrato rate to fastest vibrato rate. This task resulted in distance (i.e. individual perception of vibrato rate) and rank-difference measurements for each sample. Two generalized linear mixed effects models (GLMM) and one linear model (LM) were computed. Results for GLMM’s found significant associations between vibrato extent and vibrato rate and both individual perception of vibrato rate and rank-difference. Results for the LM found no significant relationships between demographic information and absolute total ranking error. From the results of this study, it seems that both vibrato extent and vibrato rate influence the perception of vibrato rate for professional voice teachers. Neither age nor teaching experience seemed to relate to the ability to discern vibrato rate accurately.