Components Of Model Research Articles

Comparing univariate filtration preceding and succeeding PLS-DA analysis on the differential variables/metabolites identified from untargeted LC-MS metabolomics data

BackgroundPLS-DA of high-dimensional metabolomics data is frequently employed to capture the most pertinent features to sample classification. But the presence of numerous insignificant input features could distort the PLS-DA model, blow up and scramble the selected differential features. Usually, univariate filtration is subsequently complemented to refine the selected features, but often giving unstable results. Whereas by precluding insignificant features through univariate data prefiltration assessed by FDR adjusted p-value, PLS-DA can generate more stable and reliable differential features. We explored and compared these two data analysis procedures to gain insights into the underlying mechanisms responsible for the disparate results. ResultsThe effect of univariate data filtration preceding and succeeding PLS-DA analysis on the identified discriminative features/metabolites was investigated using LC-MS data acquired on the samples of human serum and C. elegans extracts, with and without metabolite standards spiked to simulate the treated and control groups of biological samples. It was shown that the univariate data prefiltration before PLS-DA usually gave less but more stable and likely more reliable and meaningful differential features, while PLS-DA applied directly to the original data could be affected by the presence of insignificant features and orthogonal noise. Large number of insignificant variables and orthogonal noise could distort the generated PLS-DA model and affect the p(corr) value, and artificially inflate the calculated VIP values of relevant features due to the increased total number of input features for model construction, thus leading to more false positives selected by the conventional VIP threshold of 1.0. Significance and noveltyUnivariate data filtration preceding PLS-DA was important for the identification of reliable differential features if using a conventional threshold of VIP of 1.0. Presence of insignificant features could distort the PLS-DA model and inflate VIP values. Appropriate VIP threshold is associated with the numbers of input features and the model components. For PLS-DA without univariate prefiltration, threshold of VIP larger than 1.0 is recommended for the selection of discriminative features to reduce the false positives.

Perceived value in food retail: moderating effect of store format

Objective - This paper’s purpose is to measure the value perceived by food retail consumers under the moderating effect of the store format. Methodology: An online survey was administered to 493 food retail consumers. For the data analysis it was used the quantitative technique of Structural Equation Modelling (SEM). The analyses were performed with the aid of Minitab software version 12.1 and SmartPLS version 3. Results: The highest explanation degree of the perceived value for the Super / Hypermarkets format was the variables Products and Store Environment, as was the case for Wholesale self-service format. However, it should be noted that the Wholesale self-service format had the lowest structural coefficients for all paths of the model, indicating that the factors have less explanatory power on the perceived value of the consumer compared to the Super/Hypermarkets format. A notable difference between formats was noticed in the factor a Additional services variable, showing its vocation aimed at lower costs. Theoretical/methodological contributions: The investigation developed a quantitative model that demonstrates the adequacy of the model's component variables and that they vary according to the store format. The model developed can also be used as a basis for future research in other countries on food retail and other variations of retail. Relevance/Originality: This study is the first to investigate and measure the value perceived by food retail consumers, also taking into account the store format.

The influence of corporate social responsibility on repurchase intention: The mediating effect of satisfaction

The study evaluates the impact of corporate social responsibility (CSR) on customer satisfaction and repurchase intention in the fast food service business in Vietnam. This study used quantitative research methods with a sample of 414 customers aged 18 and older who have used fast food service in Vietnam. Primary data were collected based on customers’ willingness to provide information through questionnaire links on social networking platforms such as Facebook and Zalo. Structural equation modeling and mediating effect analysis were used to test the correlation between components in the research model. Research results have identified three components of CSR, including community responsibility, environmental responsibility, and ethical responsibility in business, that directly and positively influence customer satisfaction. The results validate the mediating influence of satisfaction on the correlation between CSR components (community, environmental, and ethical responsibility) and repurchase intention, which very few previous studies have performed. These findings theoretically contribute to the literature, verifying three CSR components from the customer’s point of view in the fast food service business, including community, environmental, and ethical responsibility. Expanding the theory on factors affecting customer satisfaction and promoting cause-related marketing, prosocial behavior, and competitive advantage theory is necessary. As for managerial contributions, fast food business brands are suggested to invest and increase their CSR activities.

Motivation for Scientific Research of Faculty at the National Defense and Security Training Center belongs to Higger Education Institutions

By reviewing both national and international research on lecturers' scientific research motivation, considering the components of various research models, and incorporating appropriate research results to build a model of factors that affect lecturers' scientific research motivation in a few National Security and Defense facilities based in Ha Noi, this research has investigated these factors based on the data set of 144 responds from various lecturers, in addition to in-depth interviews of 03 lecturers of National Security and Defense Center, Vietnam National University. The analyzed data reveals that suggested elements with varying degrees of influence that contribute to lecturers' scientific motivation are: i) Lecturers' awareness of scientific reasearch; ii) Lecturers' ability to conduct scientific reasearch; iii) Support from Center/Faculty to scientific reasearch; iv) Social issues from the lecturers; and v) Research environment. These factors have mostly contributed to the full explaination of the influence level on lecturers' scientific research motivation.

Movement Ecology of Adult Western Corn Rootworm: Implications for Management.

Movement of adult western corn rootworm, Diabrotica virgifera virgifera LeConte, is of fundamental importance to this species' population dynamics, ecology, evolution, and interactions with its environment, including cultivated cornfields. Realistic parameterization of dispersal components of models is needed to predict rates of range expansion, development, and spread of resistance to control measures and improve pest and resistance management strategies. However, a coherent understanding of western corn rootworm movement ecology has remained elusive because of conflicting evidence for both short- and long-distance lifetime dispersal, a type of dilemma observed in many species called Reid's paradox. Attempts to resolve this paradox using population genetic strategies to estimate rates of gene flow over space likewise imply greater dispersal distances than direct observations of short-range movement suggest, a dilemma called Slatkin's paradox. Based on the wide-array of available evidence, we present a conceptual model of adult western corn rootworm movement ecology under the premise it is a partially migratory species. We propose that rootworm populations consist of two behavioral phenotypes, resident and migrant. Both engage in local, appetitive flights, but only the migrant phenotype also makes non-appetitive migratory flights, resulting in observed patterns of bimodal dispersal distances and resolution of Reid's and Slatkin's paradoxes.

Automated remote sensing system for crops monitoring and irrigation management, based on leaf color change and piecewise linear regression models for soil moisture content predicting

Plants can serve as biological sensors if their “readings” and the feedback they provide us through changes in the colour of their leaves can be correctly interpreted. The study aims to predict soil moisture and, as such, the need for irrigation, using nonlinear mathematical models, describing the relationship between RGB and HSL colour model components and soil moisture and temperature. Nonlinear mathematical models used in the study are based on piecewise linear regression with breakpoint and soil moisture prediction using colour components and soil temperature with a deviation of +-6%. A system for automated irrigation was created and its control program was made, the basic control law of which is based on non-linear piecewise linear models. The automated irrigation management system includes a remote crop monitoring subsystem and an irrigation management subsystem. The program processes the photo received from the camera and activates the actuators when watering is needed. Compared to manual data collection in the first part of the study, the program calculates the average RGB model values from images in the studied row of tomato plantations with an accuracy of over 99% for the R and G components and over 92% for the B component. The program also predicts soil moisture with 98% accuracy. The practical significance of the water-saving efforts of this study lies in the development of a program-controlled automated irrigation system that utilizes plants as biological sensors, employing nonlinear mathematical models based on leaf colour changes to accurately predict soil moisture

Algorithmic Accuracy in Context: Assessing the Impact of Financially Motivated Dishonesty

Abstract Algorithms are expected to radically improve the way businesses operate. To fully realize these benefits, customers have to accept the algorithmic implementations. Accuracy is a critical component of algorithmic acceptance models. However, thus far, research into accuracy has been limited to user ratings of the accuracy of the algorithmic output despite strong evidence for customer (user) acts of embellishment and other moral hazards. This article aims to deepen the understanding of accuracy ratings by assessing the impact of variation in user input accuracy on the accuracy ratings of the algorithm’s ultimate response. Drawing on honesty, a survey was presented to 500 individuals on the Cloud Research platform. The quantitative analysis showed that the more inaccurately a user felt and behaved, the more accurately they rated the algorithmic response of ‘decline’, with contextual accuracy factors explaining up to 35% of the variation in ratings of the algorithm’s accuracy. This finding raises important implications for researchers and practitioners who want to improve algorithmic acceptance but may have limited their improvement focus to explainability or model accuracy without considering the user behavior. Algorithmic accuracy ratings and acceptance should be evaluated in the context of the user and their likelihood to provide inaccurate inputs.

Role of community-based health clubs in promoting patients’ health education for diabetes self-care management: an interventional qualitative study in a Ugandan urban setting

BackgroundSelf-care management support is a core component of the Chronic Care Model that emphasises the need for empowering and preparing patients to manage their healthcare. In diabetes mellitus (DM) management,...

The Emissions Model Intercomparison Project (Emissions-MIP): quantifying model sensitivity to emission characteristics

Abstract. Anthropogenic emissions of aerosols and precursor compounds are known to significantly affect the energy balance of the Earth–atmosphere system, alter the formation of clouds and precipitation, and have a substantial impact on human health and the environment. Global models are an essential tool for examining the impacts of these emissions. In this study, we examine the sensitivity of model results to the assumed height of SO2 injection, seasonality of SO2 and black carbon (BC) particulate emissions, and the assumed fraction of SO2 emissions that is injected into the atmosphere as particulate phase sulfate (SO4) in 11 climate and chemistry models, including both chemical transport models and the atmospheric component of Earth system models. We find large variation in atmospheric lifetime across models for SO2, SO4, and BC, with a particularly large relative variation for SO2, which indicates that fundamental aspects of atmospheric sulfur chemistry remain uncertain. Of the perturbations examined in this study, the assumed height of SO2 injection had the largest overall impacts, particularly on global mean net radiative flux (maximum difference of −0.35 W m−2), SO2 lifetime over Northern Hemisphere land (maximum difference of 0.8 d), surface SO2 concentration (up to 59 % decrease), and surface sulfate concentration (up to 23 % increase). Emitting SO2 at height consistently increased SO2 and SO4 column burdens and shortwave cooling, with varying magnitudes, but had inconsistent effects across models on the sign of the change in implied cloud forcing. The assumed SO4 emission fraction also had a significant impact on net radiative flux and surface sulfate concentration. Because these properties are not standardized across models this is a source of inter-model diversity typically neglected in model intercomparisons. These results imply a need to ensure that anthropogenic emission injection height and SO4 emission fraction are accurately and consistently represented in global models.

Assessing the Suitability and Prioritization of the Components of the media Literacy Competence Development Model for English Language Teachers

Assessing the Suitability and Prioritization of the Components of the media Literacy Competence Development Model for English Language Teachers

On Two -Way Nested Classification Random Model

Nested design is a specific kind of experimental design in which each level of each element occurs with all levels of different factors without relationships. The aim of this article is to estimate the variance components of the interferometry model by two methods and to compare them and obtain the least measurement error for these estimators, and depending on the viscosity chemical experiment we obtained the least measurement line through the comparison between the two methods. we conclude that, the variance components estimated by the maximum penalized likelihood method are less than the variance components estimated by the maximum likelihood method. Conversely, the mean squared error of the parameters estimated by the maximum likelihood method is less than the mean squared error of the parameters estimated by the maximum penalized likelihood method. Thus, the parameters estimated using maximum penalized likelihood method have the least measurement error, and therefore are more accurate and acceptable than the parameters estimated using the maximum likelihood method.

Systolic Blood Pressure Levels and Probable Dementia : Secondary Analysis of a Randomized Clinical Trial

AbstractBackgroundDespite a significant 19% reduction in risk for mild cognitive impairment, the primary Systolic Blood Pressure Intervention Trial (SPRINT) analysis lacked sufficient evidence to assert statistical significance for the 17% reduction in probable dementia alone in intensive blood pressure (BP) control, defined as a target systolic BP < 120 mmHg, compared to the standard target of < 140 mmHg (p = 0.10). However, early termination of SPRINT may have caused these analyses to be under‐powered. We applied the statistical framework of joint models for longitudinal data and time‐to‐event outcomes to analyze the effect of systolic BP levels on probable dementia using the SPRINT data.MethodsSPRINT was a multicenter randomized clinical trial that enrolled patients ≥50 years of age with hypertension and increased cardiovascular risk but without diabetes or history of stroke. In this secondary analysis, the primary outcome was occurrence of adjudicated probable dementia after randomization. Secondary outcomes included adjudicated mild cognitive impairment and a composite outcome of mild cognitive impairment or probable dementia. To analyze these outcomes, we applied joint longitudinal and survival models. In the longitudinal component of the joint model, systolic BP levels were modeled as a quadratic function of time since randomization, accounting for correlation in repeated measurements and for assignment to intensive BP control versus standard care. In the survival component, multi‐variable adjustment for race, education, and age at baseline was applied.ResultsOver a median follow up of 5.2 years, a total of 325 probable dementia and 640 mild cognitive impairment events occurred. After multi‐variable adjustment (as above), the hazard ratio per 10 mmHg decrease in systolic BP was 0.89 (95% confidence interval [CI] 0.79, 0.99) for probable dementia, 0.89 (95% CI 0.82, 0.96) for mild cognitive impairment, and 0.90 (95% CI 0.84, 0.97) for their composite outcome (Table). Dynamic predictions after 2 years of follow‐up provide a descriptive summary of systolic BP levels and predicted risk of probable dementia or mild cognitive impairment (Figure).ConclusionReducing systolic BP decreases the risk of both probable dementia and mild cognitive impairment.

Consumer behavior on the market of educational services: features of choosing a specialism and a higher educational institution

Methods. The results are obtained with the following methods: logical-analytical and generalization in the analysis of the main theoretical approaches to the formation of the model of consumer choice on the market of educational services; abstraction in determining the main components of the consumer behavior model on the market of educational services; methods of marketing research in determining the characteristics of consumer behavior on the market of educational services. Results. The segmentation of consumers of educational services was carried out and the main segments were identified, which demonstrate different types of consumer behavior, to which appropriate marketing efforts should be directed. The influence of internal and external factors on the behavior of consumers of educational services is determined. A model of the behavior of potential consumers of educational services has been developed, which depends on the motives and the presence or absence of a vector of further development. It has been established that the number of applicants is directly proportional to the level of loyalty of existing students. In order to build an effective marketing strategy, it is necessary to identify the motives that encourage applicants to choose certain specialism. Novelty. Behavior models of potential consumers of educational services depend on individually determined motives and the direction of the vector of the life trajectory (understanding of the further development of the personality). Understanding the possibilities of corporate communication and their use increases the level of satisfaction of both students and applicants. Practical value. The target segments of consumers of educational services and their behavioral characteristics at various stages of consumer choice taking into account individual motives have been determined.

Associations of urinary caffeine and caffeine metabolites with metabolic syndrome in US adults.

The relationship between caffeine and metabolic syndrome (MetS) has only been evaluated from the perspective of caffeine consumption. The association between urinary caffeine and MetS is still unclear. This study examined the associations between urinary caffeine and its metabolites and MetS and its components among adults. Data from the United States (US) National Health and Nutrition Examination Survey (NHANES) 2011-2014 was analyzed. NHANES is a stratified, multi-stage survey of all non-institutionalized persons in the US. A total of 2,394 subjects aged ≥ 18 years without missing data were selected in this study. Urinary caffeine and caffeine metabolite levels were quantified using high-performance liquid chromatography-electrospray ionization-tandem quadrupole mass spectrometry (HPLC-ESI-MS/MS) with stable isotope-labeled internal standards. We performed principal components analysis (PCA) to investigate the underlying correlation structure of 15 features of urinary caffeine and its metabolites and then used these principal components (PCs) as independent variables to conduct logistic regression analysis with or without restricted cubic spline (RCS) terms to explore the associations between caffeine metabolites and MetS. Two main PCs that were derived from the PCA explained 90.67% of the total variance of caffeine and its metabolites. The first PC (PC1, strongly correlated with 1-MU, 1,3-DMU, 1,7-DMU, 1,3,7-TMU, 1-MX, 1,3-DMX, 1,7-DMX, 1,3,7-TMX, and AAMU) was positively correlated with risk of MetS (OR = 1.27, p < 0.001) and all its components (all ORs > 1, all p-values < 0.001) in the unadjusted models, while in the adjusted models, it was positively correlated with MetS (OR = 1.16, p = 0.042) and central obesity (OR = 1.22, p < 0.001). In the unadjusted model, there were significant associations between the second PC (PC2, correlated with 3-MU, 7-MU, 3,7-DMU, 3-MX, 7-MX, and 3,7-DMX) and MetS (OR = 1.11, P = 0.030) and central obesity (OR = 1.16, P < 0.001), while in the adjusted models (adjustment variables include gender, age, race/ethnicity, education level and income-poverty ratio, smoking status, drinking, and physical activity), PC2 was positively associated with MetS (OR = 1.15, p = 0.035) and central obesity (OR = 1.15, p = 0.005) and negatively associated with raised triglycerides (TG) (OR = 0.84, p = 0.008). Moreover, we observed U-shaped associations between PC1 and the risk of raised TG both in unadjusted (Pnon-linear = 0.017) and adjusted (Pnon-linear = 0.014) models. Urinary caffeine metabolites were positively associated with the risk of MetS and its components through different linear or non-linear patterns.

Effects of low‐density polyethylene and polyamide microplastics on the microbiological and chemical characteristics of an Andisol

AbstractMicroplastics (MPs) are a significant threat to soils. However, there is scarce information on the impact of MPs on soil properties, particularly in volcanic ash‐derived soils. The objective of this study was to evaluate the impact of polyamide (PA) and low‐density polyethylene (LDPE) MPs on the biological and chemical characteristics of an Andisol from central Chile. Twenty‐one parameters were evaluated, including pH, electrical conductivity (EC), total organic carbon (TOC), dissolved organic carbon (DOC), total phosphorus (TP), available phosphorous (AP), available nitrogen (AN), inorganic nitrogen forms (NH4+, NO2− and NO3), carbohydrates (CHO), polyphenols (POLs), humic substances, soil basal respiration (SBR) and activities of soil enzymes such as dehydrogenase, β‐glucosidase, phosphatase and urease. For this, a microcosm was set up in clay pots in an incubation chamber at 21°C and 60% soil moisture, with the addition of PA and LDPE at doses of 1% and 3% w/w; a control treatment consisting of microcosm without MPs was also included. After 6 weeks of incubation, PA addition resulted in an increase in TOC and NH4+ by up to 32% and 26%, respectively, and a decrease in NO3− by 22%. AP decreased by 15%–19% with the addition of PA and LDPE. Similarly, acid phosphatase and β‐glucosidase activities decreased by 15% and 26% with PA and LDPE, respectively. The distance‐based linear model (DistLM) was used to analyse relationships in chemical and biological datasets. The analysis revealed that TOC and TP were primary components in the best model for predicting microbiological changes (R2 = .469, AICs = 16.026), indicating that MPs accumulation affects soil carbon cycling and P content. Overall, the results show that MPs addition alters soil chemical and microbiological properties in Andisols, with varying effects depending on the type and dose of MPs, with the highest dose (3%) producing the most marked negative effects.

Predictive analysis of effects of water stress on strawberry seedlings using fluorescent image channel components.

To investigate the effect of water stress on strawberry seedlings, a chlorophyll-fluorescence-image-acquisition system was developed. Strawberry seedlings of uniform growth were selected for grouped water-stress incubation experiments; the collected chlorophyll-fluorescence images of leaves were converted to red-green-blue (RGB), hue-saturation-value (HSV), and hue-saturation-intensity (HSI) color spaces and analyzed for water and chlorophyll contents measured at the same time for 14 consecutive days. The results indicate that the analysis and prediction of plant stress conditions can be effectively conducted using the channel components of the color-space model and the channel component ratios, which provide a reference for promoting agricultural development.

24‐hour rest‐activity rhythms and light exposure patterns in relation to cognitive function in a nationally representative U.S. sample

AbstractBackgroundHuman diurnal behaviors are orchestrated by the internal circadian clock and influenced by external stimuli such as light. Aging is associated with declined circadian output and shifted circadian phase, leading to weakened rest‐activity rhythms and altered light exposure patterns, which may in turn further exacerbate circadian impairment. Limited but growing research reported that characteristics of rest‐activity rhythms were associated with risk of dementia and cognitive decline in older populations. However, there are several gaps in the literature: 1) most previous studies included predominantly white participants with relatively high socioeconomic status, limiting the generalizability of study findings to more diverse populations; 2) few studies examined the 24‐hour light exposure pattern, an environmental cue essential to healthy circadian function, in relation to cognitive function.MethodsWe applied the extended cosine model, nonparametric algorithms, and functional principal component analysis (fPCA), to characterize rest‐activity rhythms and light exposure patterns in the National Health and Nutrition Examination Survey (2011‐2014). Cognitive function was measured by word learning and recall modules from the Consortium to Establish a Registry for Alzheimer’s disease (CERAD), the Animal Fluency test, and the Digit Symbol Substitution test.ResultsMultiple rest‐activity characteristics were associated with different aspects of cognitive function, generally supporting a relationship between weakened circadian rhythms and poorer cognitive performance. For example, when compared to the highest quintile, the lowest quintile of F statistic, a measure of overall rhythmicity, was associated with a significantly lower score on the CERAD delay recall test (beta (95% confidence interval (CI)), ‐0.61 (‐0.77, ‐0.18)) and an increase in the odds of having a performance score in the lowest quartile of the study sample (odds ratio (95% CI), 1.55 (1.05, 2.29)). Moreover, measures of amplitude and rhythm fragmentation derived from nonparametric algorithms and fPCA were also associated with cognitive function. Findings were generally consistent across subgroups with different gender, age, race/ethnicity and education levels, but subgroup differences were also observed. Moreover, preliminary findings suggested a relationship between light exposure patterns and cognitive function.Conclusions24‐hour rest‐activity and light exposure patterns are associated with cognitive function in a diverse and nationally representative sample.

Indonesia's Electronic Export Potential in The Global Market with Gravity Panel Model Approach

&#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; The purpose of this study is to calculate Indonesia's electronic export potential to 14 countries during the period 2007-2022 using a panel-gravity approach using the PPML technique. The results of this study confirm the positive and significant relationship between Indonesia's electronics exports and the main component of the gravity model, namely Indonesia's GDP and trading partners. Another variability is the economic openness of trading partners has a positive relationship and exchange rates have a positive but not significant relationship. Variability of distance, and trade agreements have a negative relationship with Indonesia's electronic exports. The results for electronic export potential show that Indonesian electronics still have a lot of leverage to increase its exports to 9 countries. Thus, the position of the electronics industry as a potential industry can be carried out by targeting these countries. But to capitalize on the potential benefits of electronics exports, policymakers need to strengthen infrastructure and regulations related to the electronics industry.&#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D; &#x0D;

Understanding the evolution of trust in a participatory health research partnership: A qualitative study.

Advancements in evaluating the impact of participatory health research (PHR) have been made through comprehensive models like the community-based participatory research (CBPR) conceptual model, which provides a useful framework for exploring how context and partnership processes can influence health research design and interventions. However, challenges in operationalising aspects of the model limit our understanding and evaluation of the PHR process. Trust is frequently identified as an important component of the CBPR model, which supports the development of key partnership outcomes, such as partnership synergy. However, trust continues to be limited to a binary view (as present or absent), which is problematic given its inherently dynamic and temporal nature. The aim of this qualitative study is to understand the evolution of trust in the national public and patient involvement (PPI) network in Ireland. Participants from the PPI network (n = 15/21) completed a semistructured interview discussing the evolution of trust by reviewing four social network maps derived from a previous longitudinal study. Following Braun and Clarke, we used reflexive thematic analysis, to iteratively develop, analyse and interpret our mediated reflection of the data. Participants described the evolution of trust as a function of three contextual factors: (1) the set-up and organisation of the network, (2) how people work togetherand (3) reflection on the process and outcomes. Their descriptions across these themes seemed to vary depending on partnership type with National Partners and Site Leads having more opportunities to demonstrate trust (e.g., via leadership roles or more resources), compared to Local. Thus, visibility and the opportunity to be visible, depending on the set-up and organisation of the network and how people work together, seemingly play an important role in the evolution of trust over time. Based on these findings, we provide important questions for reflection across themes that could be considered for future PHR partnerships. Given that the opportunity and visibility to build and maintain trust over time may not be equally available to all partners, it is important to find ways to invest in and commit to equitable relationships as the key to the success (i.e., longevity) of partnerships. We reflect on/offer important implications for those engaging in PHR partnerships and those who fund such research. A Research Advisory Group comprising four research partners (representing academic, serviceand community organisations) from the PPI Ignite Network provided input and approval for the research objectives of this study as well as previously published work informing this study. Informal consultation occurred with members of this group to discuss findings from this study, assisting with the way findings are presented and described, to be accessible for diverse audiences. Two Research Advisory Group members were involved in the interpretation of the results, and one is a co-author of this manuscript (Zoe Hughes).

DGD-cGAN: A dual generator for image dewatering and restoration

Underwater images are usually covered with a blue–greenish colour cast, making them distorted, blurry or low in contrast. This phenomenon occurs due to the light attenuation given by the scattering and absorption in the water column. In this paper we present an image dewatering approach motivated upon the observation that the image formation model can be used to drive the learning process by constraining the loss function and making used of paired data. To this end, we employ a conditional generative adversarial network (cGAN) with two generators. Our Dual Generator Dewatering cGAN (DGD-cGAN) removes the haze and colour cast induced by the water column and restores the true colours of underwater scenes whereby the effects of various attenuation and scattering phenomena that occur in underwater images are tackled by the two generators. The first generator takes as input the underwater image and predicts the dewatered scene, while the second generator learns the underwater image formation process by implementing a custom loss function based upon the transmission and the veiling light components of the image formation model. Extensive experiments show that DGD-cGAN consistently delivers a margin of improvement as compared with state-of-the-art methods on several widely available datasets.