Search

Bisphenol A alternatives are manufactured as potentially less harmful substitutes of bisphenol A (BPA) that offer similar functionality. These alternatives are already in the market, entering the environment and thus raising ecological concerns. However, it can be expected that levels of BPA alternatives will dominate in the future, they are limited information on their environmental safety. The EU PARC project highlights BPA alternatives as priority chemicals and consolidates information on BPA alternatives, with a focus on environmental relevance and on the identification of the research gaps. The review highlighted aspects and future perspectives. In brief, an extension of environmental monitoring is crucial, extending it to cover BPA alternatives to track their levels and facilitate the timely implementation of mitigation measures. The biological activity has been studied for BPA alternatives, but in a non-systematic way and prioritized a limited number of chemicals. For several BPA alternatives, the data has already provided substantial evidence regarding their potential harm to the environment. We stress the importance of conducting more comprehensive assessments that go beyond the traditional reproductive studies and focus on overlooked relevant endpoints. Future research should also consider mixture effects, realistic environmental concentrations, and the long-term consequences on biota and ecosystems.

IntroductionInformal markets play a crucial role in providing fresh and nutritious foods for people in low and middle-income countries. However, the safety of food sold in these markets remains a major concern, contributing to a high burden of foodborne illnesses.MethodsThis study was designed to analyze stakeholder perceptions of food safety in selected informal public markets in Kenya. Eight focus group discussions and 15 key informant interviews were conducted. In addition, two workshops were held to validate the findings.Results and discussionWe found that consumers and vendors confounded food quality, especially freshness, with safety, and paid more attention to appearance and physical qualities than food safety. Information (such as branding and expiry dates) and heuristic cues (such as the flow of buyers and the presence of flies) were used to assess food safety. Meat was perceived as the food most likely to cause disease. Chemical hazards in fresh produce and aflatoxins in maize were mentioned as priority hazards by both vendors and consumers. Stakeholders more knowledgeable about food safety considered microbial hazards as a top priority. Although diarrheal diseases were reported to be prevalent in the communities, these were often thought to be the result of agri-chemical residues in fresh produce or the consumption of high-fat foods, and not commonly linked to bacteria in food. Gaps identified during the interviews included poor infrastructure, inadequate food safety knowledge, insufficient or prohibitive policies, insufficient political will, and poor food safety-related practices. There were few mentions of lack of motivation or incentives for behavior change, or insufficient consumer demand for food safety. To fill the gaps, several opportunities were discussed, including contextualizing policies and regulations, investing in infrastructure, capacity building, and training, and promoting involvement and collaboration among various stakeholders.ConclusionThis study has highlighted gaps and misperceptions that need to be addressed through proper knowledge and awareness to effectively combat foodborne disease challenges. Behavioral change approaches to improve food safety are recommended.

Convolutional neural networks (CNNs) have played an important role in medical imaging—from diagnostics to research to data integration. This has allowed clinicians to plan operations, diagnose patients earlier, and study rare diseases in more detail. However, data quality, quantity, and imbalance all pose challenges for CNN training and accuracy; in addition, training costs can be high when many types of CNNs are needed in a health care system. Topology and geometry provide tools to ameliorate these challenges for CNNs when they are integrated into the CNN architecture, particularly in the data preprocessing steps or convolution layers. This paper reviews the current integration of geometric tools within CNN architectures to reduce the burden of large training datasets and offset computational costs. This paper also identifies fertile areas for future research into the integration of geometric tools with CNNs.

Approaches describing how the brain changes to accomplish cognitive tasks tend to rely on collapsed data. Here, authors present a new approach that maintains high dimensionality and use it to describe individual differences in how brain activity is represented and organized across different cognitive tasks.

Predictive models are used in many different fields of science and engineering and are always prone to make faulty predictions. These faulty predictions can be more or less malignant depending on the model application. We describe fibers of failure (FiFa), a method to classify failure modes of predictive processes. Our method uses Mapper, an algorithm from topological data analysis (TDA), to build a graphical model of input data stratified by prediction errors. We demonstrate two ways to use the failure mode groupings: either to produce a correction layer that adjusts predictions by similarity to the failure modes; or to inspect members of the failure modes to illustrate and investigate what characterizes each failure mode. We demonstrate FiFa on two scenarios: a convolutional neural network (CNN) predicting MNIST images with added noise, and an artificial neural network (ANN) predicting the electrical energy consumption of an electric arc furnace (EAF). The correction layer on the CNN model improved its prediction accuracy significantly while the inspection of failure modes for the EAF model provided guiding insights into the domain-specific reasons behind several high-error regions.

The theory of multidimensional persistence captures the topology of a multifiltration - a multiparameter family of increasing spaces. Multifiltrations arise naturally in the topological analysis of scientific data. In this paper, we give a polynomial time algorithm for computing multidimensional persistence. We recast this computation as a problem within computational commutative algebra and utilize algorithms from this area to solve it. While the resulting problem is EXPSPACE-complete and the standard algorithms take doubly-exponential time, we exploit the structure inherent withing multifiltrations to yield practical algorithms. We implement all algorithms in the paper and provide statistical experiments to demonstrate their feasibility.

Women with previously diagnosed gestational diabetes mellitus (GDM) are at increased risk of type-2-diabetes mellitus (T2D). We aimed to establish links between glucose tolerance (GT) and serum fatty acid (FA) profile in the transition from GDM to T2D. Six years after GDM, 221 women were grouped as having normal GT (NGT), impaired GT (IGT), or T2D based on oral GT test results. Fasting serum FAs were profiled, anthropometric measures taken, and dietary intake determined. Linoleic acid (LA) was significantly higher in NGT women (p < 0.001) compared with IGT and T2D, and emerged as a strong predictor of low glucose and insulin levels, independently of BMI. Self-reported vegetable oil consumption correlated with LA serum levels and glucose levels. Delta-6-, delta-9-, and stearoyl-CoA-desaturase activities were associated with decreased GT, and delta-5-desaturase activities with increased GT. In a subgroup of women at high risk of diabetes, low LA and high palmitic acid levels were seen in those that developed T2D, with no differences in other FAs or metabolic measurements. Results suggest that proportions of LA and palmitic acid are of particular interest in the transition from GDM to T2D. Interconversions between individual FAs regulated by desaturases appear to be relevant to glucose metabolism.