Malaria, historically a significant health burden in temperate Europe, particularly in the low-lying marshy areas, is often poorly represented in discussions of health in the pre-modern Netherlands. Although malaria does not produce pathognomonic skeletal lesions, the haemolytic anaemia associated with repeated infection is thought to contribute to the development of cribra orbitalia, making population-level patterns in this non-specific skeletal marker informative for exploring past malaria burden. This study applied a spatial epidemiological approach, which investigated (1) the spatial distribution of cribra orbitalia prevalence across 28 archaeological medieval sites in the Netherlands, and (2) whether this distribution can be explained by underlying environmental features consistent with malaria transmission and historical mosquito density. Global Moran’s I revealed a significant positive spatial autocorrelation in prevalence. Local Indicator of Spatial Association (LISA) analysis confirmed this, identifying distinct High–High clusters in the Southwest and Low–Low clusters in the East of the Netherlands. However, linear regression models using broad-scale environmental variables failed to explain these spatial patterns. This likely reflects their inability to capture the specific ecology of the local malaria mosquito, Anopheles atroparvus, which preferentially breeds in brackish environments. Consistent with this interpretation, cribra orbitalia prevalence was significantly positively correlated with historical (1938) estimates of A. atroparvus density. The observed clustering and correlation with mosquito density suggest that malaria contributed to cribra orbitalia prevalence and may have been an important disease in certain regions of the medieval Netherlands; however, interpretation is constrained by small non-adult sample sizes as well as uneven preservation across the Netherlands.

Coronaviruses of the human variety have been the culprit of global epidemics of varying levels of lethality, including COVID-19, which has impacted more than 200 countries and resulted in 5.7 million fatalities as of May 2022. Effective clinical management necessitates the allocation of sufficient resources and the employment of appropriately skilled personnel. The elderly population and individuals with diabetes are at increased risk of more severe manifestations of COVID-19. Countries with a higher gross domestic product (GDP) typically exhibit superior health outcomes and reduced mortality rates. Here, we suggest a predictive model for the density of medical doctors and nursing personnel for 134 countries using a support vector machine (SVM). The model was trained in 107 countries and tested in 27, with promising results shown by the kappa statistics and ROC analysis. The SVM model used for predictions showed promising results with a high level of agreement between actual and predicted cluster values.

This Research Communication explores the usefulness of predictive modelling to explain bacterial behaviour during cooling. A simple dynamic lag phase model was developed and validated. The model takes into account the effect of the cooling profile on the lag phase and growth in bulk tank milk. The time before the start of cooling was the most critical and should not exceed 1 h. The cooling rate between 30 and approximately 10 °C was the second most critical period. Cooling from 30 to 10 °C within 2 h ensured minimal growth of psychrotrophic bacteria in the milk. The cooling rate between 10 and 4 °C (the slowest phase of cooling) was of surprisingly little importance. Given a normal cooling profile to 10 °C, several hours of prolonged cooling time made practically no difference in psychrotrophic counts. This behaviour can be explained by the time/temperature dependence of the work needed by the bacteria to complete the lag phase at low temperature. For milk quality advisors, it is important to know that slow cooling below 10 °C does not result in high total counts of bacteria. In practice, slow cooling is occasionally found at farms with robotic milking. However, when comparing psychrotrophic growth in bulk milk tanks designed for robotic milking or conventional milking, the model predicted less growth for robotic milking for identical cooling profiles. It is proposed that due to the different rates of milk entering the tank, fewer bacteria will exit the lag phase during robotic milking and they will be more diluted than in conventional milking systems. At present, there is no international standard that specifies the cooling profile in robotic systems. The information on the insignificant effect of the cooling rate below 10 °C may be useful in the development of a standard.

For the first time, geomorphology and archaeology are combined for a 165 km long stretch of the Meuse river, resulting in a geomorphogenetic map (GKM) and a series of archaeological predictive maps (AVM). The maps cover the central part the Meuse valley, located in the province of Limburg between Mook in the north and Eijsden in the south. The area consists of fluvial and aeolian landforms of the Holocene Meuse floodplain and Younger Dryas aged terraces along it, spanning a period of approximately 15,000 years of landscape genesis and human habitation. The GKM more clearly discriminates between map units of Younger Dryas and early Holocene age than in previous mappings of the Meuse valley. The AVM series provide predictive information on the location of sites for four distinct consecutive archaeological periods and four main cultural themes. The maps contribute to a better understanding of landscape processes (fluvial and aeolian geomorphology and the impact of man on river behaviour), and the possibilities for human habitation and land use in prehistoric and historic times.

Psychiatric research has entered the age of ‘Big Data’. Datasets now routinely involve thousands of heterogeneous variables, including clinical, neuroimaging, genomic, proteomic, transcriptomic and other ‘omic’ measures. The analysis of these datasets is challenging, especially when the number of measurements exceeds the number of individuals, and may be further complicated by missing data for some subjects and variables that are highly correlated. Statistical learning-based models are a natural extension of classical statistical approaches but provide more effective methods to analyse very large datasets. In addition, the predictive capability of such models promises to be useful in developing decision support systems. That is, methods that can be introduced to clinical settings and guide, for example, diagnosis classification or personalized treatment. In this review, we aim to outline the potential benefits of statistical learning methods in clinical research. We first introduce the concept of Big Data in different environments. We then describe how modern statistical learning models can be used in practice on Big Datasets to extract relevant information. Finally, we discuss the strengths of using statistical learning in psychiatric studies, from both research and practical clinical points of view.

Underwriting in the United States (US) life insurance marketplace has evolved tremendously over the last several decades. This paper will take a brief look at that history, from the older underwriting techniques still in use today to the introduction of smoker/nonsmoker distinctions in about 1980 to the evolution of preferred underwriting in the late 1980s, and finally to a movement toward simplified issue underwriting and a new approach to older age underwriting today.

The calculus of looping sequences is a formalism for describing theevolution of biological systems by means of term rewriting rules. Inthis paper we enrich this calculus with a type discipline whichpreserves some biological properties depending on the minimum andthe maximum number of elements of some type requested by the present elements. The typesystem enforces these properties and typed reductions guarantee thatevolution preserves them. As an example, we model the hemoglobinstructure and the equilibrium between cell death and division: typedreductions prevent undesirable behaviors.

Determining the effects of genetically modified (GM) crops on non-target organisms is essential as many non-target species provide important ecological functions. However, it is simply not possible to collect field data on more than a few potential non-target species present in the receiving environment of a GM crop. While risk assessment must be rigorous, new approaches are necessary to improve the efficiency of the process. Utilisation of published information and existing data on the phenology and population dynamics of test species in the field can be combined with limited amounts of experimental biosafety data to predict possible outcomes on species persistence. This paper presents an example of an approach where data from laboratory experiments and field studies on phenology are combined using predictive modelling. Using the New Zealand native weevil species Nicaeana cervina as a case study, we could predict that oviposition rates of the weevil feeding on a GM ryegrass could be reduced by up to 30% without threat to populations of the weevil in pastoral ecosystems. In addition, an experimentally established correlation between feeding level and oviposition led to the prediction that a consistent reduction in feeding of 50% or higher indicated a significant risk to the species and could potentially lead to local extinctions. This approach to biosafety risk assessment, maximising the use of pre-existing field and laboratory data on non-target species, can make an important contribution to informed decision-making by regulatory authorities and developers of new technologies.

Sparse grids are the basis for efficient high dimensional approximation and have recently been applied successfully to predictive modelling. They are spanned by a collection of simpler function spaces represented by regular grids. The sparse grid combination technique prescribes how approximations on a collection of anisotropic grids can be combined to approximate high dimensional functions.

In this paper we study the parallelisation of fitting data onto a sparse grid. The computation can be done entirely by fitting partial models on a collection of regular grids. This allows parallelism over the collection of grids. In addition, each of the partial grid fits can be parallelised as well, both in the assembly phase, where parallelism is done over the data, and in the solution stage using traditional parallel solvers for the resulting PDEs. Using a simple timing model we confirm that the most effective methods are obtained when both types of parallelism are used.