Industrial materials images are an important application domain for content-based image retrieval. Users need to quickly search databases for images that exhibit similar appearance, properties, and/or features to reduce analysis turnaround time and cost. The images in this study are 2D images of millimeter-scale rock samples acquired at micrometer resolution with light microscopy or extracted from 3D micro-CT scans. Labeled rock images are expensive and time-consuming to acquire and thus are typically only available in the tens of thousands. Training a high-capacity deep learning (DL) model from scratch is therefore not practicable due to data paucity. To overcome this “few-shot learning” challenge, we propose leveraging pretrained common DL models in conjunction with transfer learning. The “similarity” of industrial materials images is subjective and assessed by human experts based on both visual appearance and physical qualities. We have emulated this human-driven assessment process via a physics-informed neural network including metadata and physical measurements in the loss function. We present a novel DL architecture that combines Siamese neural networks with a loss function that integrates classification and regression terms. The networks are trained with both image and metadata similarity (classification), and with metadata prediction (regression). For efficient inference, we use a highly compressed image feature representation, computed offline once, to search the database for images similar to a query image. Numerical experiments demonstrate superior retrieval performance of our new architecture compared with other DL and custom-feature-based approaches.

Bovine tuberculosis (bTB) is prevalent among livestock and wildlife in many countries including New Zealand (NZ), a country which aims to eradicate bTB by 2055. This study evaluates predictions related to the numbers of livestock herds with bTB in NZ from 2012 to 2021 inclusive using both statistical and mechanistic (causal) modelling. Additionally, this study made predictions for the numbers of infected herds between 2022 and 2059. This study introduces a new graphical method representing the causal criteria of strength of association, such as R2, and the consistency of predictions, such as mean squared error. Mechanistic modelling predictions were, on average, more frequently (3 of 4) unbiased than statistical modelling predictions (1 of 4). Additionally, power model predictions were, on average, more frequently (3 of 4) unbiased than exponential model predictions (1 of 4). The mechanistic power model, along with annual updating, had the highest R2 and the lowest mean squared error of predictions. It also exhibited the closest approximation to unbiased predictions. Notably, significantly biased predictions were all underestimates. Based on the mechanistic power model, the biological eradication of bTB from New Zealand is predicted to occur after 2055. Disease eradication planning will benefit from annual updating of future predictions.

Whole-genome sequencing (WGS) information has played a crucial role in the SARS-CoV-2 (COVID-19) pandemic by providing evidence about variants to inform public health policy. The purpose of this study was to assess the representativeness of sequenced cases compared with all COVID-19 cases in England, between March 2020 and August 2021, by demographic and socio-economic characteristics, to evaluate the representativeness and utility of these data in epidemiological analyses. To achieve this, polymerase chain reaction (PCR)-confirmed COVID-19 cases were extracted from the national laboratory system and linked with WGS data. During the study period, over 10% of COVID-19 cases in England had WGS data available for epidemiological analysis. With sequencing capacity increasing throughout the period, sequencing representativeness compared to all reported COVID-19 cases increased over time, allowing for valuable epidemiological analyses using demographic and socio-economic characteristics, particularly during periods with emerging novel SARS-CoV-2 variants. This study demonstrates the comprehensiveness of England’s sequencing throughout the COVID-19 pandemic, rapidly detecting variants of concern, and enabling representative epidemiological analyses to inform policy.

The Democratic Republic of the Congo (DRC) officially reports low coronavirus disease 19 (COVID-19) prevalence. This cross-sectional study, conducted between September and November 2021, assessed the COVID-19 seroprevalence in people attending Goma’s two largest markets, Kituku and Virunga. A similar study in a slum of Bukavu overlapped for 1 month using identical methods. COVID-19-unvaccinated participants (n = 796 including 454 vendors and 342 customers, 60% of whom were women) were surveyed. The median age of vendors and customers was 34.2 and 30.1 years, respectively. The crude and adjusted anti-SARS-CoV-2 antibody seroprevalence rates were 70.2% (95% CI 66.9–73.4%) and 98.8% (95% CI 94.1–100%), respectively, with no difference between vendors and customers. COVID-19 symptoms reported by survey participants in the previous 6 months were mild or absent in 58.9% and 41.1% of participants with anti-SARS-CoV-2 antibodies, respectively. No COVID-19-seropositive participants reported hospitalisation in the last 6 months. These findings are consistent with those reported in Bukavu. They confirm that SARS-CoV-2 spread without causing severe symptoms in densely populated settlements and markets and suggest that many COVID-19 cases went unreported. Based on these results, the relevance of an untargeted hypothetical vaccination programme in these communities should be questioned.

This study determined long-term health outcomes (≥10 years) of Q-fever fatigue syndrome (QFS). Long-term complaints, health-related quality of life (HRQL), health status, energy level, fatigue, post-exertional malaise, anxiety, and depression were assessed. Outcomes and determinants were studied for the total sample and compared among age subgroups: young (<40years), middle-aged (≥40–<65years), and older (≥65years) patients. 368 QFS patients were included. Participants reported a median number of 12.0 long-term complaints. Their HRQL (median EQ-5D-5L index: 0.63) and health status (median EQ-VAS: 50.0) were low, their level of fatigue was high, and many experienced post-exertional malaise complaints (98.9%). Young and middle-aged patients reported worse health outcomes compared with older patients, with both groups reporting a significantly worse health status, higher fatigue levels and anxiety, and more post-exertional malaise complaints and middle-aged patients having a lower HRQL and a higher depression risk. Multivariate regression analyses confirmed that older age is associated with better outcomes, except for the number of health complaints. QFS has thus a considerable impact on patients’ health more than 10 years after infection. Young and middle-aged patients experience more long-term health consequences compared with older patients. Tailored health care is recommended to provide optimalcare for each QFS patient.

This paper analyses aspects of generalized method of moments (GMM) inference in moment equality models in settings where standard regularity conditions may break down. Explicit analytic formulations for the asymptotic distributions of estimable functions of the GMM estimator and statistics based on the GMM criterion function are derived under relatively mild assumptions. The moment Jacobian is allowed to be rank deficient, so first order identification may fail, the values of the Jacobian singular values are not constrained, thereby allowing for varying levels of identification strength, the long-run variance of the moment conditions can be singular, and the GMM criterion function weighting matrix may also be chosen sub-optimally. The large-sample properties are derived without imposing a specific structure on the functional form of the moment conditions. Closed-form expressions for the distributions are presented that can be evaluated using standard software without recourse to bootstrap or simulation methods. The practical operation of the results is illustrated via examples involving instrumental variables estimation of a structural equation with endogenous regressors and a common CH features model.

Within US professional sports, trades within one’s own division are often perceived to be disadvantageous. We ask how common this practice is. To examine this question, we construct a date-stamped network of all trades in the National Basketball Association between June 1976 and May 2019. We then use season-specific weighted exponential random graph models to estimate the likelihood of teams avoiding within-division trade partners, and how consistent that pattern is across the observed period. In addition to the empirical question, this analysis serves to demonstrate the necessity and difficulty of constructing the proper baseline for statistical comparison. We find limited-to-no support for the popular perception.

We study two models of discrete height functions, that is, models of random integer-valued functions on the vertices of a tree. First, we consider the random homomorphism model, in which neighbours must have a height difference of exactly one. The local law is uniform by definition. We prove that the height variance of this model is bounded, uniformly over all boundary conditions (both in terms of location and boundary heights). This implies a strong notion of localisation, uniformly over all extremal Gibbs measures of the system. For the second model, we consider directed trees, in which each vertex has exactly one parent and at least two children. We consider the locally uniform law on height functions which are monotone, that is, such that the height of the parent vertex is always at least the height of the child vertex. We provide a complete classification of all extremal gradient Gibbs measures, and describe exactly the localisation-delocalisation transition for this model. Typical extremal gradient Gibbs measures are localised also in this case. Localisation in both models is consistent with the observation that the Gaussian free field is localised on trees, which is an immediate consequence of transience of the random walk.

In this paper, a new point process is introduced. It combines the nonhomogeneous Poisson process with the generalized Polya process (GPP) studied in recent literature. In reliability interpretation, each event (failure) from this process is minimally repaired with a given probability and GPP-repaired with the complementary probability. Characterization of the new process via the corresponding bivariate point process is presented. The mean numbers of events for marginal processes are obtained via the corresponding rates, which are used for considering an optimal replacement problem as an application.

Acute pyelonephritis (AP) epidemiology has been sparsely described. This study aimed to describe the evolution of AP patients hospitalised in France and identify the factors associated with urinary diversion and fatality, in a cross-sectional study over the 2014–2019 period. Adult patients hospitalised for AP were selected by algorithms of ICD-10 codes (PPV 90.1%) and urinary diversion procedure codes (PPV 100%). 527,671 AP patients were included (76.5% female: mean age 66.1, 48.0% Escherichia coli), with 5.9% of hospital deaths. In 2019, the AP incidence was 19.2/10,000, slightly increasing over the period (17.3/10,000 in 2014). 69,313 urinary diversions (13.1%) were performed (fatality rate 6.7%), mainly in males, increasing over the period (11.7% to 14.9%). Urolithiasis (OR [95% CI] =33.1 [32.3–34.0]), sepsis (1.73 [1.69–1.77]) and a Charlson index ≥3 (1.32 [1.29–1.35]) were significantly associated with urinary diversion, whereas E. coli (0.75 [0.74–0.77]) was less likely associated. The same factors were significantly associated with fatality, plus old age and cancer (2.38 [2.32–2.45]). This nationwide study showed an increase in urolithiasis and identified, for the first time, factors associated with urinary diversion in AP along with death risk factors, which may aid urologists in clinical decision-making.

In 2008, Tóth and Vető defined the self-repelling random walk with directed edges as a non-Markovian random walk on $\unicode{x2124}$: in this model, the probability that the walk moves from a point of $\unicode{x2124}$ to a given neighbor depends on the number of previous crossings of the directed edge from the initial point to the target, called the local time of the edge. Tóth and Vető found that this model exhibited very peculiar behavior, as the process formed by the local times of all the edges, evaluated at a stopping time of a certain type and suitably renormalized, converges to a deterministic process, instead of a random one as in similar models. In this work, we study the fluctuations of the local times process around its deterministic limit, about which nothing was previously known. We prove that these fluctuations converge in the Skorokhod $M_1$ topology, as well as in the uniform topology away from the discontinuities of the limit, but not in the most classical Skorokhod topology. We also prove the convergence of the fluctuations of the aforementioned stopping times.

The global prevalence and spread of multidrug-resistant organisms (MDROs) represent an emerging public health threat. Day care centre (DCC) attendance is a risk factor for MDRO carriage in children and their environment. This study aimed to map the epidemiology of carriage and potential transmission of these organisms within 18 Flemish DDCs (Belgium). An MDRO prevalence survey was organised between November 2018 and February 2019 among children attending the centres. Selective chromogenic culture media were used for the detection of extended-spectrum beta-lactamase-producing Enterobacterales (ESBL-E), carbapenemase-producing Enterobacterales (CPE), and vancomycin-resistant Enterococci (VRE) in faecal swabs obtained from diapers or jars (n = 448). All isolated MDROs were subjected to resistance gene sequencing. A total of 71 of 448 samples (15.8%) yielded isolates of ESBL-E with a predominance of Escherichia coli (92.2% of ESBL-E) and ESBL resistance gene blaCTX-M-15 (50.7% of ESBL coding genes in E. coli). ESBL-E prevalence varied between DCCs, ranging from 0 to 50%. Transmission, based on the clonal relatedness of ESBL-E strains, was observed. CPE was identified in only one child carrying an E. coli with an OXA-244 gene. VRE was absent from all samples. The observed prevalence of ESBL-E in Flemish DCCs is high compared with previous studies, and our findings re-emphasise the need for rigorous hygiene measures within such centres to control the further spread of MDROs in the community.

Hepatitis E virus infection is a major cause of acute hepatitis, typically self-limiting but occasionally leading to liver failure. Understanding disease progression factors could inform prevention strategies. This study aimed to analyse the characteristics of a large cohort of hospitalised hepatitis E patients in Tianjin, China, and explore factors influencing their progression to liver failure. A total of 1279 hospitalised patients with hepatitis E were included in this cross-sectional study in Tianjin, China. Student's t-test and the Mann–Whitney U-test were used for comparisons. Multiple logistic regression analysis was used to explore the association. Among these 1279 patients, 107 (8.4%) developed liver failure. Patients with diabetes mellitus (DM) (95% confidence interval [CI] 1.150–2.887, p = 0.011), liver cirrhosis (95% [CI] 2.229–7.224, p < 0.001), and hepatitis B (95% [CI] 1.159–4.512, p = 0.017) were more likely to progress to liver failure. Hepatitis E patients with comorbid DM, liver cirrhosis, or hepatitis B virus co-infection have higher risks of developing liver failure. Hepatitis E vaccination may be recommended for these vulnerable patients to curb disease severity.

A comparison theorem for state-dependent regime-switching diffusion processes is established, which enables us to pathwise-control the evolution of the state-dependent switching component simply by Markov chains. Moreover, a sharp estimate on the stability of Markovian regime-switching processes under the perturbation of transition rate matrices is provided. Our approach is based on elaborate constructions of switching processes in the spirit of Skorokhod’s representation theorem varying according to the problem being dealt with. In particular, this method can cope with switching processes in an infinite state space and not necessarily of birth–death type. As an application, some known results on the ergodicity and stability of state-dependent regime-switching processes can be improved.

Available data suggest that the immunogenicity of COVID-19 vaccines might decrease in the immunocompromised population, but data on vaccine immunogenicity and safety among people living with HIV (PLWH) are still lacking. The purpose of this meta-analysis is to compare the immunogenicity and safety of COVID-19 vaccines in PLWH with healthy controls. We comprehensively searched the following databases: PubMed, Cochrane Library, and EMBASE. The risk ratio (RR) of seroconversion after the first and second doses of a COVID-19 vaccine was separately pooled using random-effects meta-analysis. Seroconversion rate was lower among PLWH compared with healthy individuals after the first (RR = 0.77, 95% confident interval (CI) 0.64–0.92) and second doses (RR = 0.97, 95%CI 0.95–0.99). The risk of total adverse reactions among PLWH is similar to the risk in the healthy group, after the first (RR = 0.87, 95%CI 0.70–1.10) and second (RR = 0.83, 95%CI 0.65–1.07) doses. This study demonstrates that the immunogenicity and safety of SARS-CoV-2 vaccine in fully vaccinated HIV-infected patients were generally satisfactory. A second dose was related to seroconversion enhancement. Therefore, we considered that a booster dose may provide better seroprotection for PLWH. On the basis of a conventional two-dose regimen for COVID-19 vaccines, the booster dose is very necessary.

The coronavirus pandemic has created a new awareness of epidemics, and insurance companies have been reminded to consider the risk related to infectious diseases. This paper extends the traditional multi-state models to include epidemic effects. The main idea is to specify the transition intensities in a Markov model such that the impact of contagion is explicitly present in the same way as in epidemiological models. Since we can study the Markov model with contagious effects at an individual level, we consider individual risk and reserves relating to insurance products, conforming with the standard multi-state approach in life insurance mathematics. We compare our notions with other but related notions in the literature and perform numerical illustrations.