Recent studies utilizing AI-driven speech-based Alzheimer’s disease (AD) detection have achieved remarkable success in detecting AD dementia through the analysis of audio and text data. However, detecting AD at an early stage of mild cognitive impairment (MCI), remains a challenging task, due to the lack of sufficient training data and imbalanced diagnostic labels. Motivated by recent advanced developments in Generative AI (GAI) and Large Language Models (LLMs), we propose an LLM-based data generation framework, leveraging prior knowledge encoded in LLMs to generate new data samples. Our novel LLM generation framework introduces two novel data generation strategies, namely, the cross-lingual and the counterfactual data generation, facilitating out-of-distribution learning over new data samples to reduce biases in MCI label prediction due to the systematic underrepresentation of MCI subjects in the AD speech dataset. The results have demonstrated that our proposed framework significantly improves MCI Detection Sensitivity and F1-score on average by a maximum of 38% and 31%, respectively. Furthermore, key speech markers in predicting MCI before and after LLM-based data generation have been identified to enhance our understanding of how the novel data generation approach contributes to the reduction of MCI label prediction biases, shedding new light on speech-based MCI detection under low data resource constraint. Our proposed methodology offers a generalized data generation framework for improving downstream prediction tasks in cases where limited and/or imbalanced data have presented significant challenges to AI-driven health decision-making. Future study can focus on incorporating more datasets and exploiting more acoustic features for speech-based MCI detection.

We show that every $(n,d,\lambda )$-graph contains a Hamilton cycle for sufficiently large $n$, assuming that $d\geq \log ^{6}n$ and $\lambda \leq cd$, where $c=\frac {1}{70000}$. This significantly improves a recent result of Glock, Correia, and Sudakov, who obtained a similar result for $d$ that grows polynomially with $n$. The proof is based on a new result regarding the second largest eigenvalue of the adjacency matrix of a subgraph induced by a random subset of vertices, combined with a recent result on connecting designated pairs of vertices by vertex-disjoint paths in $(n,d,\lambda )$-graphs. We believe that the former result is of independent interest and will have further applications.

Let X be the sum of a diffusion process and a Lévy jump process, and for any integer $n\ge 1$ let $\phi_n$ be a function defined on $\mathbb{R}^2$ and taking values in $\mathbb{R}$, with adequate properties. We study the convergence of functionals of the type

\begin{align*} \Delta_n\sum_{i=1}^{[t/\Delta_n]} \phi_n\!\left(\alpha_n\left[\frac{X_{(i-1)\Delta_n}}{\alpha_n}\right], \:\frac{\alpha_n}{\sqrt{\Delta_n}}\left(\left[\frac{X_{i\Delta_n}}{\alpha_n}\right]-\left[\frac{X_{(i-1)\Delta_n}}{\alpha_n}\right]\right)\right),\end{align*}

$(\Delta_n)$

$(\alpha_n)$

$n\to +\infty$

$\frac{\alpha_n}{\sqrt{\Delta_n}}$

$[0,+\infty)$]

We study an optimal inventory control problem under a reflected jump–diffusion netflow process with state-dependent jumps, in which the intensity of the jump process can depend on the inventory level. We examine the well-posedness of the associated integro-differential Hamilton–Jacobi–Bellman (ID-HJB) equation with Neumann boundary condition in the classical sense. To achieve this, we first establish the existence of viscosity solutions to the ID-HJB equation of an auxiliary control problem with a compact policy space, which is proved to be equivalent to the primal problem. We reformulate the ID-HJB equation as a Neumann HJB equation with the (non-local) integral term expressed in terms of the value function of the auxiliary problem and prove the existence of a unique classical solution to the Neumann HJB equation. Then, the well-posedness of the primal ID-HJB equation follows from the unique classical solution of the Neumann HJB equation and the existence of viscosity solutions to the auxiliary ID-HJB equation. Based on this classical solution, we characterize the optimal (admissible) inventory control strategy and show the verification result for the primal control problem.

This study examined global trends in influenza-associated lower respiratory infections (LRIs) deaths from 1990 to 2019 using data from the GBD 2019. The annual percentage change (APC) and average annual percentage change (AAPC) were used to analyze age-standardized death rates (ASDR). Globally, the ASDR of influenza-associated LRIs was 3.29/100,000 in 2019, which was higher in the African region (6.57/100,000) and among adults aged 70 years and older (29.88/100,000). The ASDR of influenza-associated LRIs decreased significantly from 1990 to 2019 (AAPC = −1.88%, P < 0.05). However, it was significantly increased among adults aged 70 years and older during 2017–2019 (APC = 2.31%, P < 0.05), especially in Western Pacific Region and South-East Asia Regions. The ratio of death rates between adults aged 70 years and older and children aged under 5 years increased globally from 1.63 in 1990 to 5.34 in 2019, and the Western Pacific Region experienced the most substantial increase, with the ratio soaring from 1.83 in 1990 to 12.98 in 2019. Despite a decline in the global ASDR of influenza-associated LRIs, it continues to impose a significant burden, particularly in the African, Western Pacific regions and among the elderly population.

This paper is concerned with the growth rate of susceptible–infectious–recovered epidemics with general infectious period distribution on random intersection graphs. This type of graph is characterised by the presence of cliques (fully connected subgraphs). We study epidemics on random intersection graphs with a mixed Poisson degree distribution and show that in the limit of large population sizes the number of infected individuals grows exponentially during the early phase of the epidemic, as is generally the case for epidemics on asymptotically unclustered networks. The Malthusian parameter is shown to satisfy a variant of the classical Euler–Lotka equation. To obtain these results we construct a coupling of the epidemic process and a continuous-time multitype branching process, where the type of an individual is (essentially) given by the length of its infectious period. Asymptotic results are then obtained via an embedded single-type Crump–Mode–Jagers branching process.

The efficacy of COVID-19 vaccines against the Delta variant has been observed to be high, both against severe disease and infection. The full population level vaccine effectiveness, however, also contains the indirect effects of vaccination, which require analysis of transmission dynamics to uncover. Finland was close to naïve to SARS-CoV-2 infections before the Delta dominant era, and non-pharmaceutical interventions (NPIs) were at an internationally low level. We utilize Finnish register data and a mathematical model for transmission and COVID-19 disease burden to construct a completely unvaccinated control population and estimate the different components of the vaccine effectiveness. The estimated direct effectiveness was 72% against COVID-19 cases and 87–96% against severe disease outcomes, but the estimated indirect effectiveness was even better, 93% against cases and 94–97% against severe disease. The total and overall effectiveness, including both direct and indirect effects of vaccination, were thus excellent. Our results show how well the population was protected by vaccination during the Delta era, especially by the indirect effectiveness, providing protection also to the unvaccinated part of the population. The estimated averted numbers of hospitalizations, ICU admissions, and deaths in Finland during the Delta era under the implemented NPIs were about 100 times the observed numbers.

The study aims were to present in vitro susceptibilities of clinical isolates from Gram-negative bacteria bloodstream infections (GNBSI) collected in China. GNBSI isolates were collected from 18 tertiary hospitals in 7 regions of China from 2018 to 2020. Minimum inhibitory concentrations were assessed using a Trek Diagnostic System. Susceptibility was determined using CLSI broth microdilution, and breakpoints were interpreted using CLSI M100 (2021). A total of 1,815 GNBSI strains were collected, with E. coli (42.4%) and Klebsiella pneumoniae (28.6%) being the most prevalent species, followed by P. aeruginosa (6.7%). Susceptibility analyses revealed low susceptibilities (<40%) of ESBL-producing E. coli and K. pneumonia to third-/fourth-generation cephalosporins, monobactamases, and fluoroquinolones. High susceptibilities to colistin (95.0%) and amikacin (81.3%) were found for K. pneumoniae, while Acinetobacter baumannii exhibited a high susceptibility (99.2%) to colistin but a low susceptibility to other antimicrobials (<27.5%). Isolates from ICUs displayed lower drug susceptibility rates of K. pneumoniae and A. baumannii than isolates from non-ICUs (all P < 0.05). Carbapenem-resistant and ESBL-producing K. pneumoniae detection was different across regions (both P < 0.05). E. coli and K. pneumoniae were major contributors to GNBSI, while A. baumannii exhibited severe drug resistance in isolates obtained from ICU departments.

A tantalizing open problem, posed independently by Stiebitz in 1995 and by Alon in 1996 and again in 2006, asks whether for every pair of integers $s,t \ge 1$ there exists a finite number $F(s,t)$ such that the vertex set of every digraph of minimum out-degree at least $F(s,t)$ can be partitioned into non-empty parts $A$ and $B$ such that the subdigraphs induced on $A$ and $B$ have minimum out-degree at least $s$ and $t$, respectively.

In this short note, we prove that if $F(2,2)$ exists, then all the numbers $F(s,t)$ with $s,t\ge 1$ exist and satisfy $F(s,t)=\Theta (s+t)$. In consequence, the problem of Alon and Stiebitz reduces to the case $s=t=2$. Moreover, the numbers $F(s,t)$ with $s,t \ge 2$ either all exist and grow linearly, or all of them do not exist.

We use recent advances in polynomial diffusion processes to develop a continuous-time joint mortality model for the actuarial valuation and risk analysis of life insurance liabilities. The model considers the stochastic nature of future mortality improvements and introduces a common subordinator for the marginal survival processes, resulting in a nontrivial dependence structure between the survival of pairs of individuals. Polynomial diffusion processes can be used to derive closed-form formulae for standard actuarial quantities. The model fits well with a classic dataset provided by a Canadian insurer and can be used to evaluate products issued to multiple lives, as shown through numerical applications.

This study proposes two novel time-varying model-averaging methods for time-varying parameter regression models. When the number of predictors is small, we propose a novel time-varying complete subset-averaging (TVCSA) procedure, where the optimal time-varying subset size is obtained by minimizing the local leave-h-out cross-validation criterion. The TVCSA method is asymptotically optimal for achieving the lowest possible local mean squared error. When the number of predictors is relatively large, we propose a factor TVCSA method to reduce the computational burden by first reducing the dimension of predictors by extracting a few factors using principal component analysis and then obtaining the TVCSA forecasts from time-varying models with the generated factors. We show that the TVCSA estimator remains asymptotically optimal in the presence of generated factors. Monte Carlo simulation studies have provided favorable evidence for the TVCSA methods relative to the popular model-averaging methods in the literature. Empirical applications to equity premiums and inflation forecasting highlight the practical merits of the proposed methods.

Invasive group A Streptococcal (iGAS) outbreaks have been linked to Community Healthcare Services Delivered at Home (CHSDH). There is, however, very limited evidence describing the epidemiology and mortality of iGAS cases associated with CHSDH. We used routine data to describe iGAS cases in adults who had received CHSDH prior to onset and compare characteristics between CHSDH-outbreak and non-outbreak CHSDH cases, in South East England between December 2021 and December 2023. There were 80/898 (8.9%) iGAS case episodes with CHSDH prior to onset; cases were in elderly people (50% aged 85 and over), and had primarily received wound or ulcer care (93.8%), with almost all care delivered by community nurses (98.8%). The 30-day all-cause case fatality was 26.3%. Emm 1.0 was the most common type (17.5%). In this period, 5/11 iGAS outbreaks (45.4%) were CHSDH-associated, and 25 cases with receipt of CHSDH prior to onset (31.3%, Confidence Interval [CI] 21.3–42.6%) were linked to these outbreaks. On univariate analysis, CHSDH-outbreak case episodes were more likely to be associated with emm pattern genotype E (OR 6.1 95% CI 1.8–20.9), and skin or soft tissue infection clinical presentation (OR 3.6, 95% CI 1.1–12.0) than non-outbreak CHSDH cases. There may be an increased risk of propagation of iGAS outbreaks in patients receiving CHSDH, emphasizing the need for rigorous early infection prevention and control, and outbreak surveillance.