Despite major advances in artificial intelligence (AI) for medicine and healthcare, the deployment and adoption of AI technologies remain limited in real-world clinical practice. In recent years, concerns have been raised about the technical, clinical, ethical and legal risks associated with medical AI. To increase real world adoption, it is essential that medical AI tools are trusted and accepted by patients, clinicians, health organisations and authorities. This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare. The FUTURE-AI consortium was founded in 2021 and currently comprises 118 inter-disciplinary experts from 51 countries representing all continents, including AI scientists, clinicians, ethicists, and social scientists. Over a two-year period, the consortium defined guiding principles and best practices for trustworthy AI through an iterative process comprising an in-depth literature review, a modified Delphi survey, and online consensus meetings. The FUTURE-AI framework was established based on 6 guiding principles for trustworthy AI in healthcare, i.e. Fairness, Universality, Traceability, Usability, Robustness and Explainability. Through consensus, a set of 28 best practices were defined, addressing technical, clinical, legal and socio-ethical dimensions. The recommendations cover the entire lifecycle of medical AI, from design, development and validation to regulation, deployment, and monitoring. FUTURE-AI is a risk-informed, assumption-free guideline which provides a structured approach for constructing medical AI tools that will be trusted, deployed and adopted in real-world practice. Researchers are encouraged to take the recommendations into account in proof-of-concept stages to facilitate future translation towards clinical practice of medical AI.

Recent advances in the study of microstates for 1/16-BPS black holes have inspired renewed interest in the analysis of heavy operators. For these operators, traditional techniques that work effectively in the planar limit are no longer applicable. Methods that are sensitive to finite N effects are required. In particular, trace relations that connect different multi-trace operators must be carefully considered. A powerful approach to tackling this challenge, which utilizes the representation theory of the symmetric group, is provided by restricted Schur polynomials. In this review, we develop these methods with the goal of providing the background needed for their application to 1/16-BPS black holes.

Brain tumors are among the deadliest cancers worldwide, with particularly devastating impact in Sub-Saharan Africa (SSA) where limited access to medical imaging infrastructure and expertise often delays diagnosis and treatment planning. Accurate brain tumor segmentation is crucial for treatment planning, surgical guidance, and monitoring disease progression, yet manual segmentation is time-consuming and subject to inter-observer variability. Recent advances in deep learning, based on Convolutional Neural Networks (CNNs) and Transformers have demonstrated significant potential in automating this critical task. This study evaluates three state-of-the-art architectures, SwinUNETR-v2, nnUNet, and MedNeXt for automated brain tumor segmentation in multi-parametric Magnetic Resonance Imaging (MRI) scans. We trained our models on the BraTS-Africa 2024 and BraTS2021 datasets, and performed validation on the BraTS-Africa 2024 validation set. We observed that training on a mixed dataset (BraTS-Africa 2024 and BraTS2021) did not yield improved performance on the SSA validation set in all tumor regions compared to training solely on SSA data with well-validated methods. Ensembling predictions from different models also lead to notable performance increases. Our best-performing model, a finetuned MedNeXt, achieved an average lesion-wise Dice score of 0.84, with individual scores of 0.81 (enhancing tumor), 0.81 (tumor core), and 0.91 (whole tumor). While further improvements are expected with extended training and larger datasets, these results demonstrate the feasibility of deploying deep learning for reliable tumor segmentation in resource-limited settings. We further highlight the need to improve local data acquisition protocols to support the development of clinically relevant, region-specific AI tools.

Cervical dystonia, a debilitating neurological disorder marked by involuntary muscle contractions and chronic pain, presents significant treatment challenges despite advances in botulinum toxin therapy. While botulinum toxin type B has emerged as one of the leading treatments, comparative efficacy across doses and the influence of demographic factors for personalized medicine remain understudied. This study aimed to: (1) compare the efficacy of different botulinum toxin type B doses using Bayesian methods, (2) evaluate demographic and clinical factors affecting treatment response, and (3) establish a probabilistic framework for personalized cervical dystonia management. We analyzed data from a multicenter randomized controlled trial involving 109 patients assigned to placebo, 5,000 units, or 10,000 units of botulinum toxin type B groups. The primary outcome was the Toronto Western Spasmodic Torticollis Rating Scale measured over 16 weeks. Bayesian hierarchical modeling assessed treatment effects while accounting for patient heterogeneity. Lower botulinum toxin type B doses (5,000 units) showed greater overall Toronto Western Spasmodic Torticollis Rating Scale score reductions (treatment effect: -2.39, 95% Probability Interval: -4.10 to -0.70). Male patients demonstrated better responses (5.2% greater improvement) than female patients. Substantial between-patient variability and site-specific effects were observed, highlighting the need for personalized protocols. The study confirms botulinum toxin type B's dose-dependent efficacy while identifying key modifiable factors in treatment response. Bayesian methods provided nuanced insights into uncertainty and heterogeneity, paving the way for personalized medicine in cervical dystonia management.

We present the findings of "The Alzheimer's Disease Prediction Of Longitudinal Evolution" (TADPOLE) Challenge, which compared the performance of 92 algorithms from 33 international teams at predicting the future trajectory of 219 individuals at risk of Alzheimer's disease. Challenge participants were required to make a prediction, for each month of a 5-year future time period, of three key outcomes: clinical diagnosis, Alzheimer's Disease Assessment Scale Cognitive Subdomain (ADAS-Cog13), and total volume of the ventricles. The methods used by challenge participants included multivariate linear regression, machine learning methods such as support vector machines and deep neural networks, as well as disease progression models. No single submission was best at predicting all three outcomes. For clinical diagnosis and ventricle volume prediction, the best algorithms strongly outperform simple baselines in predictive ability. However, for ADAS-Cog13 no single submitted prediction method was significantly better than random guesswork. Two ensemble methods based on taking the mean and median over all predictions, obtained top scores on almost all tasks. Better than average performance at diagnosis prediction was generally associated with the additional inclusion of features from cerebrospinal fluid (CSF) samples and diffusion tensor imaging (DTI). On the other hand, better performance at ventricle volume prediction was associated with inclusion of summary statistics, such as the slope or maxima/minima of biomarkers. TADPOLE's unique results suggest that current prediction algorithms provide sufficient accuracy to exploit biomarkers related to clinical diagnosis and ventricle volume, for cohort refinement in clinical trials for Alzheimer's disease. However, results call into question the usage of cognitive test scores for patient selection and as a primary endpoint in clinical trials.

This research set out to explore and delineate spatial patterns and mortality distributions for various cancer types across U.S. states between 1999 and 2021. The aim was to uncover region-specific cancer burdens and inform geographically targeted prevention efforts. We analyzed state-level cancer mortality records sourced from the CDC WONDER platform, concentrating on cancer sites consistently reported across the 48 contiguous states and Washington, D.C., excluding Hawaii, Alaska, and Puerto Rico. Multivariate clustering using Mahalanobis distance grouped states according to similarities in mortality profiles. Spatial autocorrelation was examined for each cancer type using both Global Moran's I and Local Indicators of Spatial Association (LISA). Additionally, the Getis-Ord statistic was applied to detect cancer-specific hotspots and cold spots.

We study an optimal stopping problem when the state process is governed by a general Feller process. In particular, we examine viscosity properties of the associated value function with no a priori assumption on the stochastic differential equation satisfied by the state process. Our approach relies on properties of the Feller semigroup. We present conditions on the state process under which the value function is the unique viscosity solution to an Hamilton-Jacobi-Bellman (HJB) equation associated with a particular operator. More specifically, assuming that the state process is a Feller process, we prove uniqueness of the viscosity solution which was conjectured in [26]. We then apply our results to study viscosity property of optimal stopping problems for some particular Feller processes, namely diffusion processes with piecewise coefficients and semi-Markov processes. Finally, we obtain explicit value functions for optimal stopping of straddle options, when the state process is a reflected Brownian motion, Brownian motion with jump at boundary and regime switching Feller diffusion, respectively (see Section 8).

Transforming food systems is essential to bring about a healthier, equitable, sustainable, and resilient future, including achieving global development and sustainability goals. To date, no comprehensive framework exists to track food systems transformation and their contributions to global goals. In 2021, the Food Systems Countdown to 2030 Initiative (FSCI) articulated an architecture to monitor food systems across five themes: 1 diets, nutrition, and health; 2 environment, natural resources, and production; 3 livelihoods, poverty, and equity; 4 governance; and 5 resilience and sustainability. Each theme comprises three-to-five indicator domains. This paper builds on that architecture, presenting the inclusive, consultative process used to select indicators and an application of the indicator framework using the latest available data, constructing the first global food systems baseline to track transformation. While data are available to cover most themes and domains, critical indicator gaps exist such as off-farm livelihoods, food loss and waste, and governance. Baseline results demonstrate every region or country can claim positive outcomes in some parts of food systems, but none are optimal across all domains, and some indicators are independent of national income. These results underscore the need for dedicated monitoring and transformation agendas specific to food systems. Tracking these indicators to 2030 and beyond will allow for data-driven food systems governance at all scales and increase accountability for urgently needed progress toward achieving global goals.

Computer Vision for Analyzing and Classifying cells and tissues often require rigorous lab procedures and so automated Computer Vision solutions have been sought. Most work in such field usually requires Feature Extractions before the analysis of such features via Machine Learning and Machine Vision algorithms. We developed a Convolutional Neural Network that classifies 5 types of epithelial breast cell lines comprised of two human cancer lines, 2 normal immortalized lines, and 1 immortalized mouse line (MDA-MB-468, MCF7, 10A, 12A and HC11) without requiring feature extraction. The Multiclass Cell Line Classification Convolutional Neural Network extends our earlier work on a Binary Breast Cancer Cell Line Classification model. CellLineNet is 31-layer Convolutional Neural Network trained, validated and tested on a 3,252 image dataset of 5 types of Epithelial Breast cell Lines (MDA-MB-468, MCF7, 10A, 12A and HC11) in an end-to-end fashion. End-to-End Learning enables CellLineNet to identify and learn on its own, visual features and regularities most important to Breast Cancer Cell Line Classification from the dataset of images. Using Transfer Learning, the 28-layer MobileNet Convolutional Neural Network architecture with pre-trained ImageNet weights is extended and fine tuned to the Multiclass Epithelial Breast cell Line Classification problem. CellLineNet simply requires an imaged Cell Line as input and it outputs the type of breast epithelial cell line (MDA-MB-468, MCF7, 10A, 12A or HC11) as predicted probabilities for the 5 classes. CellLineNet scored a 96.67% Accuracy.

We investigated the electronic and optical properties of bilayer AB stacked Boron and Nitrogen vacancies in hexagonal Boron Nitride (h-BN) using density functional theory (DFT). The density of states (DOS) and electronic band structure showed that Boron vacancy in bilayer h-BN results in a magnetic and conducting ground state. The band gap energy ranges from 4.56 eV for the pristine BN bilayer to 0.12 eV for a single Nitrogen vacancy in the bilayer. Considering the presence of 1,3,4-Boron vacancy, half metallic character is observed. However, the 2-boron vacancy configuration resulted in metallic character. The bilayers with 1,2,3,4- Nitrogen vacancy has a band gap of 0.39, 0.33, 0.28 and 0.12eV respectively, which is significantly less than the pristine band gap. Also B and N vacancy induces ferromagnetism in the h-BN bilayer. The maximum total magnetic moment for the Boron vacant system is 6.583uB in case of 4-Boron vacancy configuration. In case of Nitrogen vacancy system it is 3.926uB for 4-Nitrogen vacancy configuration. The optical response of the system is presented in terms of the absorption coefficient, refractive index and dielectric constant for pristine as well as the defective configurations. Negative value of dielectric constant for Boron vacant system in the energy range 0.9-1.4 eV and for Nitrogen vacant system in the energy range 0.5-0.8 eV opens an opportunity for it to be utilized for negative index optical materials. The current study shows that B and N vacancies in bilayer h-BN could have potential applications in nano-structure based electronics, optoelectronics and spintronic devices.

The study investigates the effects of crude oil prices on inflation and interest rate in Ghana using data obtained from Bank of Ghana data repository. The Augmented Dickey-Fuller and the Phillips-Perron tests were used to test the presence or otherwise of unit root relationship among the variables. The stationarity test showed that the variables are either integrated of order one or integrated of order zero. Autoregressive distributed lag bounds test approach was adopted to examine cointegration among the variables. The results showed a positive relationship between crude oil prices and inflation in the long-run. The short-run show the coefficient of the first period lag value of inflation is negative, but statistically insignificant in the short-run. However, the second period lag value of inflation is positive and significant. The result also shows negative relationship between crude oil price and interest rate. Based on the findings, it is recommended that the government of Ghana should provide and strengthen the efficiency of the public transport system to help reduce transport fares in order to shield the poor from the implications of oil price increases in Ghana.

Failure to receive post-natal care within first week of delivery causes a 3% increase in the possibility of Acute Respiratory Infection in children under five. Mothers with unpaid maternity leave put their children at a risk of 3.9% increase in the possibility of ARI compared to those with paid maternity leave.

In many statistical problems, several estimators are usually available for interval estimation of a parameter of interest, and hence, the selection of an appropriate estimator is important. The criterion for a good estimator is to have a high coverage probability close to the nominal level and a shorter interval length. However, these two concepts are in opposition to each other: high and low coverages are associated with longer and shorter interval lengths respectively. Some methods, such as bootstrap calibration, modify the nominal level to improve the coverage and thereby allow the selection of intervals based on interval lengths only. Nonetheless, these methods are computationally expensive. In this paper, we propose an index which offers an easy to compute approach of comparing confidence interval estimators based on a compromise between the coverage probability and the confidence interval length. We illustrate that the confidence interval index has range of values within the neighborhood of the range of the coverage probability, [0,1]. In addition, a good confidence interval estimator has an index value approaching 1; and a bad confidence interval has an index value approaching 0. A simulation study was conducted to assess the finite sample performance of the index. The proposed index is illustrated with a practical example from the literature

This paper reviews the state of the art in satellite and machine learning based poverty estimates and finds some interesting results. The most important factors correlated to the predictive power of welfare in the reviewed studies are the number of pre-processing steps employed, the number of datasets used, the type of welfare indicator targeted, and the choice of AI model. As expected, studies that used hard indicators as targets achieved better performance in predicting welfare than those that targeted soft ones. Also expected was the number of pre-processing steps and datasets used having a positive and statistically significant relationship with welfare estimation performance. Even more important, we find that the combination of ML and DL significantly increases predictive power by as much as 15 percentage points compared to using either alone. Surprisingly, we find that the spatial resolution of the satellite imagery used is important but not critical to the performance as the relationship is positive but not statistically significant. The finding of no evidence indicating that predictive performance of a statistically significant effect occurs over time was also unexpected. These findings have important implications for future research in this domain. For example, the level of effort and resources devoted to acquiring more expensive, higher resolution SI will have to be reconsidered given that medium resolutions ones seem to achieve similar results. The increasingly popular approach of combining ML, DL, and TL, either in a concurrent or iterative manner, might become a standard approach to achieving better results.

In extreme value analysis, the extreme value index plays a vital role as it determines the tail heaviness of the underlying distribution and is the primary parameter required for the estimation of other extreme events. In this paper, we review the estimation of the extreme value index when observations are subject to right random censoring and the presence of covariate information. In addition, we propose some estimators of the extreme value index, including a maximum likelihood estimator from a perturbed Pareto distribution. The existing estimators and the proposed ones are compared through a simulation study under identical conditions. The results show that the performance of the estimators depend on the percentage of censoring, the underlying distribution, the size of extreme value index and the number of top order statistics. Overall, we found the proposed estimator from the perturbed Pareto distribution to be robust to censoring, size of the extreme value index and the number of top order statistics.

Malaria remains a serious health challenge in the Comoros Islands, despite ongoing control efforts. Past studies have shown reductions in cases due to prevention and treatment measures, but little work has been done to forecast future malaria deaths and assess the long-term impact of these measures. Malaria mortality data from 1990 to 2019 were analyzed using an ARIMA(1,0,0) model. The model was validated through diagnostic tests, ensuring reliability for forecasting trends. The study confirmed significant reductions in malaria cases, such as in Grand Comoro, where cases fell from 235.36 to 5.47 per 1,000 people. The ARIMA model predicted that fatalities will remain low if current control measures, including bed nets, indoor spraying, and mass drug administration, are sustained. The findings highlight the success of these interventions in reducing malaria mortality. However, challenges like drug and insecticide resistance and financial limitations pose risks to further progress. Continued support and adaptation of strategies are essential to address these challenges and sustain low malaria mortality rates. The study demonstrates the effectiveness of malaria control efforts in the Comoros and underscores the importance of maintaining these measures to achieve malaria elimination and improve public health outcomes.

In this paper we consider the semiclassical version of pseudo-differential operators on the lattice space $\hbar \mathbb{Z}^n$. The current work is an extension of a previous work and agrees with it in the limit of the parameter $\hbar \rightarrow 1$. The various representations of the operators will be studied as well as the composition, transpose, adjoint and the link between ellipticity and parametrix of operators. We also give the conditions for the $\ell^p(\hbar \mathbb{Z}^n)$, weighted $\ell^2(\hbar \mathbb{Z}^n)$ boundedness and $\ell^p(\hbar \mathbb{Z}^n)$ compactness of operators. We investigate the relation between the classical and semi-classical quantization and employ its applications to Schatten-Von Neumann classes on $\ell^2( \hbar \mathbb{Z}^n)$. We establish Gårding and sharp Gårding inequalities, with an application to the well-posedness of parabolic equations on the lattice $\hbar \mathbb{Z}^n$. Finally we verify that in the limiting case where $\hbar \rightarrow 0$ the semi-classical calculus of pseudo-differential operators recovers the classical Euclidean calculus, but with a twist.

In late 2019, a novel coronavirus, the SARS-CoV-2 outbreak was identified in Wuhan, China and later spread to every corner of the globe. Whilst the number of infection-induced deaths in Ghana, West Africa are minimal when compared with the rest of the world, the impact on the local health service is still significant. Compartmental models are a useful framework for investigating transmission of diseases in societies. To understand how the infection will spread and how to limit the outbreak. We have developed a modified SEIR compartmental model with nine compartments (CoVCom9) to describe the dynamics of SARS-CoV-2 transmission in Ghana. We have carried out a detailed mathematical analysis of the CoVCom9, including the derivation of the basic reproduction number, $\mathcal{R}_{0}$. In particular, we have shown that the disease-free equilibrium is globally asymptotically stable when \mathcal{R}_{0}<1 via a candidate Lyapunov function. Using the SARS-CoV-2 reported data for confirmed-positive cases and deaths from March 13 to August 10, 2020, we have parametrised the CoVCom9 model. The results of this fit show good agreement with data. We used Latin hypercube sampling-rank correlation coefficient (LHS-PRCC) to investigate the uncertainty and sensitivity of $\mathcal{R}_{0}$ since the results derived are significant in controlling the spread of SARS-CoV-2. We estimate that over this five month period, the basic reproduction number is given by $\mathcal{R}_{0} = 3.110$, with the 95\% confidence interval being $2.042 \leq \mathcal{R}_0 \leq 3.240$, and the mean value being $\mathcal{R}_{0}=2.623$. Of the 32 parameters in the model, we find that just six have a significant influence on $\mathcal{R}_{0}$, these include the rate of testing, where an increasing testing rate contributes to the reduction of $\mathcal{R}_{0}$.

In this article, we prove a local large deviation principle (LLDP) for the empirical locality measure of typed random networks on $n$ nodes conditioned to have a given \emph{ empirical type measure} and \emph{ empirical link measure.} From the LLDP, we deduce a full large deviation principle for the typed random graph, and the classical Erdos-Renyi graphs, where $nc/2$ links are inserted at random among $n$ nodes. No topological restrictions are required for these results.