The role of xenobiotic disruption of microbiota, corresponding dysbiosis, and potential links to host metabolic diseases are of critical importance. In this study, we used a widely prescribed antipsychotic drug, risperidone, known to influence weight gain in humans, to induce weight gain in C57BL/6J female mice. We hypothesized that microbes essential for maintaining gut homeostasis and energy balance would be depleted following treatment with risperidone, leading to enhanced weight gain relative to controls. Thus, we performed metagenomic analyses on stool samples to identify microbes that were excluded in risperidone-treated animals but remained present in controls. We identified multiple taxa including Limosilactobacillus reuteri as a candidate for further study. Oral supplementation with L. reuteri protected against risperidone-induced weight gain (RIWG) and was dependent on cellular production of a specialized metabolite, reutericyclin. Further, synthetic reutericyclin was sufficient to mitigate RIWG. Both synthetic reutericyclin and L. reuteri restored energy balance in the presence of risperidone to mitigate excess weight gain and induce shifts in the microbiome associated with leanness. In total, our results identify reutericyclin production by L. reuteri as a potential probiotic to restore energy balance induced by risperidone and to promote leanness.

Accurate analytical methods are crucial to assess human exposure to micro- and nanoplastics (MNPs). Quantitative pyrolysis-gas chromatography coupled with mass spectrometry (Py-GC-MS) has recently been used for quantifying MNPs in human blood. However, pyrolysis introduces complex effects such as secondary reactions between matrix compounds and polymers. This work introduces a non-targeted and multivariate approach to improve the identification and quantification of polyethylene (PE), poly(vinyl chloride) (PVC) and polyethylene terephthalate (PET). After spiking of extracted blood samples, PARADISe was used for componentization and integration of 417 features detected with Py-GC-MS. Quantification based on multivariate calibration models demonstrated a superior performance when compared to univariate regression. Feature selection approaches were used to identify optimal feature subsets, which reduced quantification errors by 30 % for PE, 10 % for PVC and 38 % for PET. In addition, chemical insight into pyrolysis processes was obtained by studying the matrix effects (MEs) of blood. The pyrolysis of PE and PVC appeared to be minimally affected (MEs = 81-154 %), while PET exhibited complex interactions with the matrix (MEs = 40-9031 %), impacting its quantification accuracy. In conclusion, this research highlights the importance of accounting for secondary effects during pyrolysis and introduces a multivariate approach for more accurate and robust quantification of MNPs in blood.

Understanding access to and use of antibiotics in livestock production systems is critical for guiding antimicrobial stewardship programmes and animal health services. We analysed antibiotic use practices among smallholder-intensive poultry farms in Kenya and characterised access to veterinary supply chains by calculating travel time to drug stores. Data were collected from 766 poultry farms across 15 Kenyan counties, representing all production types, between May 2021 and February 2022. We also collected antibiotic sales and geolocation data from 321 veterinary drug stores in Nakuru and Kilifi counties, representing areas with high and low-intensity poultry production, respectively. Using a machine learning framework, we predicted farm-level antibiotic use based on collected demographic and production traits. We also built geospatial models to characterise farmer travel time to drug stores with motorised transport. Half of farms used antibiotics at least once in the last two months, mostly for self-administered therapeutic reasons. Random forest analysis predicted that farms using disinfectants in cleaning, keeping other poultry species, with rodents in the chicken house and vaccinating their birds had the highest likelihood of antibiotic use. 95.4 % of farmers lived within one hour of a veterinary drug store, with 40 % residing within 15 min. Antibiotic use is integrated in smallholder poultry production, emphasising the need for prioritizing biosecurity, regulatory and socio-behavioural interventions, and economic incentives to enhance stewardship. Spatial maps suggests both risks and opportunities for antibiotic access and veterinary care.

The aim of this study was to explore and develop a preoperative and noninvasive model for predicting spread through air spaces (STAS) status in lung adenocarcinoma (LUAD) with diameter ≤ 3 cm.

This multicenter retrospective study included 640 LUAD patients. Center I included 525 patients (368 in the training cohort and 157 in the validation cohort); center II included 115 patients (the test cohort). We extracted radiomics features from the intratumor, extended tumor and peritumor regions. Multivariate logistic regression and boruta algorithm were used to select clinical independent risk factors and radiomics features, respectively. We developed a clinical model and four radiomics models (the intratumor model, extended tumor model, peritumor model and fusion model). A nomogram based on prediction probability value of the optimal radiomics model and clinical independent risk factors was developed to predict STAS status.

Maximum diameter and nodule type were clinical independent risk factors. The extended tumor model achieved satisfactory STAS status discrimination performance with the AUC of 0.74, 0.71 and 0.80 in the three cohorts, respectively, performed better than other radiomics models. The integrated discrimination improvement value revealed that the nomogram outperformed compared to the clinical model with the value of 12 %. Patients with high nomogram score (≥ 77.31) will be identified as STAS-positive.

Peritumoral information is significant to predict STAS status. The nomogram based on the extended tumor model and clinical independent risk factors provided good preoperative prediction of STAS status in LUAD with diameter ≤ 3 cm, aiding surgical decision-making.

Timely and accurate outcome prediction is essential for clinical decision-making for ischemic stroke patients in the intensive care unit (ICU). However, the interpretation and translation of predictive models into clinical applications are equally crucial. This study aims to develop an interpretable machine learning (IML) model that effectively predicts in-hospital mortality for ischemic stroke patients.

In this study, an IML model was developed and validated using multicenter cohorts of 3225 ischemic stroke patients admitted to the ICU. Nine machine learning (ML) models, including logistic regression (LR), K-nearest neighbors (KNN), naive Bayes (NB), decision tree (DT), support vector machine (SVM), random forest (RF), XGBoost, LightGBM, and artificial neural network (ANN), were developed to predict in-hospital mortality using data from the MIMIC-IV and externally validated in Shanghai Changhai Hospital. Feature selection was conducted using three algorithms. Model's performance was assessed using area under the receiver operating characteristic (AUROC), accuracy, sensitivity, specificity and F1 score. Calibration curve and Brier score were used to evaluate the degree of calibration of the model, and decision curve analysis were generated to assess the net clinical benefit. Additionally, the SHapley Additive exPlanations (SHAP) method was employed to evaluate the risk of in-hospital mortality among ischemic stroke patients admitted to the ICU.

Mechanical ventilation, age, statins, white blood cell, blood urea nitrogen, hematocrit, warfarin, bicarbonate and systolic blood pressure were selected as the nine most influential variables. The RF model demonstrated the most robust predictive performance, achieving AUROC values of 0.908 and 0.858 in the testing set and external validation set, respectively. Calibration curves also revealed a high consistency between observations and predictions. Decision curve analysis showed that the model had the greatest net benefit rate when the prediction probability threshold is 0.10 ∼ 0.80. SHAP was employed to interpret the RF model. In addition, we have developed an online prediction calculator for ischemic stroke patients.

This study develops a machine learning-based calculator to predict the probability of in-hospital mortality among patients with ischemic stroke in ICU. The calculator has the potential to guide clinical decision-making and improve the care of patients with ischemic stroke by identifying patients at a higher risk of in-hospital mortality.

Neutrophil extracellular traps (NETs) play a central role in chronic airway diseases. However, the precise genetic basis linking NETs to the development of severe asthma remains elusive. This study aims to unravel the molecular characterization of NET-related genes (NRGs) in severe asthma and to reliably identify relevant molecular clusters and biomarkers. We analyzed RNA-seq data from the Gene Expression Omnibus database. Interaction analysis revealed fifty differentially expressed NRGs (DE-NRGs). Subsequently, the non-negative matrix factorization algorithm categorized samples from severe asthma patients. A machine learning algorithm then identified core NRGs that were highly associated with severe asthma. DE-NRGs were correlated and subjected to protein-protein interaction analysis. Unsupervised consensus clustering of the core gene expression profiles delineated two distinct clusters (C1 and C2) characterizing severe asthma. Functional enrichment highlighted immune-related pathways in the C2 cluster. Core gene selection included the Boruta algorithm, support vector machine, and least absolute contraction and selection operator algorithms. Diagnostic performance was assessed by receiver operating characteristic curves. This study addresses the molecular characterization of NRGs in adult severe asthma, revealing distinct clusters based on DE-NRGs. Potential biomarkers (TIMP1 and NFIL3) were identified that may be important for early diagnosis and treatment of severe asthma.

Despite optimal antimicrobial therapy, the treatment failure rate of hospital-acquired pneumonia (HAP) routinely reaches 40% in critically ill patients. Subphenotypes have been identified within sepsis and acute respiratory distress syndrome with important predictive and possibly therapeutic implications. We derived prognosis subphenotypes for HAP and explored whether they were associated with biological markers and response to treatment.

We separately analysed data from four cohorts of critically ill patients in France (PNEUMOCARE, n = 511, ATLANREA, n = 401), Netherlands (MARS, n = 1351) and Europe-South America (ENIRRI, n = 900) to investigate HAP heterogeneity using unsupervised clustering based on clinical and routine biological variables available at HAP diagnosis. Then, we developed a machine learning-based workflow to create a simplified classification model using discovery data sets. This model was validated by applying it to an independent replication data set from an international randomized clinical trial comparing linezolid and tedizolid for the treatment of HAP (VITAL, n = 726 patients). The primary outcome was the association of subphenotypes with 28-day all-cause mortality. Secondary analyses included subphenotype associations with treatment failure at test-of-cure, respiratory microbiome and cytokine profiles in the ATLANREA subgroup, and treatment response in the VITAL trial.

We tested twelve metrics and determined that a two-cluster model best fits all cohorts. HAP subphenotype 2 had greater disease severity, lower body temperature, and worse PaO2/FiO2 ratio than subphenotype 1 patients. Although the prevalence of subphenotype 2 ranged from 26.9 to 66.9% across the four derivation cohorts, the rates of 28-day mortality and treatment failure at test-of-cure were consistently higher to subphenotype 1 (p < 0.01 for all comparisons). Subphenotype 2 was associated with greater respiratory microbiome dysbiosis and higher levels of proinflammatory cytokines in the ATLANREA cohort, as well as with statistically significant tedizolid effect modification in the VITAL trial (Relative Risk of treatment failure with tedizolid = 1.52; 95% CI 1.12-2.06 in subphenotype 1 vs. = 0.98; 95% CI 0.7-1.38 in subphenotype 2).

We identified two robust clinical subphenotypes by extensively analyzing HAP data sets. Their associations with respiratory microbiome composition, systemic inflammation, and treatment efficacy in independent data sets highlight their potential for prognostic value and predictive enrichment in future clinical trials aimed at personalized therapies.

Efficiently assessing glucose handling capacity is a critical public health challenge. This study assessed the utility of relatively easy-to-measure continuous glucose monitoring (CGM)-derived indices in estimating glucose handling capacities calculated from resource-intensive clamp tests.

We conducted a prospective study of 64 individuals without prior diabetes diagnosis. The study performed CGM, oral glucose tolerance tests (OGTT), and hyperglycemic and hyperinsulinemic-euglycemic clamp tests. We validated CGM-derived indices characteristics using an independent dataset from another country and mathematical models with simulated data.

A CGM-derived index reflecting the autocorrelation function of glucose levels (AC_Var) is significantly correlated with clamp-derived disposition index (DI), a well-established measure of glucose handling capacity and predictor of diabetes onset. Multivariate and machine learning models indicate AC_Var's contribution to predicting clamp-derived DI independent from other CGM-derived indices. The model using CGM-measured glucose standard deviation and AC_Var outperforms models using commonly used diabetes diagnostic indices, such as fasting blood glucose, HbA1c, and OGTT measures, in predicting clamp-derived DI. Mathematical simulations also demonstrate the association of AC_Var with DI.

CGM-derived indices, including AC_Var, serve as valuable tools for predicting glucose handling capacities in populations without prior diabetes diagnosis. We develop a web application that calculates these CGM-derived indices ( https://cgm-ac-mean-std.streamlit.app/ ).

Assessing the generalizability of randomized clinical trials (RCTs) to real-world patients remains challenging. We propose a multidimensional metric to quantify the representativeness of an RCT cohort in an electronic health record (EHR) population and estimate real-world effects based on individualized treatment effects observed in the RCT.

We identified 65 clinical prerandomization characteristics of patients with heart failure with preserved ejection fraction within the TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist Trial) and extracted those features in similar patients in EHR data from 4 hospitals in the Yale New Haven Health System. We then assessed the real-world generalizability of TOPCAT by developing a novel statistic, the phenotypic distance metric, to quantify the representation of TOPCAT participants within EHR patients. Finally, applying a machine learning method to learn individualized treatment effect in TOPCAT participants stratified by region, the United States and Eastern Europe, we predicted spironolactone benefit within the EHR cohorts.

There were 3445 patients in TOPCAT (median age 69, interquartile range [IQR], 61-76 years, 52% female) and 8121 patients with heart failure with preserved ejection fraction across 4 hospitals (median age range 77, IQR, 68-86; years to 85; IQR, 77-91 years, 54% to 62% female). Across covariates, the EHR patients were more similar to each other than the TOPCAT-US participants (median standardized mean difference 0.065, IQR, 0.011-0.144 versus median standardized mean difference 0.186, IQR, 0.040-0.479). The phenotypic distance metric found a higher generalizability of the TOPCAT-US participants to the EHR patients than the TOPCAT-EE participants. Using a TOPCAT-US-derived model of individualized treatment effect, all EHR patients were predicted to derive benefit from spironolactone treatment, while a TOPCAT-EE-derived model predicted 13% of EHR patients to derive benefit.

This novel multidimensional metric evaluates the real-world representativeness of RCT participants against corresponding patients in the EHR, enabling the evaluation of an RCT's implication for real-world patients.

Multiple machine learning techniques were employed to identify key long non-coding RNA (lncRNA) biomarkers associated with cuproptosis in Diffuse Large B-Cell Lymphoma (DLBCL). Data from the TCGA and GEO databases facilitated the identification of 126 significant cuproptosis-related lncRNAs. Various feature selection methods, such as Univariate Filtering, Lasso, Boruta, and Random Forest, were integrated with a Transformer-based model to develop a robust prognostic tool. This model, validated through fivefold cross-validation, demonstrated high accuracy and robustness in predicting risk scores. MALAT1 was pinpointed using permutation feature importance from machine learning methods and was further validated in DLBCL cell lines, confirming its substantial role in cell proliferation. Knockdown experiments on MALAT1 led to reduced cell proliferation, underscoring its potential as a therapeutic target. This integrated approach not only enhances the precision of biomarker identification but also provides a robust prognostic model for DLBCL, demonstrating the utility of these lncRNAs in personalized treatment strategies. This study highlights the critical role of combining diverse machine learning methods to advance DLBCL research and develop targeted cancer therapies.

Residential fine particulate matter (PM2.5) poses a substantial health hazard, yet disparities across urban and peri-urban areas remain unclear in China due to the absence of systematic monitoring data. Key knowledge gaps persist regarding the characteristics, influencing factors, and health burdens of residential PM2.5 pollution, particularly in peri-urban regions. Between 2018 and 2019, 746 urban and 339 peri-urban households in 15 typical Chinese cities were randomly selected through the Chinese Indoor Environment and Health Surveillance Project (CIEHS). The relationships between influencing factors and residential PM2.5 were assessed using a general linear model (GLM), and factors importance was identified using the Boruta algorithm with 10-fold cross-validation and 100 replications. Disability-adjusted life-years (DALYs) attributable to the residential PM2.5 were estimated using the population-attributable fraction method. The mean residential PM2.5 concentration was 62.2 μg/m3, with higher levels in peri-urban (Odds ratio [OR] = 1.0705 95 % Confidence interval [95 % CI]: 1.0615, 1.0795) and during cold seasons (OR = 1.3499 95 %CI: 1.3388, 1.3610). Environmental factors, building conditions, family-related information, and lifestyle behaviors were all associated with residential PM2.5, notably, with the five most influential factors remaining consistent across urban and peri-urban areas. The mean DALY rate was 2160 per 100,000 in urban areas, slightly higher at 2186 per 100,000 in peri-urban areas. According to China's seventh national census, residential PM2.5 exposure accounted for 19.48 million urban and 11.14 million peri-urban DALYs nationally. These findings underscore the urgency for differentiated environmental management and targeted mitigation policies to address the disproportionate health burden in peri-urban areas.

Establishing an accurate prognosis remains challenging in older patients with cancer because of the population's heterogeneity and the current predictive models' reduced ability to capture the complex interactions between oncologic and geriatric predictors. We aim to develop and externally validate a new predictive score (the Geriatric Cancer Scoring System [GCSS]) to refine individualized prognosis for older patients with cancer during the first year after a geriatric assessment (GA).

Data were collected from two French prospective multicenter cohorts of patients with cancer 70 years and older, referred for GA: ELCAPA (training set January 2007-March 2016) and ONCODAGE (validation set August 2008-March 2010). Candidate predictors included baseline oncologic and geriatric factors and routine biomarkers. We built predictive models using Cox regression, single decision tree (DT), and random survival forest (RSF) methods, comparing their predictive performance for 3-, 6-, and 12-month mortalities by computing time-dependent area under the receiver operator curve (tAUC).

A total of 2,012 and 1,397 patients were included in the training and validation set, respectively (mean age: 81 ± 6 years/78 ± 5 years; women: 47%/70%; metastatic cancer: 50%/34%; 12-month mortality: 43%/16%). Tumor site/metastatic status, cancer treatment, weight loss, ≥five prescription drugs, impaired functional status and mobility, abnormal G-8 score, low creatinine clearance, and elevated C-reactive protein (CRP)/albumin were identified as relevant predictors in the Cox model. DT and RSF identified more complex combinations of features, with G-8 score, tumor site/metastatic status, and CRP/albumin ratio contributing most to the predictions. The RSF approach gave the highest tAUC (12 months: 0.87 [RSF], 0.82 [Cox], 0.82 [DT]) and was retained as the final model.

The GCSS on the basis of a machine learning approach applied to two large French cohorts gave an accurate externally validated mortality prediction. The GCSS might improve decision making and counseling in older patients with cancer referred for pretherapeutic GA. GCSS's generalizability must now be confirmed in an international setting.

The emergence of Batrachochytrium salamandrivorans (Bsal) poses an imminent threat to caudate biodiversity worldwide, particularly through anthropogenic-mediated means such as the pet trade. Bsal is a fungal panzootic that has yet to reach the Americas, Africa, and Australia, presenting a significant biosecurity risk to naïve amphibian populations lacking the innate immune defenses necessary for combating invasive pathogens. We explored the capability of near-infrared spectroscopy (NIRS) coupled with predictive modeling as a rapid, non-invasive Bsal screening tool in live caudates. Using eastern newts (Notopthalmus viridescens) as a model species, NIR spectra were collected in tandem with dermal swabs used for confirmatory qPCR analysis. We identified that spectral profiles differed significantly by physical location (chin, cloaca, tail, and foot) as well as by Bsal pathogen status (control vs. exposed individuals; p < 0.05). The support vector machine algorithm achieved a mean classification accuracy of 80% and a sensitivity of 92% for discriminating Bsal-control (-) from Bsal-exposed (+) individuals. This approach offers a promising method for identifying Bsal-compromised populations, potentially aiding in early detection and mitigation efforts alongside existing techniques.

Predicting disease progression in metabolic dysfunction-associated steatotic liver disease (MASLD) is challenging, and current non-invasive tests (NITs) lack the precision to replace liver biopsy. This study aimed to identify plasma biomarkers for different stages of fibrosis using affinity-based proteomics in two biopsy-proven cohorts. The primary objective was to identify biomarkers capable of distinguishing between low-to-no fibrosis (F0-1) and significant fibrosis (F2-4) in MASLD.

Participants in the discovery cohort were recruited from Uppsala University Hospital and Swedish CArdioPulmonary bioImage Study (SCAPIS), while the validation cohort was included from Linköping University Hospital. All participants diagnosed with MASLD underwent liver biopsy and were categorized by fibrosis stage (F0-1 or F2-4). A total of 276 plasma proteins were analyzed using Olink® panels, with biomarkers identified through ordinal logistic regression, random forest (RF) analysis and the Boruta algorithm.

The discovery cohort included 60 participants, with 60% having fibrosis stage F0-1 and 40% having F2-4. The validation cohort had 59 participants, of whom 35 had fibrosis stage F0-1 (59.3%) and 24 had stage F2-4 (40.7%). Five biomarkers were significantly associated with fibrosis stage in the discovery cohort, with four confirmed in the validation cohort. A model combining angiotensin converting enzyme-2 (ACE2), hepatocyte growth factor (HGF) and insulin-like growth factor-binding protein-7 (IGFBP-7) demonstrated strong predictive performance for significant fibrosis (c-statistics 0.82-0.83), outperforming fibrosis-4 (FIB-4) (c-statistics 0.61-0.72).

A biomarker model including ACE2, HGF and IGFBP7 shows promise in distinguishing between low-stage and significant fibrosis.

This study aimed to identify dosiomic features that have a significant impact on biochemical failure (BCF) following low-dose rate (LDR) brachytherapy treatment using Iodine-125 seeds for prostate cancer and to provide insights into LDR brachytherapy treatment efficacy using a dosiomic approach.

Between January 2005 and February 2015, 1,205 patients with localized prostate cancer underwent Iodine-125 seed implantation without combined external irradiation. A total of 96 patients were selected for this study, including 48 with BCF and 48 without BCF. The patients were divided into two cohorts: derivation and validation. Dose distribution images (DDs) were calculated from computed tomography (CT) images taken one month after implantation. A total of 1,130 dosiomic features, including shape-and-size, histogram, and texture features, were extracted from these DDs, their wavelet-transformed images, and Laplacian-of-Gaussian (LoG)-filtered images. The features obtained were categorized into three groups: shape-and-size (S), histogram (H), and texture (T). The Boruta algorithm was used to eliminate less important features. Two analyses were performed: Analysis A performed a multivariate logistic regression analysis using data from the validation cohort to identify significant features. Analysis B generated logistic regression models using derivation cohort data. The accuracy of BCF prediction was assessed using the validation cohort, with performance measured using the area under the receiver operating characteristic curve (AUC).

After the feature reduction process, two, two, and four features remained in the S, H, and T feature groups, respectively. In analysis A, the multivariate logistic regression identified four dominant features, two from each of the S and T groups. In analysis B, the AUC of the logistic regression prediction models using S, H, and all four features were 0.81, 0.77, and 0.86, respectively.

Four significant dosiomic features were identified. Notably, three features-elongation, Maximum2DDiameterRow, and wavelet-HHL_Skewness-strongly distinguished patients with favorable prognoses from others. These findings suggest that dosiomic features from postimplant CT and dose distribution may serve as effective factors for evaluating LDR brachytherapy outcomes in patients with prostate cancer.

The design of sampling methods is crucial in digital soil mapping for soil organic carbon (SOC), as it directly affects prediction precision and reliability. While sampling methods based on environmental variables are widely used, the spatial heterogeneity of soil properties poses challenges by introducing variability in influential driving factors across subregions, potentially reducing prediction accuracy. To address this, a partitioned conditioned Latin hypercube sampling (PcLHS) method explicitly considering spatial heterogeneity is proposed. PcLHS first employs the regionalization with dynamically constrained agglomerative clustering and partitioning (REDCAP) method to partition the study area into relatively homogeneous subregions. Key environmental variables are then identified using the Boruta and the Variance Inflation Factor method, followed by conditioned Latin hypercube sampling (cLHS) to select training points within each subregion. Finally, the selected training points are combined to form the complete training dataset. A case study on SOC sampling in northeastern France demonstrated that PcLHS consistently outperformed traditional sampling methods, achieving lower root mean square error (RMSE, 0.40-0.43), higher coefficient of determination (R2, 0.36-0.44), and improved concordance correlation coefficient (CCC, 0.58-0.63). Compared to other methods, PcLHS reduced RMSE by 4-11%, increased R2 by 18-46%, and improved CCC by 14-29%. These results highlight the necessity of considering spatial heterogeneity in soil sampling design and establish PcLHS as an effective method for SOC prediction in heterogeneous landscapes.

Insulin resistance (IR) is an important pathologic component in the occurrence and development of cardiovascular disease (CVD). The estimated glucose disposal rate (eGDR) is a measure of glucose handling capacity, that has demonstrated utility as a reliable marker of IR. The study aimed to determine the predictive utility of IR assessed by eGDR for CVD risk.

This nationwide prospective cohort study utilized data of 6416 participants from the China Health and Retirement Longitudinal Study (CHARLS) who were free of CVD but had complete data on eGDR at baseline. The Boruta algorithm was performed for feature selection. Multivariate Cox proportional hazards regression models and restricted cubic spline (RCS) analysis were conducted to examine the associations between eGDR and CVD, and the results were expressed with hazard ratio (HR) and 95% confidence interval (CI) values. The area under the receiver operating characteristic (ROC) curve (AUC), calibration curve, Hosmer-Lemeshow test, net reclassification improvement (NRI), and decision curve analysis (DCA) were employed to evaluate the clinical efficacy of eGDR in identifying CVD. Subgroup analysis was performed to explore the potential association of with CVD in different populations.

During a median follow-up of 106.5 months, 1339 (20.87%) incident CVD cases, including 1025 (15.96%) heart disease and 439 (6.84%) stroke, were recorded from CHARLS. The RCS curves demonstrated a significant and linear relationship between eGDR and all endpoints (all P for nonlinear > 0.05). After multivariate adjustment, the lower eGDR levels were found to be significantly associated with a greater prevalence of CVD. Compared to the lowest quartile, the highest eGDR quartile was associated with a decreased risk of CVD (HR 0.686, 95% CI 0.545-0.862). When assessed as a continuous variable, individuals with a unit increasement in eGDR was related to a 21.2% (HR 0.788, 95% CI 0.669-0.929) lower risk of CVD, a 18.3% (HR 0.817, 95% CI 0.678-0.985) decreased risk of heart disease, and 39.5% (HR 0.705, 95% CI 0.539-0.923) lower risk of stroke. The eGDR had an excellent predictive performance according to the results of ROC (AUC = 0.712) and χ2 likelihood ratio test (χ2 = 4.876, P = 0.771). NRI and DCA analysis also suggested the improvement from eGDR to identify prevalent CVD and the favorable clinical efficacy of the multivariate model. Subgroup analysis revealed that the trend in incident CVD risk were broadly consistent with the main results across subgroups.

A lower level of eGDR was found to be associated with increased risk of incident CVD, suggesting that eGDR may serve as a promising and preferable predictor for CVD.

Formulas for the extraction of continuous and binary effect sizes that are entered into a meta-analysis are readily available. Only some formulas for the extraction of count outcomes have been presented previously. The purpose of this methodological article is to present formulas for extracting effect sizes and their standard errors for studies of count outcomes with person-time denominators.

Formulas for the calculation of the number of events in a study and the corresponding person time in which these events occurred are presented. These formulas are then used to estimate the relevant effect sizes and standard errors of interest. These effect sizes are rates, rate ratios and rate differences for a two-group comparison and rate ratios and rate differences for a difference-in-difference design.

Two studies from the field of suicide prevention are used to demonstrate the extraction of the information required to estimate effect sizes and standard errors. In the first example, the rate ratio for a two-group comparison was 0.957 (standard error of the log rate ratio, 0.035), and the rate difference was -0.56 per 100,000 person years (standard error 0.44). In the second example, the rate ratio for a difference-in-difference analysis was 0.975 (standard error of the log rate ratio 0.036) and the rate difference was -0.30 per 100,000 person years (standard error 0.42).

The application of these formulas enables the calculation of effect sizes that may not have been presented in the original study. This reduces the need to exclude otherwise eligible studies from a meta-analysis, potentially reducing one source of bias.

As global vaccination rates have reached their lowest point in nearly 15 years, effective interventions are being required globally to promote vaccination; however, there is a lack of rigorous evaluation of the effect of various interventions. Through a global synthesis, we analysed data from approximately 6 125 795 participants across 319 studies in 41 countries to reveal the global landscape of four intervention themes and to assess their effectiveness in increasing vaccination rates. We found an overall positive effect of the interventions across four main themes on improving vaccination. Specifically, dialogue-based interventions increased vaccination rates by 43.1% (95% CI: 29.8 to 57.9%, with effect sizes measured as relative risks (RRs)), though they may not always be effective in adolescents or in the sample with a higher percentage of male participants. Incentive-based interventions, whether implemented alone or combined with other intervention themes, failed to demonstrate a significant effect in children. Reminder/recall-based interventions were also effective for promoting vaccination (38.5% increase, 95% CI: 28.9 to 48.9%), particularly for completing vaccine series. Multi-component interventions exhibited excellent effectiveness in vaccination (54.3% increase, 95% CI: 40.5 to 69.6%), with the combination of dialogue, incentive and reminder/recall proving more effective than other multi-component interventions, but showing no significant effects in populations with high initial vaccination rates. However, we found that in most cases combining additional interventions with a single intervention may not significantly improve their effectiveness, especially for incentive-based interventions, but dialogue-based and reminder/recall-based interventions appear to be beneficial in some specific combinations. These findings underscore the importance of governments, public health officials and advocacy groups implementing appropriate vaccine interventions by selecting interventions tailored to specific populations, strategically promoting the completion of vaccine series and effectively combining interventions to promote global vaccination and save more lives.

Large presence of unlicensed powered two- and three-wheeler (PTW) drivers in China pose a significant threat to road safety. In this study, a customized Deep Forest Model (DF-ptw) is constructed to investigate the effect of unlicensed PTW drivers on crash severity in two-vehicle crashes, using a recent 3-year historical crash data. SHapley Additive explanation (SHAP) and Partial Dependence Plot (PDP) analysis reveal that unlicensed motorcyclists are significantly more likely to suffer serious injuries in two-vehicle crashes compared to unlicensed auto-rickshaw drivers. Additionally, factors such as drunk driving, fatigued driving, and being an unlicensed driver over the age of 53 notably elevate the risk of serious injury or death, with unlicensed motorcyclists being disproportionately affected. Moreover, self-employed unlicensed PTW drivers face a higher probability of serious injury or fatality in crashes compared to farmers, blue-collar, and white-collar workers. Unlicensed PTW drivers are also more susceptible to severe or fatal injuries on national and provincial roads, in low visibility conditions, during late-night hours, on non-separated roads, and at dusk or dawn. Based on these findings, this study proposes to reduce the frequency and severity of crashes involving unlicensed PTW drivers by focusing on more stringent eligibility checks, increasing safety awareness, and implementing advanced safety measures.

Investigating brain-enriched proteins with machine learning methods may enable a brain-specific understanding of brain aging and provide insights into the molecular mechanisms and pathological pathways of dementia. The study aims to analyze associations of brain-specific plasma proteomic aging signature with risks of incident dementia. In 45,429 dementia-free UK Biobank participants at baseline, we generated a brain-specific biological age using 63 brain-enriched plasma proteins with machine learning methods. The brain age gap was estimated, and Cox proportional hazards models were used to study the association with incident all-cause dementia, Alzheimer's disease (AD), and vascular dementia. Per-unit increment in the brain age gap z-score was associated with significantly higher risks of all-cause dementia (hazard ratio [95% confidence interval], 1.67 [1.56-1.79], P < 0.001), AD (1.85 [1.66-2.08], P < 0.001), and vascular dementia (1.86 [1.55-2.24], P < 0.001), respectively. Notably, 2.1% of the study population exhibited extreme old brain aging defined as brain age gap z-score > 2, correlating with over threefold increased risks of all-cause dementia and vascular dementia (3.42 [2.25-5.20], P < 0.001, and 3.41 [1.05-11.13], P = 0.042, respectively), and fourfold increased risk of AD (4.45 [2.32-8.54], P < 0.001). The associations were stronger among participants with healthier lifestyle factors (all P-interaction < 0.05). These findings were corroborated by magnetic resonance imaging assessments showing that a higher brain age gap aligns global pathophysiology of dementia, including global and regional atrophy in gray matter, and white matter lesions (P < 0.001). The brain-specific proteomic age gap is a powerful biomarker of brain aging, indicative of dementia risk and neurodegeneration.

The prognostication of long-term functional outcomes remains challenging in patients with traumatic brain injury (TBI). Our aim was to demonstrate that intensive care unit (ICU) variables are not efficient to predict 6-month functional outcome in survivors with moderate to severe TBI (msTBI) but are mostly associated with mortality, which leads to a mortality bias for models predicting a composite outcome of mortality and severe disability.

We analyzed the data from the multicenter randomized controlled Continuous Hyperosmolar Therapy in Traumatic Brain-Injured Patients trial and developed predictive models using machine learning methods and baseline characteristics and predictors collected during ICU stay. We compared our models' predictions of 6-month binary Glasgow Outcome Scale extended (GOS-E) score in all patients with msTBI (unfavorable GOS-E 1-4 vs. favorable GOS-E 5-8) with mortality (GOS-E 1 vs. GOS-E 2-8) and binary functional outcome in survivors with msTBI (severe disability GOS-E 2-4 vs. moderate to no disability GOS-E 5-8). We investigated the link between ICU variables and long-term functional outcomes in survivors with msTBI using predictive modeling and factor analysis of mixed data and validated our hypotheses on the International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT) model.

Based on data from 370 patients with msTBI and classically used ICU variables, the prediction of the 6-month outcome in survivors was inefficient (mean area under the receiver operating characteristic 0.52). Using factor analysis of mixed data graph, we demonstrated that high-variance ICU variables were not associated with outcome in survivors with msTBI (p = 0.15 for dimension 1, p = 0.53 for dimension 2) but mostly with mortality (p < 0.001 for dimension 1), leading to a mortality bias for models predicting a composite outcome of mortality and severe disability. We finally identified this mortality bias in the IMPACT model.

We demonstrated using machine learning-based predictive models that classically used ICU variables are strongly associated with mortality but not with 6-month outcome in survivors with msTBI, leading to a mortality bias when predicting a composite outcome of mortality and severe disability.

Patients with gastric cancer (GC) who experience early recurrence (ER) within 2 years postoperatively have poor prognoses. This study aimed to analyze and predict ER after curative surgery for patients with GC using machine learning (ML) methods.

This multicenter population-based cohort study included data from ten large tertiary regional medical centers in China. The clinical, pathological, and laboratory parameters were retrospectively collected from the records of 11,615 patients. The patients were randomly divided into training (70%) and test (30%) cohorts. A total of ten ML models were developed and validated to predict the ER. Model performance was evaluated using area under the receiver operating characteristic curve (AUC), calibration plots, and Brier score (BS). SHapley Additive exPlanations (SHAP) was used to rank the input features and interpret predictions.

ER was reported in 1794 patients (15%) during follow-up. The stacking ensemble model achieved AUCs of 1.0 and 0.8 in the training and testing cohorts, respectively, with a BS of 0.113. SHAP dependency plots revealed that tumor staging, elevated tumor marker levels, lymphovascular invasion, perineural invasion, and tumor size > 5 cm were associated with higher ER risk. The impact of age and the number of lymph nodes harvested on ER risk exhibited a "U-shaped distribution." Additionally, an online prediction tool based on the best model was developed to facilitate clinical applications.

We developed a robust clinical model for predicting the risk of ER after surgery for GC, which may aid in individualized clinical decision-making.

For efficient decision-making and optimal land management trajectories, information on soil properties in relation to safety guidelines should be processed from point inventories to surface predictive maps. For large-scale predictive mapping, very few practical implementations have attempted to clarify how well indicator models can be built from large covariate sets combined with spatial proxies. This paper summarizes the performance of the weighted indicator-based random forest model which was used to predict exceedance probabilities for several potentially toxic elements (PTEs) in Czech farmland. The method was implemented for data mining in the Czech high-density monitoring data which had to be firstly regressed to achieve analytical harmony, and the reliability of the regression-based harmonisation was used as the input weights for the final model. The indicator-based models were trained for each PTE (As, Be, Cd, Co, Cr, Cu, Hg, Ni, Pb, V, and Zn) with two different sets of indicators, reflecting the two-tier nature of the Czech safety guidelines, which differentiate between soil textures of topsoil. The two separate predictive outputs are combined into a single probability map using a pragmatic meta-model of linear weights derived from a soil texture map generated by a compositional spatial model. Through validation with data splitting, the accuracy of the models showed relatively high predictive power for the probability distributions, but with pronounced differences between PTEs as the root mean square error in terms of exceedance probabilities ranged from 11 % (V) to 32 % (Cd and Cr) for independent validation. In addition, models based on high-resolution auxiliary variables allowed a meaningful and quantitative identification of the most important natural and anthropogenic drivers for areas with an increased rate of non-compliance with the protection thresholds for cultivated soils. Variable importance calculations showed the dominant influence of spatially explicit covariates (represented by geographical distances to quantile-based groups of points), but still significant contributions from other predictors. Among the natural factors, lithological information came to the fore, mainly due to continuous response variables such as mineral exploration density or geophysical ancillary variables (from remotely sensed gravimetry and radiometry). Among anthropogenic factors, particulate matter in the atmosphere was identified as the most important human-related pressure, followed by several land-use effects.

Bedside lung ultrasound (LUS) and computed tomography (CT) imaging are valuable modalities in screening and diagnosis of pulmonary diseases. This study aims to investigate the prognostic value of integrating LUS and CT imaging findings with clinical features to predict poor outcomes upon ER admission in COVID-19.

Patients visiting the study center with clinical presentation and laboratory findings compatible with COVID-19 between April 2020 to January 2022 were considered for this study. Several imaging findings (ground glass opacity, consolidation, atelectatic bands, mosaic attenuation, ARDS pattern, crazy paving, pleural thickening in CT and A-line, comet-tail artifact, confluent B-Line in BLUS, pleural thickening and Consolidation in both modalities) were evaluated, alongside clinical assessments upon admission, to assess their prognostic value. The top radiological, LUS findings, and clinical signs were integrated in a nomogram for predicting mortality.

A total of 1230 patients were included in the analyses. Among the findings, consolidation in BLUS and CT imaging, and absence of A-lines were associated with mortality. In addition to these findings, ground-glass opacities, atelectatic band, mosaic attenuation, crazy paving, and confluent B-line were also associated with ICU hospitalization. Although, the prognostic value of individual markers was poor and comparable (AUC < 0.65), the combined use of top clinical and imaging findings in the associated nomogram led to a high accuracy in predicting mortality (Area under curve: 87.3%).

BLUS and CT imaging findings alone provide limited utility in stratifying patients for higher mortality and ICU admission risk and should not be used for risk stratification alone outside the context of each patient and their clinical presentations in suspected COVID-19 patients.

While the Cardiometabolic Index (CMI) serves as a novel marker for assessing adipose tissue distribution and metabolic function, its prognostic utility for cardiovascular disease (CVD) events remains incompletely understood. This investigation sought to elucidate the predictive capabilities of CMI for cardiovascular outcomes and explore underlying mechanistic pathways to establish a comprehensive risk prediction framework.

The study encompassed 7,822 individuals from a national health and retirement longitudinal cohort, with participants stratified by CMI quartiles. Following baseline characteristic comparisons and CVD incidence rate calculations, we implemented multiple Cox regression models to assess CMI's cardiovascular risk prediction capabilities. For nomogram construction, we utilized an ensemble machine learning framework, combining Boruta algorithm-based feature selection with Random Forest (RF) and XGBoost analyses to determine key predictive parameters.

Throughout the median follow-up duration of 84 months, we documented 1,500 incident CVD cases, comprising 1,148 cardiac events and 488 cerebrovascular events. CVD incidence demonstrated a positive gradient across ascending CMI quartiles. Multivariate Cox regression analysis, adjusting for potential confounders, confirmed a significant association between CMI and CVD risk. Notably, mediation analyses revealed that hypertension and glycated hemoglobin (HbA1c) potentially serve as mechanistic intermediaries in the CMI-CVD relationship. Sex-stratified analyses suggested differential predictive patterns between gender subgroups. Given CMI's robust and consistent predictive capability for stroke outcomes, we developed a machine learning-derived nomogram incorporating five key predictors: age, CMI, hypertension status, high-sensitivity C-reactive protein (hsCRP) and renal function (measured as serum creatinine). The nomogram demonstrated strong discriminative ability, achieving areas under the receiver operating characteristic curve (AUC) of 0.76 (95% CI: 0.56-0.97) and 0.74 (95% CI: 0.66-0.81) for 2-year and 6-year stroke prediction, respectively.

Our findings establish CMI as a significant predictor of cardiovascular events in the aging population, with the relationship partially mediated through hypertension and insulin resistance pathways. The validated nomogram, developed using longitudinal data from a substantial elderly cohort, incorporates CMI to enable preclinical risk stratification, supporting timely preventive strategies.

The formation of memories is a complex, multi-scale phenomenon, especially when it involves integration of information from various brain systems. We have investigated the differences between a novel and consolidated association of spatial cues and amphetamine administration, using an in situ hybridisation method to track the short-term dynamics during the recall testing. We have found that remote recall group involves smaller, but more consolidated groups of neurons, which is consistent with their specialisation. By employing machine learning analysis, we have shown this pattern is especially pronounced in the VTA; furthermore, we also uncovered significant activity patterns in retrosplenial and prefrontal cortices, as well as in the DG and CA3 subfields of the hippocampus. The behavioural propensity towards the associated localisation appears to be driven by the nucleus accumbens, however, further modulated by a trio of the amygdala, VTA and hippocampus, as the trained association is confronted with test experience. Moreover, chemogenetic analysis revealed central amygdala as critical for linking appetitive emotional states with spatial contexts. These results show that memory mechanisms must be modelled considering individual differences in motivation, as well as covering dynamics of the process.

Multiple organ dysfunction syndrome (MODS) is a critical complication in trauma-induced sepsis patients and is associated with a high mortality rate. This study aimed to develop and validate predictive models for MODS in this patient population using a nomogram and machine learning approaches.

This retrospective cohort study utilized data from the Medical Information Mart for Intensive Care-IV 2.2 database, focusing on trauma patients diagnosed with sepsis within the first day of intensive care unit (ICU) admission. Predictive variables were extracted from the initial 24 h of ICU data. The dataset (2008-2019) was divided into a training set (2008-2016) and a temporal validation set (2017-2019). Feature selection was conducted using the Boruta algorithm. Predictive models were developed and validated using a nomogram and various machine learning techniques. Model performance was evaluated based on discrimination, calibration, and decision curve analysis.

Among 1295 trauma patients with sepsis, 349 (26.95%) developed MODS. The 28-day mortality rates were 11.21% for non-MODS patients and 23.82% for MODS patients. Key predictors of MODS included the simplified acute physiology score II score, use of mechanical ventilation, and vasopressor administration. In temporal validation, all models significantly outperformed traditional scoring systems (all P <0.05). The nomogram achieved an area under the curve (AUC) of 0.757 (95% confidence interval [CI]: 0.700 to 0.814), while the random forest model demonstrated the highest performance with an AUC of 0.769 (95% CI: 0.712 to 0.826). Calibration plots showed excellent agreement between predicted and observed probabilities, and decision curve analysis indicated a consistently higher net benefit for the newly developed models.

The nomogram and machine learning models provide enhanced predictive accuracy for MODS in trauma-induced sepsis patients compared to traditional scoring systems. These tools, accessible via web-based applications, have the potential to improve early risk stratification and guide clinical decision-making, ultimately enhancing outcomes for trauma patients. Further external validation is recommended to confirm their generalizability.

This study aims to develop and validate a machine learning model for identifying individuals within the nursing population experiencing severe subjective cognitive decline (SCD) during the menopause transition, along with their associated factors.

A secondary analysis was performed using cross-sectional data from 1,264 nurses undergoing the menopause transition. The data set was randomly split into training (75%) and validation sets (25%), with the Bortua algorithm employed for feature selection. Seven machine learning models were constructed and optimized. Model performance was assessed using area under the receiver operating characteristic curve, accuracy, sensitivity, specificity, and F1 score. Shapley Additive Explanations analysis was used to elucidate the weights and characteristics of various factors associated with severe SCD.

The average SCD score among nurses in the menopause transition was (5.38 ± 2.43). The Bortua algorithm identified 13 significant feature factors. Among the seven models, the support vector machine exhibited the best overall performance, achieving an area under the receiver operating characteristic curve of 0.846, accuracy of 0.789, sensitivity of 0.753, specificity of 0.802, and an F1 score of 0.658. The two variables most strongly associated with SCD were menopausal symptoms and the stage of menopause.

The machine learning models effectively identify individuals with severe SCD and the related factors associated with severe SCD in nurses during the menopause transition. These findings offer valuable insights for the management of cognitive health in women undergoing the menopause transition.

Emerging infectious diseases are increasingly recognized as a significant threat to global biodiversity conservation. Elucidating the relationship between pathogens and the host microbiome could lead to novel approaches for mitigating disease impacts. Pathogens can alter the host microbiome by inducing dysbiosis, an ecological state characterized by a reduction in bacterial alpha diversity, an increase in pathobionts, or a shift in beta diversity. We used the snake fungal disease (SFD; ophidiomycosis), system to examine how an emerging pathogen may induce dysbiosis across two experimental scales. We used quantitative polymerase chain reaction, bacterial amplicon sequencing, and a deep learning neural network to characterize the skin microbiome of free-ranging snakes across a broad phylogenetic and spatial extent. Habitat suitability models were used to find variables associated with fungal presence on the landscape. We also conducted a laboratory study of northern watersnakes to examine temporal changes in the skin microbiome following inoculation with Ophidiomyces ophidiicola. Patterns characteristic of dysbiosis were found at both scales, as were nonlinear changes in alpha and alterations in beta diversity, although structural-level and dispersion changes differed between field and laboratory contexts. The neural network was far more accurate (99.8% positive predictive value [PPV]) in predicting disease state than other analytic techniques (36.4% PPV). The genus Pseudomonas was characteristic of disease-negative microbiomes, whereas, positive snakes were characterized by the pathobionts Chryseobacterium, Paracoccus, and Sphingobacterium. Geographic regions suitable for O. ophidiicola had high pathogen loads (>0.66 maximum sensitivity + specificity). We found that pathogen-induced dysbiosis of the microbiome followed predictable trends, that disease state could be classified with neural network analyses, and that habitat suitability models predicted habitat for the SFD pathogen.

Predicting vancomycin exposure is essential for optimizing dosing regimens in sepsis patients. While population pharmacokinetic (PPK) models are commonly used, their performance is limited. Machine learning (ML) models offer advantages over PPK models, but it remains unclear which model-PPK, Bayesian, ML, or hybrid PPK-ML-is best for predicting vancomycin exposure across different clinical scenarios in sepsis patients. This study compares the performance of these models in predicting the 24 hour area under the blood concentration curve (AUC24) to support precision dosing in sepsis care. Data from sepsis patients treated with intravenous vancomycin were sourced from the MIMIC-IV database. The data set was split into training and testing sets, and four models-PPK, Bayesian, ML, and hybrid-were developed. In the testing set, AUC24 was predicted using all models, and performance was evaluated using mean absolute error, mean squared error, root mean squared error, mean absolute percentage error (MAPE), and R². A total of 4,059 patients were included. In the absence of vancomycin concentration data, the hybrid model outperformed both PPK and Bayesian models, with MAPE improvements of 58% and 17%, respectively. When vancomycin concentration data were available, the Bayesian model demonstrated the best performance (MAPE: 13.37% vs 68.17%, 34.17%, and 28.52% for PPK, Random Forest, and hybrid models). The hybrid model is recommended to predict AUC24 when concentration data were unavailable, while the Bayesian model should be used when concentrations were available, offering robust strategies for precise vancomycin dosing in sepsis patients.

This study evaluates and compares the performance of four models-PPK, Bayesian, ML, and hybrid PPK-ML-in predicting vancomycin exposure (AUC24) in sepsis patients using real-world data from the MIMIC-IV database. These results underscore the importance of selecting appropriate models based on the availability of concentration data, providing valuable guidance for precision dosing strategies in sepsis care. This work contributes to advancing personalized vancomycin therapy, optimizing dosing regimens, and improving clinical outcomes in sepsis patients.

Machine learning is frequently used to make decisions based on big data. Among these techniques, random forest is particularly prominent. Although random forest is known to have many advantages, one aspect that is often overseen is that it is a non-deterministic method that can produce different models using the same input data. This can have severe consequences on decision-making processes. In this study, we introduce a method to quantify the impact of non-determinism on predictions, variable importance estimates, and decisions based on the predictions or variable importance estimates. Our findings demonstrate that increasing the number of trees in random forests enhances the stability in a non-linear way while computation time increases linearly. Consequently, we conclude that there exists an optimal number of trees for any given data set that maximises the stability without unnecessarily increasing the computation time. Based on these findings, we have developed the R package optRF which models the relationship between the number of trees and the stability of random forest, providing recommendations for the optimal number of trees for any given data set.

Phenotypic investigations have shown that actively engaging with music, i.e., playing a musical instrument or singing may be protective of motor decline in aging. For example, music training associated with enhanced sensorimotor skills accompanied by changes in brain structure and function. Although it is possible that the benefits of active music engagement "transfer" to benefits in the motor domain, it is also possible that the genetic architecture of motor behaviour and the motor system structure may influence active music engagement. This study investigated whether polygenic scores (PGS) for five behavioural motor traits, 12 neuromotor structural brain traits, and seven rates of change in brain structure traits trained from existing discovery genome-wide association studies (GWAS) predict active music engagement outcomes in four independent cohorts of unrelated individuals of European ancestry: the Canadian Longitudinal Study on Aging (CLSA; N=22,198), Wisconsin Longitudinal Study (WLS; N=4,605), Vanderbilt's BioVU Repository (BioVU; N=6,150), and Vanderbilt's Online Musicality study (OM; N=1,559). Results were meta-analyzed for each PGS main effect across outcomes and cohorts, revealing that PGS for a faster walking pace was associated with higher amounts of active music engagement. Within CLSA, a higher PGS for walking pace was associated with greater odds of engaging with music. Findings suggest a shared genetic architecture between motor function and active music engagement. Future intervention-based research should consider the genetic underpinnings of motor behavior when evaluating the effects of music engagement on motor function across the lifespan.

Human microglial heterogeneity is only beginning to be appreciated at the molecular level. Here, we present a large, single-cell atlas of expression signatures from 441,088 live microglia broadly sampled across a diverse set of brain regions and neurodegenerative and neuroinflammatory diseases obtained from 161 donors sampled at autopsy or during a neurosurgical procedure. Using single-cell hierarchical Poisson factorization (scHPF), we derived a 23-factor model for continuous gene expression signatures across microglia which capture specific biological processes (e.g., metabolism, phagocytosis, antigen presentation, inflammatory signaling, disease-associated states). Using external datasets, we evaluated the aspects of microglial phenotypes that are encapsulated in various in vitro and in vivo microglia models and identified and replicated the role of two factors in human postmortem tissue of Alzheimer's disease (AD). Further, we derived a complex network of transcriptional regulators for all factors, including regulators of an AD-related factor enriched for the mouse disease-associated microglia 2 (DAM2) signature: ARID5B, CEBPA, MITF, and PPARG. We replicated the role of these four regulators in the AD-related factor and then designed a multiplexed MERFISH panel to assess our microglial factors using spatial transcriptomics. We find that, unlike cells with high expression of the interferon-response factor, cells with high expression of the AD DAM2-like factor are widely distributed in neocortical tissue. We thus propose a novel analytic framework that provides a taxonomic approach for microglia that is more biologically interpretable and use it to uncover new therapeutic targets for AD.

While echocardiography is pivotal for detecting left ventricular hypertrophy (LVH), it struggles with etiology differentiation. To enhance LVH assessment, we aimed to develop an artificial intelligence algorithm using echocardiography-based radiomics. This algorithm is designed to detect LVH and differentiate its common etiologies, such as hypertrophic cardiomyopathy (HCM), cardiac amyloidosis (CA), and hypertensive heart disease (HHD), based on echocardiographic images.

The developmental data sets from multiple medical centers included 867 subjects, with an independent external test set from a single tertiary medical center containing 619 subjects. Radiomic feature analysis was conducted on 4 echocardiographic views, extracting both conventional and harmonization-driven myocardial textures along with myocardial geographic features. Then, we developed classification models for each condition. Variable contributions were evaluated using Shapley Additive Explanations analysis.

The radiomics-based LightGBM model, selected from internal validation, maintained strong performance in the external test set (area under the curve of 0.96 for HCM, 0.89 for CA, and 0.86 for HHD). Compared with the logistic regression model using conventional echocardiographic parameters (left ventricular ejection fraction, left ventricular mass index, left atrial volume index, and E/e'), the final model demonstrated superior sensitivity (0.89 versus 0.80 for HCM, 0.80 versus 0.80 for CA, and 0.75 versus 0.33 for HHD) and F1-score (0.87 versus 0.57 for HCM, 0.84 versus 0.72 for CA, and 0.82 versus 0.50 for HHD). Feature analysis highlighted that harmonization-driven textures played a key role in differentiating HCM, while conventional textures and myocardial thickness were influential in differentiating CA and HHD.

This study confirms that artificial intelligence-enhanced echocardiography-based radiomics effectively differentiate the etiology of LVH, highlighting the potential of artificial intelligence-driven texture and geographic analysis in LVH evaluation.