Elastic Net Model Research Articles

Genomic Profiling of Antimicrobial Resistance Genes in Clinical Salmonella Isolates from Cattle in the Texas Panhandle, USA.

Rising antimicrobial resistance (AMR) in Salmonella serotypes host-adapted to cattle is of increasing concern to the beef and dairy industry. The bulk of the existing literature focuses on AMR post-slaughter. In comparison, the understanding of AMR in Salmonella among pre-harvest cattle is still limited, particularly in Texas, which ranks top five in beef and dairy exports in the United States; inherently, the health of Texas cattle has nationwide implications for the health of the United States beef and dairy industry. In this study, long-read whole genome sequencing and bioinformatic methods were utilized to analyze antimicrobial resistance genes (ARGs) in 98 isolates from beef and dairy cattle in the Texas Panhandle. Fisher exact tests and elastic net models accounting for population structure were used to infer associations between genomic ARG profiles and antimicrobial phenotypic profiles and metadata. Gene mapping was also performed to assess the role of mobile genetic elements in harboring ARGs. Antimicrobial resistance genes were found to be statistically different between the type of cattle operation and Salmonella serotypes. Beef operations were statistically significantly associated with more ARGs compared to dairy operations. Salmonella Heidelberg, followed by Salmonella Dublin isolates, were associated with the most ARGs. Additionally, specific classes of ARGs were only present within mobile genetic elements.

Association of metabolic signatures of air pollution with MASLD: Observational and Mendelian randomization study

Background & AimsTo identify metabolic signatures associated with exposure to ambient air pollution and to explore their associations with risk of metabolic dysfunction-associated steatotic liver disease (MASLD). MethodsWe utilized data from the UK Biobank Cohort. Annual mean concentrations of PM2.5, PM10, NO2 and NOx were assessed for each participant using bilinear interpolation. The Elastic Net regression model was used to identify metabolites associated with four air pollutants and to construct metabolic signatures, respectively. Associations between air pollutants, metabolic signatures and MASLD were analyzed using Cox models. Mendelian randomization (MR) analysis was used to examine potential causality. Mediation analysis was employed to examine the role of metabolic signatures in the association between air pollutants and MASLD. ResultsA total of 244,842 participants from the UK Biobank were included in this analysis. We identified 87, 65, 76, and 71 metabolites as metabolic signatures of PM2.5, PM10, NO2, and NOx, respectively. Metabolic signatures were associated with risk of MASLD, with hazard ratios (HRs) and 95% confidence intervals (95% CIs) were 1.10 (1.06, 1.14), 1.06 (1.02, 1.10), 1.24 (1.20, 1.29) and 1.14 (1.10, 1.19). The four pollutants were associated with increased risk of MASLD, with HRs (95% CIs) of 1.03 (1.01, 1.05), 1.02 (1.01, 1.04), 1.01 (1.01, 1.02) and 1.01 (1.00, 1.01). MR analysis indicated an association between PM2.5, NO2 and NOx-related metabolic signatures and MASLD. Metabolic signatures mediated the association of PM2.5, PM10, NO2 and NOx with MASLD. ConclusionThere may be association between PM2.5, PM10, NO2 and NOx-related metabolic signatures and MASLD, and metabolic signatures mediate the increase of PM2.5, PM10, NO2 and NOx in the risk of MASLD. Impact and implicationsAir pollution is a significant public health issue and an important risk factor for metabolic dysfunction-associated steatotic liver disease (MASLD), however, the mechanism by which air pollution affects MASLD remains unclear. Our study used integrated serological metabolic data of 251 metabolites from a large-scale cohort study to demonstrate that metabolic signatures play a crucial role in the elevated risk of MASLD caused by air pollution. These results are relevant to patients and policymakers because they suggest that air pollution-related metabolic signatures are not only potentially associated with MASLD but also involved in mediating the process by which PM2.5, PM10, NO2, and NOx increase the risk of MASLD. Focusing on changes in air pollution-related metabolic signatures may offer a new perspective for preventing air pollution-induced MASLD and serve as protective measures to address this emerging public health challenge.

392. MACHINE LEARNING TO PREDICT LYMPH NODE METASTASIS IN T1 ESOPHAGEAL SQUAMOUS CELL CARCINOMA: A MULTICENTER STUDY

Abstract Background Lymph node metastases (LNM) is a relatively uncommon but possible complication of T1 esophageal squamous cell carcinoma (ESCC). Existing models do poorly when it comes to quantifying this risk. As a branch of artificial intelligence, machine learning (ML) algorithms have developed and become more prominent. Up to now, no individualized ML prediction model aimed at predicting LNM in T1 esophageal squamous cell carcinoma patients is available. This study aimed to develop a machine learning model for LNM in patients with T1 esophageal squamous cell carcinoma. Methods Patients with T1 squamous cell carcinoma treated with surgery between January 2010 and September 2021 from 3 institutions were included. Machine-learning models were developed using data on patients’ age and sex, depth of tumor invasion, tumor size, tumor location, macroscopic tumor type, lymphatic and vascular invasion, and histologic grade. Elastic net regression (ELR), random forest (RF), extreme gradient boosting (XGB), and a combined (ensemble) model of these was generated. Use Area Under Curve (AUC) to evaluate the predictive ability of models. The contribution of each factor was calculated. To meet clinical needs, we designed the model as a user-friendly website. Results 926 and 341 patients were included in the derivation and validation cohorts. The lymph node metastasis rate was 14.8%. The models all exhibited potent discriminating power. The elastic net regression model performed best with externally validated AUC of 0.803 (AUC 0.773 for RF, 0.791 for XGB, 0.799 for ensemble) whereas the NCCN guidelines identified patients with LNM with an AUC of 0.576 and logistic model with an AUC of 0. 670. The most important features were lymphatic and vascular invasion as well as depth of tumor invasion. Conclusions The model created utilizing machine learning approaches had excellent performance estimating the likelihood of LNM in T1 ESCC, which can help patients achieve personalized treatment options.

Quantifying seasonal variations in pollution sources with machine learning-enhanced positive matrix factorization

As the pace of industrialization and urbanization accelerates, water quality management faces increasing challenges, with traditional methods for pollutant source apportionment often proving inadequate in handling complex environmental data. This study enhances the precision and reliability of pollutant source identification by integrating Positive Matrix Factorization (PMF) models with diverse machine learning techniques. Utilizing data from 17 water quality monitoring stations along the Wuding River from 2017 to 2021, we employed Random Forest (RF), Support Vector Machine (SVM), Elastic Net (EN), and Extreme Gradient Boosting (XGBoost) models to predict the Water Quality Index (WQI) during dry and wet seasons. Results indicate that the RF model exhibited optimal performance in the dry season (R2 = 0.93), while the SVM was superior in the wet season (R2 = 0.94). SHAP (SHapley Additive exPlanations) value analysis identified CODMn and NH3-N as significant influencers on WQI in the dry season, whereas COD, BOD, and TP gained prominence during the wet season. SHAP values reveal the contribution of each feature to the model output, thereby enhancing the model’s transparency and interpretability. Additionally, feature importance identified by machine learning was utilized as weights to optimize the contribution rates predicted by the PMF model. The optimised model was able to identify the contribution of domestic and farm effluent discharges more accurately in the dry season, with a significant increase in the percentage of identification from 19.4 % to 45.4 %, and an increase in the percentage of contribution from agricultural non-point sources and domestic effluent in the rainy season. This research offers a novel perspective on the characteristics of river water pollution and holds significant implications for formulating data-driven environmental management strategies.

Ancestry-aligned polygenic scores combined with conventional risk factors improve prediction of cardiometabolic outcomes in African populations

BackgroundCardiovascular diseases (CVD) are a major health concern in Africa. Improved identification and treatment of high-risk individuals can reduce adverse health outcomes. Current CVD risk calculators are largely unvalidated in African populations and overlook genetic factors. Polygenic scores (PGS) can enhance risk prediction by measuring genetic susceptibility to CVD, but their effectiveness in genetically diverse populations is limited by a European-ancestry bias. To address this, we developed models integrating genetic data and conventional risk factors to assess the risk of developing cardiometabolic outcomes in African populations.MethodsWe used summary statistics from a genome-wide association meta-analysis (n = 14,126) in African populations to derive novel genome-wide PGS for 14 cardiometabolic traits in an independent African target sample (Africa Wits-INDEPTH Partnership for Genomic Research (AWI-Gen), n = 10,603). Regression analyses assessed relationships between each PGS and corresponding cardiometabolic trait, and seven CVD outcomes (CVD, heart attack, stroke, diabetes mellitus, dyslipidaemia, hypertension, and obesity). The predictive utility of the genetic data was evaluated using elastic net models containing multiple PGS (MultiPGS) and reference-projected principal components of ancestry (PPCs). An integrated risk prediction model incorporating genetic and conventional risk factors was developed. Nested cross-validation was used when deriving elastic net models to enhance generalisability.ResultsOur African-specific PGS displayed significant but variable within- and cross- trait prediction (max.R2 = 6.8%, p = 1.86 × 10−173). Significantly associated PGS with dyslipidaemia included the PGS for total cholesterol (logOR = 0.210, SE = 0.022, p = 2.18 × 10−21) and low-density lipoprotein (logOR = − 0.141, SE = 0.022, p = 1.30 × 10−20); with hypertension, the systolic blood pressure PGS (logOR = 0.150, SE = 0.045, p = 8.34 × 10−4); and multiple PGS associated with obesity: body mass index (max. logOR = 0.131, SE = 0.031, p = 2.22 × 10−5), hip circumference (logOR = 0.122, SE = 0.029, p = 2.28 × 10−5), waist circumference (logOR = 0.013, SE = 0.098, p = 8.13 × 10−4) and weight (logOR = 0.103, SE = 0.029, p = 4.89 × 10−5). Elastic net models incorporating MultiPGS and PPCs significantly improved prediction over MultiPGS alone. Models including genetic data and conventional risk factors were more predictive than conventional risk models alone (dyslipidaemia: R2 increase = 2.6%, p = 4.45 × 10−12; hypertension: R2 increase = 2.6%, p = 2.37 × 10−13; obesity: R2 increase = 5.5%, 1.33 × 10−34).ConclusionsIn African populations, CVD and associated cardiometabolic trait prediction models can be improved by incorporating ancestry-aligned PGS and accounting for ancestry. Combining PGS with conventional risk factors further enhances prediction over traditional models based on conventional factors. Incorporating data from target populations can improve the generalisability of international predictive models for CVD and associated traits in African populations.

Improving genetic risk modeling of dementia from real-world data in underrepresented populations

Genetic risk modeling for dementia offers significant benefits, but studies based on real-world data, particularly for underrepresented populations, are limited. We employ an Elastic Net model for dementia risk prediction using single-nucleotide polymorphisms prioritized by functional genomic data from multiple neurodegenerative disease genome-wide association studies. We compare this model with APOE and polygenic risk score models across genetic ancestry groups (Hispanic Latino American sample: 610 patients with 126 cases; African American sample: 440 patients with 84 cases; East Asian American sample: 673 patients with 75 cases), using electronic health records from UCLA Health for discovery and the All of Us cohort for validation. Our model significantly outperforms other models across multiple ancestries, improving the area-under-precision-recall curve by 31–84% (Wilcoxon signed-rank test p-value <0.05) and the area-under-the-receiver-operating characteristic by 11–17% (DeLong test p-value <0.05) compared to the APOE and the polygenic risk score models. We identify shared and ancestry-specific risk genes and biological pathways, reinforcing and adding to existing knowledge. Our study highlights the benefits of integrating functional mapping, multiple neurodegenerative diseases, and machine learning for genetic risk models in diverse populations. Our findings hold potential for refining precision medicine strategies in dementia diagnosis.

Explainable machine learning framework to predict the risk of work-related neck and shoulder musculoskeletal disorders among healthcare professionals.

This study aims to develop risk prediction models for neck and shoulder musculoskeletal disorders among healthcare professionals. A stratified sampling method was employed to select employees from medical institutions in Nanning City, yielding 617 samples. The Boruta algorithm was used for feature selection, and various models, including Tree-Based Models, Single Hidden-Layer Neural Network Models (MLP), Elastic Net Models (ENet), and Support Vector Machines (SVM), were applied to predict the selected variables, utilizing SHAP algorithms for individual-level local explanations. The SVM model excels in both Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) and exhibits more stable performance when generalizing to unseen data. The Random Forest model exhibited relatively high overall performance on the training set. The MLP model emerges as the most consistent and accurate in predicting shoulder musculoskeletal disorders, while the SVM model shows strong fitting capabilities during the training phase, with occupational factors identified as the main contributors to WMSDs. This study successfully constructs work-related musculoskeletal disorder risk prediction models for healthcare professionals, enabling a quantitative analysis of the impact of occupational factors. This advancement is beneficial for future economical and convenient work-related musculoskeletal disorder screening in healthcare professions.

An epigenome-wide study of selenium status and DNA methylation in the Strong Heart Study

BackgroundSelenium (Se) is an essential nutrient linked to adverse health endpoints at low and high levels. The mechanisms behind these relationships remain unclear and there is a need to further understand the epigenetic impacts of Se and their relationship to disease. We investigated the association between urinary Se levels and DNA methylation (DNAm) in the Strong Heart Study (SHS), a prospective study of cardiovascular disease (CVD) among American Indians adults. MethodsSelenium concentrations were measured in urine (collected in 1989–1991) using inductively coupled plasma mass spectrometry among 1,357 participants free of CVD and diabetes. DNAm in whole blood was measured cross-sectionally using the Illumina MethylationEPIC BeadChip (850 K) Array. We used epigenome-wide robust linear regressions and elastic net to identify differentially methylated cytosine-guanine dinucleotide (CpG) sites associated with urinary Se levels. ResultsThe mean (standard deviation) urinary Se concentration was 51.8 (25.1) μg/g creatinine. Across 788,368 CpG sites, five differentially methylated positions (DMP) (hypermethylated: cg00163554, cg18212762, cg11270656, and hypomethylated: cg25194720, cg00886293) were significantly associated with Se in linear regressions after accounting for multiple comparisons (false discovery rate p-value: 0.10). The top hypermethylated DMP (cg00163554) was annotated to the Disco Interacting Protein 2 Homolog C (DIP2C) gene, which relates to transcription factor binding. Elastic net models selected 425 hypo- and hyper-methylated DMPs associated with urinary Se, including three sites (cg00163554 [DIP2C], cg18212762 [MAP4K2], cg11270656 [GPIHBP1]) identified in linear regressions. ConclusionsUrinary Se was associated with minimal changes in DNAm in adults from American Indian communities across the Southwest and the Great Plains in the United States, suggesting that other mechanisms may be driving health impacts. Future analyses should explore other mechanistic biomarkers in human populations, determine these relationships prospectively, and investigate the potential role of differentially methylated sites with disease endpoints.

Is more care recipient time at home also a family caregiver-centered quality of life measure?

Time in healthcare facilities is associated with worse patient quality of life (QoL); however, impact on family caregiver QoL is unknown. We evaluate care recipient days not at home-days in the emergency department (ED), inpatient (IP) care, and post-acute care (PAC)-to understand how care recipient days not at home correspond to family caregiver QoL. Secondary data were linked to care recipient utilization data. Elastic net machine learning models were used to evaluate the impact of a single day of utilization in each setting on binary QoL outcomes. We also compared composite weighted and unweighted "days not at home" variables. Two time periods, 6 and 18 months, were used to predict three caregiver QoL measures (self-rated health, depressive symptoms, and subjective burden). In the 6-month timeframe, a single day of ED utilization was associated with increased likelihood of poor QoL for all three assessed outcomes (range: 1.4%-3.2%). A day of PAC was associated to a modest degree with increased likelihood of caregiver burden (0.2%) and depressive symptoms (0.1%), with a slight protective effect for self-rated health (-0.1%). An IP day had a slight protective effect (-0.2 to -0.1%). At 18 months, ED and IP had similar, albeit more muted, relationships with caregiver burden and depressive symptoms. PAC had a slight protective effect for caregiver burden (-0.1%). Cumulative days in all settings combined generally was not associated with caregiver QoL. Whereas total care recipient time away from home had some negative spillovers to family caregivers, the countervailing effects of unique settings on caregiver QoL may mask net QoL effects. This finding limits the utility of a single care recipient home time measure as a valid caregiver-centered measure. Considering cumulative care recipient time in individual settings separately may be needed to reveal the true net effects on caregiver QoL.

Comparison of Machine Learning Models for Predicting Interstitial Glucose Using Smart Watch and Food Log

This study examines the performance of various machine learning (ML) models in predicting Interstitial Glucose (IG) levels using data from wrist-worn wearable sensors. The insights from these predictions can aid in understanding metabolic syndromes and disease states. A public dataset comprising information from the Empatica E4 smart watch, the Dexcom Continuous Glucose Monitor (CGM) measuring IG, and a food log was utilized. The raw data were processed into features, which were then used to train different ML models. This study evaluates the performance of decision tree (DT), support vector machine (SVM), Random Forest (RF), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Gaussian Naïve Bayes (GNB), lasso cross-validation (LassoCV), Ridge, Elastic Net, and XGBoost models. For classification, IG labels were categorized into high, standard, and low, and the performance of the ML models was assessed using accuracy (40–78%), precision (41–78%), recall (39–77%), F1-score (0.31–0.77), and receiver operating characteristic (ROC) curves. Regression models predicting IG values were evaluated based on R-squared values (−7.84–0.84), mean absolute error (5.54–60.84 mg/dL), root mean square error (9.04–68.07 mg/dL), and visual methods like residual and QQ plots. To assess whether the differences between models were statistically significant, the Friedman test was carried out and was interpreted using the Nemenyi post hoc test. Tree-based models, particularly RF and DT, demonstrated superior accuracy for classification tasks in comparison to other models. For regression, the RF model achieved the lowest RMSE of 9.04 mg/dL with an R-squared value of 0.84, while the GNB model performed the worst, with an RMSE of 68.07 mg/dL. A SHAP analysis identified time from midnight as the most significant predictor. Partial dependence plots revealed complex feature interactions in the RF model, contrasting with the simpler interactions captured by LDA.

A personalised algorithm predicting the risk of intravenous corticosteroid failure in acute ulcerative colitis.

An episode of acute ulcerative colitis (UC) represents an important watershed moment in a patient's disease course. To derive a personalised algorithm for identifying patients at high risk of corticosteroid non-response from variables available at hospital presentation using a large prospectively collected acute UC patient database and machine learning-based techniques. We analysed data from 682 consecutive presentations of acute UC. We used an Akaike information criterion-based elastic net model to select variables based on the 419 earliest presentations of acute UC (1996-2017). We constructed two risk-scoring algorithms, with and without utilising additional endoscopic variables, using logistic regression models. We validated these risk scores on separate cohorts of 181 (2018-2022) and 82 (2015-2022) acute UC presentations. The partial risk of rescue (ROR) score included the admission indices of oral corticosteroid treatment, bowel frequency ≥6/24 h, albumin, CRP ≥12 mg/mL and log10CRP. The full ROR score incorporates the same variables with the addition of the Mayo endoscopic subscore and disease extent. The AUCs in the main validation cohort were 0.76 (95% CI: 0.69-0.83) and 0.78 (95% CI: 0.71-0.85) for the partial and full ROR scores, respectively. These pragmatic personalised risk scores (available at www.severecolitis.com) have comparably strong performance characteristics and usability enabling the identification of individuals at high risk of corticosteroid non-response before or after endoscopic assessment. The ROR scores have the potential to challenge conventional acute UC treatment paradigms by identifying patients who may benefit from early rescue therapy or participation in relevant clinical trials.

Visual analytics of an interpretable prediction model for the glass transition temperature of fluoroelastomers

In development, precise measurement and control of Tg is essential for researchers to thoroughly understand and master how variations in formulation and production processes impact product performance. Given the complexity of industrial production, most manufacturers repeatedly produce fluorine materials and test their Tg until it satisfies the application environment’s demands. Consequently, this study utilized an FKM dataset from actual synthesis trials, subjected to data classification, alignment, and feature filtering, to develop a predictive model based on Elastic Net Regression, enhancing the accuracy of Tg predictions for FKM. We evaluated the performance of several regression machine learning algorithms, including Random Forest, Support Vector Machines, Decision Trees, Extreme Gradient Boosting, and Ensemble Learning. The findings show that Random Forest and Elastic Net models exceeded other models by approximately 5% to 10 % in the coefficient of determination, indicating superior predictive capabilities. The Elastic Net Regression model excelled particularly in terms of evaluation coefficients and error rates. To boost the model’s interpretability, this research developed and implemented a visualization technology scheme for the excellently performing models, integrating SHAP value analysis to vividly depict the feature space, prediction processes, and experimental outcomes, thereby aiding the exploration of Tg influencing factors. This research not only offers an innovative machine learning framework for Tg prediction in FKM but also enhances the model’s transparency and credibility using visualization technology, holding significant theoretical and practical implications for materials science research.

Serum protein responses to Dietary Approaches to Stop Hypertension (DASH) and DASH-Sodium trials and associations with blood pressure changes.

The Dietary Approaches to Stop Hypertension (DASH) diet reduces blood pressure, but the mechanisms underlying DASH diet-blood pressure relations are not well understood. Proteomic measures may provide insights into the pathophysiological mechanisms through which the DASH diet reduces blood pressure. The DASH (1994-1996) and DASH-Sodium (1997-1999) trials were multicenter, randomized-controlled feeding trials. Proteomic profiling was conducted in serum collected at the end of the feeding period (DASH, N = 215; DASH-Sodium, N = 390). Multivariable linear regression models were used to identify interactions between 71 DASH diet-related proteins and changes in systolic and diastolic blood pressure. Estimates were meta-analyzed across both trials. Elastic net models were used to identify proteins that predict changes in blood pressure. Ten significant interactions were identified [systolic blood pressure: seven proteins; diastolic blood pressure: three proteins], which represented nine unique proteins. A high level of renin at the end of the feeding period was associated with greater reductions in diastolic blood pressure in individuals consuming the control than DASH diets. A high level of procollagen c-endopeptidase enhancer 1 (PCOLCE) and collagen triple helix repeat-containing protein 1 (CTHRC1) were associated with greater reductions in systolic blood pressure in individuals consuming the DASH than control diets, and with elevations in systolic blood pressure in individuals consuming the control diets (P for interaction for all tests < 0.05). Elastic net models identified six additional proteins that predicted change in blood pressure. Several novel proteins were identified that may provide some insight into the relationship between the DASH diet and blood pressure.

A Novel Machine-Learning Algorithm to Predict the Early Termination of Nutrition Support Team Follow-Up in Hospitalized Adults: A Retrospective Cohort Study.

For hospitalized adults, it is important to initiate the early reintroduction of oral food in accordance with nutrition support team guidelines. The aim of this study was to develop and validate a machine learning-based algorithm that predicts the early termination of medical nutritional therapy (the transition to oral feeding). This retrospective cohort study included consecutive adult patients admitted to the Hacettepe hospital (from 1 January 2018 to 31 December 2022). The outcome of the study was the prediction of an early transition to adequate oral feeding before discharge. The dataset was randomly (70/30) divided into training and test datasets. We used six ML algorithms with multiple features to construct prediction models. ML model performance was measured according to the accuracy, area under the receiver operating characteristic curve, and F1 score. We used the Boruta Method to determine the important features and interpret the selected features. A total of 2298 adult inpatients who were followed by a nutrition support team for medical nutritional therapy were included. Patients received parenteral nutrition (1471/2298, 64.01%), enteral nutrition (717/2298, 31.2%), or supplemental parenteral nutrition (110/2298, 4.79%). The median (interquartile range) Nutritional Risk Screening (NRS-2002) score was 5 (1). Six prediction algorithms were used, and the artificial neural network and elastic net models achieved the greatest area under the ROC in all outcomes (AUC = 0.770). Ranked by z-value, the 10 most important features in predicting an early transition to oral feeding in the artificial neural network and elastic net algorithms were parenteral nutrition, surgical wards, surgical outcomes, enteral nutrition, age, supplemental parenteral nutrition, digestive system diseases, gastrointestinal complications, NRS-2002, and impaired consciousness. We developed machine learning models for the prediction of an early transition to oral feeding before discharge. Overall, there was no discernible superiority among the models. Nevertheless, the artificial neural network and elastic net methods provided the highest AUC values. Since the machine learning model is interpretable, it can enable clinicians to better comprehend the features underlying the outcomes. Our study could support personalized treatment and nutritional follow-up strategies in clinical decision making for the prediction of an early transition to oral feeding in hospitalized adult patients.

Associations of movement behaviours and dietary intake with arterial stiffness: results from the ORISCAV-LUX 2 cross-sectional study

ObjectiveAdopting a physically active lifestyle and maintaining a diet rich in antioxidants can reduce the risk of vascular diseases. Arterial stiffness is an early marker for cardiovascular diseases, indicating vascular...

The Predictive Performance of General Surgery Milestones on Post-Graduation Outcomes.

To establish whether Accreditation Council for Graduate Medical Education Milestones predict future performance of general surgery trainees. Milestones provide bi-annual assessments of trainee progress across six competencies. It is unknown whether the Milestones predict surgeon performance after the transition to independent practice. We performed a retrospective cohort study of surgeons with complete Milestone assessments in the fourth and fifth clinical years who treated patients in acute care hospitals within Florida, New York, and Pennsylvania, 2015-2018. To account for the multiple ways in which the Milestone assessments might predict post-graduation outcomes, we included 120 Milestones features in our elastic net machine learning models. The primary outcome was risk-adjusted patient death or serious morbidity. 278 general surgeons were included in the study. Milestone assessments 6-months into the fourth clinical year displayed a normal score distribution while multicollinearity and low score discrimination at the final assessment period were detected. Individual Milestones features from the Patient Care, Professionalism, and Systems-based Practice domains were most predictive of patient-related outcomes. For example, surgeons with worse patient outcomes had significantly lower scores in Patient Care 3 when compared to surgeons with better patient outcomes (High DSM, yes: 2.86 vs. no: 3.04, P=0.011). The Milestones features that were most predictive of better patient outcomes related to intraoperative skills, ethical principles, and patient navigation and safety, measured 12-18 months prior to graduation. The development of a parsimonious set of evidence-based Milestones that better correlate with surgeon experience could enhance surgical education.

Individual Deviation-Based Functional Hypergraph for Identifying Subtypes of Autism Spectrum Disorder.

Heterogeneity has been one of the main barriers to understanding and treatment of autism spectrum disorder (ASD). Previous studies have identified several subtypes of ASD through unsupervised clustering analysis. However, most of them primarily depicted the pairwise similarity between individuals through second-order relationships, relying solely on patient data for their calculation. This leads to an underestimation of the complexity inherent in inter-individual relationships and the diagnostic information provided by typical development (TD). To address this, we utilized an elastic net model to construct an individual deviation-based hypergraph (ID-Hypergraph) based on functional MRI data. We then conducted a novel community detection clustering algorithm to the ID-Hypergraph, with the aim of identifying subtypes of ASD. By applying this framework to the Autism Brain Imaging Data Exchange repository data (discovery: 147/125, ASD/TD; replication: 134/132, ASD/TD), we identified four reproducible ASD subtypes with roughly similar patterns of ALFF between the discovery and replication datasets. Moreover, these subtypes significantly varied in communication domains. In addition, we achieved over 80% accuracy for the classification between these subtypes. Taken together, our study demonstrated the effectiveness of identifying subtypes of ASD through the ID-hypergraph, highlighting its potential in elucidating the heterogeneity of ASD and diagnosing ASD subtypes.

A Penalized Regression Method for Genomic Prediction Reduces Mismatch between Training and Testing Sets.

Genomic selection (GS) is changing plant breeding by significantly reducing the resources needed for phenotyping. However, its accuracy can be compromised by mismatches between training and testing sets, which impact efficiency when the predictive model does not adequately reflect the genetic and environmental conditions of the target population. To address this challenge, this study introduces a straightforward method using binary-Lasso regression to estimate β coefficients. In this approach, the response variable assigns 1 to testing set inputs and 0 to training set inputs. Subsequently, Lasso, Ridge, and Elastic Net regression models use the inverse of these β coefficients (in absolute values) as weights during training (WLasso, WRidge, and WElastic Net). This weighting method gives less importance to features that discriminate more between training and testing sets. The effectiveness of this method is evaluated across six datasets, demonstrating consistent improvements in terms of the normalized root mean square error. Importantly, the model's implementation is facilitated using the glmnet library, which supports straightforward integration for weighting β coefficients.

DNA Damage Response in Early Breast Cancer: A Phase III Cohort in the Phobos Study.

We assessed the impact of DNA damage response and repair (DDR) biomarker expressions in 222 node-positive early breast cancer (BC) patients from a previous Phase III GOIM 9902 trial of adjuvant taxanes. At a median follow-up of 64 months, the original study showed no disease-free survival (DFS) or overall survival (OS) differences with the addition of docetaxel (D) to epirubicine-cyclophosphamide (EC). Immunohistochemistry was employed to assess the expression of DDR phosphoproteins (pATM, pATR, pCHK1, γH2AX, pRPA32, and pWEE1) in tumor tissue, and their association with clinical outcomes was evaluated through the Cox elastic net model. Over an extended follow-up of 234 months, we confirmed no significant differences in DFS or OS between patients treated with EC and those receiving D → EC. A DDR risk score, inversely driven by ATM and ATR expression, emerged as an independent prognostic factor for both DFS (HR = 0.41, p < 0.0001) and OS (HR = 0.61, p = 0.046). Further validation in a public adjuvant BC cohort was possible only for ATM, confirming its protective role. Overall, our findings confirm the potential role of the DDR pathway in BC prognostication and in shaping treatment strategies advocating for an integrated approach, combining molecular markers with clinical-pathological factors.

Large-Scale Proteomics in Early Pregnancy and Hypertensive Disorders of Pregnancy

There is no consensus regarding the best method for prediction of hypertensive disorders of pregnancy (HDP), including gestational hypertension and preeclampsia. To determine predictive ability in early pregnancy of large-scale proteomics for prediction of HDP. This was a nested case-control study, conducted in 2022 to 2023, using clinical data and plasma samples collected between 2010 and 2013 during the first trimester, with follow-up until pregnancy outcome. This multicenter observational study took place at 8 academic medical centers in the US. Nulliparous individuals during first-trimester clinical visits were included. Participants with HDP were selected as cases; controls were selected from those who delivered at or after 37 weeks without any HDP, preterm birth, or small-for-gestational-age infant. Age, self-reported race and ethnicity, body mass index, diabetes, health insurance, and fetal sex were available covariates. Proteomics using an aptamer-based assay that included 6481 unique human proteins was performed on stored plasma. Covariates were used in predictive models. Prediction models were developed using the elastic net, and analyses were performed on a randomly partitioned training dataset comprising 80% of study participants, with the remaining 20% used as an independent testing dataset. Primary measure of predictive performance was area under the receiver operating characteristic curve (AUC). This study included 753 HDP cases and 1097 controls with a mean (SD) age of 26.9 (5.5) years. Maternal race and ethnicity were 51 Asian (2.8%), 275 non-Hispanic Black (14.9%), 275 Hispanic (14.9%), 1161 non-Hispanic White (62.8% ), and 88 recorded as other (4.8%), which included those who did not identify according to these designations. The elastic net model, allowing for forced inclusion of prespecified covariates, was used to adjust protein-based models for clinical and demographic variables. Under this approach, no proteins were selected to augment the clinical and demographic covariates. The predictive performance of the resulting model was modest, with a training set AUC of 0.64 (95% CI, 0.61-0.67) and a test set AUC of 0.62 (95% CI, 0.56-0.68). Further adjustment for study site yielded only minimal changes in AUCs. In this case-control study with detailed clinical data and stored plasma samples available in the first trimester, an aptamer-based proteomics panel did not meaningfully add to predictive utility over and above clinical and demographic factors that are routinely available.