Performance Of Prediction Models Research Articles

Mobile single-lead ECG atrial fibrillation prediction enhancement integrated by standard ECG algorithm with Deep learning model

Abstract Background Artificial intelligence (AI) using electrocardiogram (ECG) enabled to predict atrial fibrillation (AF) in patients without documented AF. Mobile single-lead ECG is more convenient to surveil cardiac rhythm with simple measurement. However, AI-enabled arrhythmia predictability by mobile ECG is limited due to single channel utilization and longer duration for arrhythmia diagnosis. We aimed to enhance mobile single-lead ECG AF prediction AI algorithm integrated with 12-lead ECG using deep learning model. Method Based on 552,372 12-lead ECG data of 318,321 patients, a statistical AF prediction model employing a deep-learning approach was constituted. Out of single-lead 13,509 ECGs from a total of 6,719 patients, we utilized a total of 10,287 normal sinus rhythm ECGs from 5,170 patients. Resnet structure was utilized to distinguish subtle changes of the vicinity of P-wave. Single-lead mobile ECGs were adjusted with noise filtering and segmented every 10 seconds. A random under-sampling was applied to reduce bias from data imbalance. Both 12-lead ECG and single-lead ECG were allocated to training, validation, testing datasets in a 6:2:2 ratio. Then, we conducted transfer learning using the standard 12-lead ECG’s deep learning model to improve performance of single-lead mobile ECG deep learning model. Results AF was annotated in 26,541 (4.8%) with 12-lead ECG whereas 1,443 (21.2%) with single-lead mobile ECG. The area under the curve (AUC) value for predicting AF was 0.910 with 12-lead ECG, and 0.742 with mobile ECG. The predictive performance of mobile ECG was 67.5% in accuracy, 64.2% in sensitivity and 66.8% in F1-score. The AUC value of mobile ECG after applying transfer learning based on 12-lead ECG for AF prediction was increased to 0.790 with accuracy of 73.8%, sensitivity of 65.5% and F1-score of 71.0%. Conclusion Integration with deep learning algorithm of standard 12-lead ECG significantly improved the model performance of single-lead ECG AF prediction model compared to single-lead mobile ECG only based model. Easy application of mobile ECG with enhanced AF predictability might serve a more convenient method as a pre-emptive assistive tool to provide probabilistic prediction for PAF screening rather than 12-lead ECG.

A radiomics-based interpretable machine learning model to predict the HER2 status in bladder cancer: a multicenter study

ObjectiveTo develop a computed tomography (CT) radiomics-based interpretable machine learning (ML) model to preoperatively predict human epidermal growth factor receptor 2 (HER2) status in bladder cancer (BCa) with multicenter validation.MethodsIn this retrospective study, 207 patients with pathologically confirmed BCa were enrolled and divided into the training set (n = 154) and test set (n = 53). Least absolute shrinkage and selection operator (LASSO) regression was used to identify the most discriminative features in the training set. Five radiomics-based ML models, namely logistic regression (LR), support vector machine (SVM), k-nearest neighbors (KNN), eXtreme Gradient Boosting (XGBoost) and random forest (RF), were developed. The predictive performance of established ML models was evaluated by the area under the receiver operating characteristic curve (AUC). The Shapley additive explanation (SHAP) was used to analyze the interpretability of ML models.ResultsA total of 1218 radiomics features were extracted from the nephrographic phase CT images, and 11 features were filtered for constructing ML models. In the test set, the AUCs of LR, SVM, KNN, XGBoost, and RF were 0.803, 0.709, 0.679, 0.794, and 0.815, with corresponding accuracies of 71.7%, 69.8%, 60.4%, 75.5%, and 75.5%, respectively. RF was identified as the optimal classifier. SHAP analysis showed that texture features (gray level size zone matrix and gray level co-occurrence matrix) were significant predictors of HER2 status.ConclusionsThe radiomics-based interpretable ML model provides a noninvasive tool to predict the HER2 status of BCa with satisfactory discriminatory performance.Critical relevance statementAn interpretable radiomics-based machine learning model can preoperatively predict HER2 status in bladder cancer, potentially aiding in the clinical decision-making process.Key PointsThe CT radiomics model could identify HER2 status in bladder cancer.The random forest model showed a more robust and accurate performance.The model demonstrated favorable interpretability through SHAP method.Graphical

Improving the prediction of major adverse cardiovascular events in patients with hypertrophic cardiomyopathy by adding plasma SVEP1 to conventional clinical model including NT-proBNP

Abstract Background Hypertrophic cardiomyopathy (HCM) is the most common genetic heart disease with a prevalence between 1 in 200 to 500 [1]. HCM can lead to major adverse cardiovascular events (MACE) such as heart failure (HF) hospitalization and cardiac death [2,3]. A protein named "sushi, von Willebrand factor type A, EGF and pentraxin domain containing 1" (SVEP1) is a large extracellular matrix protein whose activity can be measured in plasma [4,5]. SVEP1 promotes inflammation and atherosclerosis [4,5]. SVEP1 is associated with risk of HF hospitalization and cardiac death in patients with HF with reduced ejection fraction [6]. However, no prior study has assessed the prognostic value of SVEP1 in patients with HCM. Purpose To test the hypothesis that the addition of SVEP1 improves the predictive performance of the conventional model based on clinical risk factors plus NT-proBNP for future MACE in patients with HCM. Methods We performed a multicenter prospective cohort study of patients with HCM [7]. The outcome was MACE, defined as a composite of HF hospitalization or cardiac death [6]. After dividing the cohort into four groups using the median values of SVEP1 and NT-proBNP, we compared the risk of the outcome event using the Cox proportional hazards model adjusting for 15 clinical risk factors known to be associated with MACE in HCM (Figure 1) [8-11]. We also developed the Lasso-regularized Cox proportional hazards model to predict the time to first MACE by adding SVEP1 to the 15 clinical risk factors with or without NT-proBNP in the training cohort (randomly selected 75% of the cohort). We then compared the predictive performance in the test cohort (25% of the cohort) using the C-statistic of each model. Results During a median follow-up of 1.9 (25 percentile-75 percentile, 0.7-4.3) years, the outcome event occurred in 63 (10%) of 610 patients. The four groups stratified by the median of SVEP1 and NT-proBNP levels had different risks of the outcome event (log-rank p<0.001: Figure 1). Even in the groups with lower-than-median NT-proBNP levels, the high-SVEP1 group had higher risks of MACE compared with the low-SVEP1 group (adjusted hazard ratio 4.52, p=0.042: Figure 1). In the prediction of time to first MACE, the addition of SVEP1 improved the C-statistic of the clinical model (p<0.001) and that of the clinical plus NT-proBNP model (p=0.01: Figure 2). The clinical plus SVEP1 model also outperformed the clinical plus NT-proBNP model (p=0.005: Figure 2). Conclusions In the present multicenter prospective cohort study of 610 patients with HCM, SVEP1 improved MACE risk stratification when added to the conventional model consisting of clinical risk factors plus NT-proBNP. SVEP1 also exhibited a superior predictive ability than NT-proBNP to predict MACE in patients with HCM. These data collectively indicate that SVEP1 provides additional risk stratification and may be a better biomarker than NT-proBNP for the prognostication of patients with HCM.Figure 1.Kaplan-Meier curves for MACEFigure 2.Time-dependent AUROC curves

Visual field prediction based on temporal-spatial feature learning

Glaucoma stands as the leading irreversible cause of blindness worldwide. Regular visual field examinations play a crucial role in both diagnosing and treating glaucoma. Predicting future visual field changes can assist clinicians in making timely interventions to manage the progression of this disease. To integrate temporal and spatial features from past visual field examination results and enhance visual field prediction, a convolutional long short-term memory (ConvLSTM) network was employed to construct a predictive model. The predictive performance of the ConvLSTM model was validated and compared with other methods using a dataset of perimetry tests from the Humphrey field analyzer at the University of Washington (UWHVF). Compared to traditional methods, the ConvLSTM model demonstrated higher prediction accuracy. Additionally, the relationship between visual field series length and prediction performance was investigated. In predicting the visual field using the previous three visual field results of past 1.5~6.0 years, it was found that the ConvLSTM model performed better, achieving a mean absolute error of 2.255 dB, a root mean squared error of 3.457 dB, and a coefficient of determination of 0.960. The experimental results show that the proposed method effectively utilizes existing visual field examination results to achieve more accurate visual field prediction for the next 0.5~2.0 years. This approach holds promise in assisting clinicians in diagnosing and treating visual field progression in glaucoma patients.

Ensemble approach for predicting the diagnosis of osteoarthritis using physical activity factors.

Osteoarthritis (OA) is a common degenerative disease of the joints. Risk factors for OA include non-modifiable factors such as age and sex, as well as modifiable factors like physical activity. this study aimed to construct a soft voting ensemble model to predict OA diagnosis using variables related to individual characteristics and physical activity and identify important variables in constructing the model through permutation importance. By using the recursive feature elimination, cross-validated technique, the variables with the best predictive performance were selected among variables, and an ensemble model combining RandomForest, XGBoost, and LightGBM algorithms was constructed. The predictive performance and permutation importance of each variable were evaluated. The variables selected to construct the model were age, sex, grip strength, and quality of life, and the accuracy of the ensemble model was 0.828. The most important variable in constructing the model was age (0.199), followed by grip strength (0.053), quality of life (0.043), and sex (0.034). The performance of the model for predicting OA was relatively good. If this model is continuously used and updated, it could be used to predict OA diagnosis, and the predictive performance of the OA model may be further improved.

Predicting the fatigue life of additively manufactured AlSi10Mg alloy using a physics-informed neural network incorporating continuous damage mechanics

The material characteristics of additively manufactured AlSi10Mg alloy, including the random distribution of process-induced micro defects, microstructual anisotropy, and grain morphologies with complex diversity, poses a significant challenge in accurately predicting its fatigue life, limiting its application in the aircraft field. This work herein introduces a novel approach incorporating a physics-informed neural network (PINN), in which an artificial neural network (ANN) is embedded with a damage mechanics model (CDM) and the critical process parameters of laser powder bed fusion (L-PBF) additive manufacturing. The partial differential equations of the updated CDM model are introduced into the training procedures of the PINN to building a loss function, effectively “teaching” the PINN to learn the physical knowledge. A comparison of the fatigue life prediction results of ANN and the proposed PINN models shows that the PINN model outperforms its counterparts with a 38.71% higher prediction accuracy. The effects of L-PBF process parameters on the fatigue life of as-built AlSi10Mg is examined using both ANN and PINN, proving the better predictive performance and data-physics consistency of the PINN model. The scheme of this work can inform the efficient prediction of fatigue life of additively manufactured alloys and also reverse prediction or fast finding of optimized process parameters for additive manufacturing.

Adaptive Hybrid Prediction Model for Adapting to Data Distribution Shifts in Machining Quality Prediction

Abstract The accuracy of data-driven intelligent prediction for machining quality relies on the training samples. However, in actual applications, the continuous operation of machining equipment leads to gradual distribution shifts between the process data and the training samples for modeling. The shifts result in a degradation in the performance of predictive model, previous studies have often overlooked this issue. To tackle with the intricate problem, this research proposes a real-time model optimization approach. Firstly, a method for detecting machining data distribution shifts based on the two-sample Kolmogorov-Smirnov (KS) test is proposed. Then, an adaptive hybrid prediction model (AHPM) capable of real-time optimization is developed. This model consists of a deep neural network (DNN) and a broad learning system (BLS). DNN plays a primary role in prediction within the hybrid model with excellent generalization capability. BLS quickly completes optimization prior to DNN with its unique parameter update mechanism to compensate for prediction loss. Experimental results indicate that AHPM achieves the shortest optimization time while maintaining high accuracy, with post-optimization error reduction rates for MSE, MAE, and MAPE all exceeding 10%. In the test of application to actual machining cases, accuracy improved by 8.88% compared to traditional methods without optimization.

Experimental study evaluating the performance of thermal conductivity prediction models for air–water saturated weathered sandstone heritage

As a critical parameter, thermal conductivity directly determines the heat transfer and temperature variation within rocks, which can lead to mechanical damage and chemical corrosion. Consequently, understanding the thermal conductivity of stone heritage is vital for assessing their deterioration mechanisms and developing effective conservation strategies. This study obtained sandstone samples from the Yungang Grottoes and subjected them to freeze–thaw cycle experiments to generate weathered sandstone samples. Subsequently, the thermal conductivity of these samples was measured under both dry and water-saturated state using the transient plane source method. To analyze the relationship between air–water saturation, porosity, and thermal conductivity, a saturation influence coefficient was introduced. Thereafter, the effectiveness and applicability of 13 commonly used thermal conductivity mixing law prediction models were evaluated based on experimental data. The results suggested that the influence of water saturation on the thermal conductivity of rocks varies with porosity, and water saturation significantly enhances the thermal conductivity of weathered sandstone. Among the 13 common models, the Geometric mean model was found to be more accurate than other models, with superior performance in both dry (MAE, RMSE, MAPE are 0.148, 0.214, 5.59% respectively) and water-saturated (MAE, RMSE, MAPE are 0.244, 0.170, 8.4% respectively) state. The Albert model demonstrates a good fit in the dry state, whereas the Walsh model (with maximum effect), Ribaud model, and Huang model also exhibit good fitting efficacy in the water-saturated state. This study provides a solid foundation for better predicting the thermal conductivity of weathered stone heritage and developing effective preventive conservation strategies.

Machine learning algorithms to predict treatment success for patients with pulmonary tuberculosis.

Despite advancements in detection and treatment, tuberculosis (TB), an infectious illness caused by the Mycobacterium TB bacteria, continues to pose a serious threat to world health. The TB diagnosis phase includes a patient's medical history, physical examination, chest X-rays, and laboratory procedures, such as molecular testing and sputum culture. In artificial intelligence (AI), machine learning (ML) is an advanced study of statistical algorithms that can learn from historical data and generalize the results to unseen data. There are not many studies done on the ML algorithm that enables the prediction of treatment success for patients with pulmonary TB (PTB). The objective of this study is to identify an effective and predictive ML algorithm to evaluate the detection of treatment success in PTB patients and to compare the predictive performance of the ML models. In this retrospective study, a total of 1236 PTB patients who were given treatment under a randomized controlled clinical trial at the ICMR-National Institute for Research in Tuberculosis, Chennai, India were considered for data analysis. The multiple ML models were developed and tested to identify the best algorithm to predict the sputum culture conversion of TB patients during the treatment period. In this study, decision tree (DT), random forest (RF), support vector machine (SVM) and naïve bayes (NB) models were validated with high performance by achieving an area under the curve (AUC) of receiver operating characteristic (ROC) greater than 80%. The salient finding of the study is that the DT model was produced as a better algorithm with the highest accuracy (92.72%), an AUC (0.909), precision (95.90%), recall (95.60%) and F1-score (95.75%) among the ML models. This methodology may be used to study the precise ML model classification for predicting the treatment success of TB patients during the treatment period.

Developing a suicide risk model for use in the Indian Health Service

We developed and evaluated an electronic health record (EHR)-based model for suicide risk specific to an American Indian patient population. Using EHR data for all patients over 18 with a visit between 1/1/2017 and 10/2/2021, we developed a model for the risk of a suicide attempt or death in the 90 days following a visit. Features included demographics, medications, diagnoses, and scores from relevant screening tools. We compared the predictive performance of logistic regression and random forest models against existing suicide screening, which was augmented to include the history of previous attempts or ideation. During the study, 16,835 patients had 331,588 visits, with 490 attempts and 37 deaths by suicide. The logistic regression and random forest models (area under the ROC (AUROC) 0.83 [0.80–0.86]; both models) performed better than enhanced screening (AUROC 0.64 [0.61–0.67]). These results suggest that an EHR-based suicide risk model can add value to existing practices at Indian Health Service clinics.

Assessing the relative importance of vitamin D deficiency in cardiovascular health.

Previous research has suggested a potential link between vitamin D (VD) deficiency and adverse cardiovascular health outcomes, although the findings have been inconsistent. This study investigates the association between VD deficiency and cardiovascular disease (CVD) within the context of established CVD risk factors. We utilized a Random Forest model to predict both CVD and VD deficiency risks, using a dataset of 1,078 observations from a rural Chinese population. Feature importance was evaluated using SHapley Additive exPlanations (SHAP) to discern the impact of various risk factors on the model's output. The results showed that the model for CVD prediction achieved a high accuracy of 87%, demonstrating robust performance across precision, recall, and F1 score metrics. Conversely, the VD deficiency prediction model exhibited suboptimal performance, with an accuracy of 52% and lower precision, recall, and F1 scores. Feature importance analysis indicated that traditional risk factors such as systolic blood pressure, diastolic blood pressure, age, body mass index, and waist-to-hip ratio significantly influenced CVD risk, collectively contributing to 70% of the model's predictive power. Although VD deficiency was associated with an increased risk of CVD, its importance in predicting CVD risk was notably low. Similarly, for VD deficiency prediction, CVD risk factors such as systolic blood pressure, glucose levels, diastolic blood pressure, and body mass index emerged as influential features. However, the overall predictive performance of the VD deficiency prediction model was weak (52%), indicating the absence of VD deficiency-related risk factors. Ablation experiments confirmed the relatively lower importance of VD deficiency in predicting CVD risk. Furthermore, the SHAP partial dependence plot revealed a nonlinear relationship between VD levels and CVD risk. In conclusion, while VD deficiency appears directly or indirectly associated with increased CVD risk, its relative importance within predictive models is considerably lower when compared to other risk factors. These findings suggest that VD deficiency may not warrant primary focus in CVD risk assessment and prevention strategies, however, further research is needed to explore the causal relationship between VD deficiency and CVD risk.

Deep spatial-temporal information fusion dynamic graph convolutional network for traffic flow prediction

Abstract Predicting traffic flow is vital for advancing intelligent transportation systems, but achieving precise forecasts remains challenging. To deeply explore the complex and diverse dynamic spatio-temporal correlation of traffic data, this study proposes a multi-information fusion spatio-temporal dynamic graph convolution model for traffic flow prediction. In the model construction, firstly, a dynamic graph generation module is designed to fully capture the complex spatial correlation of the traffic network and model the changing spatio-temporal interaction; secondly, the spatial self-attention mechanism is used to dynamically focus on the spatial relationship between different nodes, and combined with the dynamic graph convolution network to extract the spatial dynamic correlation features of the road network to study deep spatiotemporal information; Finally, existing models rarely consider other traffic information related to traffic flow, and a spatiotemporal information fusion module is designed to enhance the ability of the model to perceive information, thereby improving the performance of traffic flow prediction models. Experiments are conducted on four real traffic datasets using three performance evaluation metrics and 18 baseline models. The results show that the model has higher prediction accuracy.

A Novel Neural Network Prediction Model for Mineral Content Based on the Data of Elemental Logging and Well Logging

The mineral content of shale plays an important role in the study of reservoir characteristics. Obtaining high-precision mineral content data from a simple and straightforward method is helpful for the study of reservoir properties. Due to the complex and varied elemental combination of minerals and the lithological diversity in different formations, there are limitations in the traditional statistical methods for predicting mineral content. In this study, a multi-source fusion dataset is constructed first using a detailed workflow to process multi-source data from elemental logging and conventional well logging from the Funing shale formations in China. A hybrid neural network model consisting of a back-propagation artificial neural network (BP-ANN) and gated recurrent unit (GRU) structure, with a custom constraint loss function, is developed to predict the mineral contents. The results show that the predictive performance of this hybrid multi-source model performs statistically significantly better than that of the BP-ANN and GRU models using single source data. In addition, the accuracy of the hybrid model is further improved by using the custom constraint loss function. This method provides high-resolution distributions of mineral content compared to the X-ray powder diffraction testing method. It will help enhance the understanding of the formation.

Prediction Models for Post-Stroke Hospital Readmission: A Systematic Review.

This study aims to evaluate the predictive performance and methodological quality of post-stroke readmission prediction models, identify key predictors associated with readmission, and provide guidance for selecting appropriate risk assessment tools. A comprehensive literature search was conducted from inception to February 1, 2024. Two independent researchers screened the literature and extracted relevant data using the CHARMS checklist. Eleven studies and 16 prediction models were included, with sample sizes ranging from 108 to 803,124 cases and outcome event incidences between 5.2% and 50.0%. The four most frequently included predictors in the models were length of stay, hypertension, age, and functional disability. Twelve models reported an area under the curve (AUC) ranging from 0.520 to 0.940, and five models provided calibration metrics. Only one model included both internal and external validation, while six models had internal validation. Eleven studies were found to have a high risk of bias (ROB), predominantly in the area of data analysis. This systematic review included 16 readmission prediction models for stroke, which generally exhibited good predictive performance and can effectively identify high-risk patients likely to be readmitted. However, the generalizability of these models remains uncertain due to methodological limitations. Rather than developing new readmission prediction models for stroke, the focus should shift toward external validation and the iterative adaptation of existing models. These models should be tailored to local settings, extended with new predictors if necessary, and presented in an interactive graphical user interface.

Development of post-transplant diabetes mellitus in US Liver Transplant recipients is influenced by OPTN region

Background: Post-transplant diabetes mellitus (PTDM) is associated with significant morbidity and mortality in liver transplant recipients (LTRs). We utilized the Organ Procurement and Transplantation Network (OPTN) database to compare incidence of developing PTDM across the US and develop a risk prediction model for new-onset PTDM using OPTN region as well as donor-, recipient-, and transplant-related factors. Methods: All US adult, primary, deceased donor, LTRs between January 1, 2007, and December 31, 2016, with no prior history of diabetes noted, were identified. Kaplan Meier estimators were utilized to calculate cumulative incidence of PTDM, stratified by OPTN region. Multivariable Cox-proportional hazards models were fitted to estimate hazards of PTDM in each OPTN region and build a risk prediction model, via backward selection. Results: Cumulative incidence of PTDM at 1-year, 3-years and 5-years post-transplant was 12.0%, 16.1%, and 18.9%, respectively. Region 3, followed by region 8, 2 and 9, had the highest adjusted hazards of developing PTDM. Inclusion of OPTN region in a risk prediction model for PTDM in LTRs (including recipient age, sex, race, education, insurance coverage, BMI, primary liver disease, cold ischemia time and donor history of diabetes), modestly improved performance (C-statistic=0.60). Conclusions: In patients without pre-existing, confirmed DM, the incidence of PTDM in LTRs varied across OPTN regions, with highest hazards in region 3, followed by region 8, 2 and 9. Performance of a novel risk prediction model for PTDM in LTRs has improved performance with the inclusion of OPTN region. Vigilance is recommended to centers in high-risk regions to identify PTDM and mitigate its development.

Research progress of cardiovascular disease risk prediction models among patients with chronic kidney disease

Patients with chronic kidney disease (CKD) have a relatively high risk of cardiovascular disease (CVD). Risk stratification guided by CVD risk prediction models is essential for managing CKD populations. We reviewed the outcome events, predictive variables, modeling methods, and predictive performance of CVD risk prediction models in CKD populations. We found a large variability in predictive outcomes, number of predictors, and sample sizes across studies. The models tended to overestimate the CVD risk of CKD populations. There are few independently validated or constructed CVD risk prediction models for CKD populations in developing countries, and in particular, there is a lack of independent external validation studies of model calibration. Future studies should comply with the reporting standards of risk prediction models to better support the application of CVD risk prediction models for CKD populations.

Forecasting financial distress in listed Chinese construction firms: leveraging ensemble learning and non-financial variables

The construction industry, characterized by high business failure rates due to complex, capital-intensive projects, is crucial for China’s economy. Predicting financial distress and early warnings is crucial to prevent crises. While financial ratios are commonly used, non-financial information remains underexplored, with limited empirical evidence supporting its effectiveness in enhancing predictive accuracy. Additionally, most studies on financial distress prediction focus on constructing single classifiers without utilizing ensemble models that integrate multiple algorithms, resulting in suboptimal prediction accuracy. To address these problems, this study presents a model that uses accounting variables and firm characteristics, construction market dynamics, and macroeconomic indicators to predict the probability of failure one to two years in advance. The model uses a soft voting-based ensemble algorithm, financial ratios refined through the recursive feature elimination with cross-validation (RFECV) algorithm, and dataset balancing via Synthetic Minority Over-Sampling Technique + Tomek Link (SMOTETomek). The comparison results indicate that the predictive performance of the soft voting ensemble model outperforms all single classifiers across all combinations of input variables and prediction years. Additionally, incorporating non-financial variables, such as firm characteristics, construction market dynamics, and macroeconomic indicators, further enhances the model’s accuracy. The proposed model can be effectively employed to help stakeholders mitigate the risks associated with the financial distress of construction companies before the project implementation phase.

Predicting 3-month poor functional outcomes of acute ischemic stroke in young patients using machine learning

BackgroundPrediction of short-term outcomes in young patients with acute ischemic stroke (AIS) may assist in making therapy decisions. Machine learning (ML) is increasingly used in healthcare due to its high accuracy. This study aims to use a ML-based predictive model for poor 3-month functional outcomes in young AIS patients and to compare the predictive performance of ML models with the logistic regression model.MethodsWe enrolled AIS patients aged between 18 and 50 years from the Third Chinese National Stroke Registry (CNSR-III), collected between 2015 and 2018. A modified Rankin Scale (mRS) ≥ 3 was a poor functional outcome at 3 months. Four ML tree models were developed: The extreme Gradient Boosting (XGBoost), Light Gradient Boosted Machine (lightGBM), Random Forest (RF), and The Gradient Boosting Decision Trees (GBDT), compared with logistic regression. We assess the model performance based on both discrimination and calibration.ResultsA total of 2268 young patients with a mean age of 44.3 ± 5.5 years were included. Among them, (9%) had poor functional outcomes. The mRS at admission, living alone conditions, and high National Institutes of Health Stroke Scale (NIHSS) at discharge remained independent predictors of poor 3-month outcomes. The best AUC in the test group was XGBoost (AUC = 0.801), followed by GBDT, RF, and lightGBM (AUCs of 0.795, 0, 794, and 0.792, respectively). The XGBoost, RF, and lightGBM models were significantly better than logistic regression (P < 0.05).ConclusionsML outperformed logistic regression, where XGBoost the boost was the best model for predicting poor functional outcomes in young AIS patients. It is important to consider living alone conditions with high severity scores to improve stroke prognosis.

Novel plasma protein biomarkers: A time-dependent predictive model for Alzheimer's disease

BackgroundThe accurate prediction of Alzheimer's disease (AD) is crucial for the efficient management of its progression. The objective of this research was to construct a new risk predictive model utilizing novel plasma protein biomarkers for predicting AD incidence in the future and analyze their potential biological correlation with AD incidence. MethodsA cohort of 440 participants aged 60 years and older from the Alzheimer's Disease Neuroimaging Initiative (ADNI) longitudinal cohort was utilized. The baseline plasma proteomics data was employed to conduct Cox regression, LASSO regression, and cross-validation to identify plasma protein signatures predictive of AD risk. Subsequently, a multivariable Cox proportional hazards model based on these signatures was constructed. The performance of the risk prediction model was evaluated using time-dependent receiver operating characteristic (t-ROC) curves and Kaplan-Meier curves. Additionally, we analyzed the correlations between protein signature expression in plasma and predicted AD risk, the time of AD onset, the expression of protein signatures in cerebrospinal fluid (CSF), the expression of CSF and plasma biomarkers, and APOE ε4 genotypes. Colocalization and Mendelian randomization analyses was conducted to investigate the association between protein features and AD risk. GEO database was utilized to analyze the differential expression of protein features in the blood and brain of AD patients. ResultsWe identified seven protein signatures (APOE, CGA, CRP, CCL26, CCL20, NRCAM, and PYY) that independently predicted AD incidence in the future. The risk prediction model demonstrated area under the ROC curve (AUC) values of 0.77, 0.76, and 0.77 for predicting AD incidence at 4, 6, and 8 years, respectively. Furthermore, the model remained stable in the range of the 3rd to the 12th year (ROC ≥ 0.74). The low-risk group, as defined by the model, exhibited a significantly later AD onset compared to the high-risk group (P < 0.0001). Moreover, all protein signatures exhibited significant correlations with AD risk (P < 0.001) and the time of AD onset (P < 0.01). There was no strong correlation between the protein expression levels in plasma and CSF, as well as AD CSF biomarkers. APOE, CGA, and CRP exhibited significantly lower expression levels in APOE ε4 positive individuals (P < 0.05). Additionally, colocalization analysis reveals a significant association between AD and SNP loci in APOE. Mendelian randomization analysis shows a negative correlation between NRCAM and AD risk. Transcriptomic analysis indicates a significant downregulation of NRCAM and PYY in the peripheral blood of AD patients (P < 0.01), while APOE, CGA, and NRCAM are significantly downregulated in the brains of AD patients (P < 0.0001). ConclusionOur research has successfully identified protein signatures in plasma as potential risk biomarkers that can independently predict AD onset in the future. Notably, this risk prediction model has demonstrated commendable predictive performance and stability over time. These findings underscore the promising utility of plasma protein signatures in dynamically predicting the risk of AD, thereby facilitating early screening and intervention strategies.

Impacts of age and BMI on vancomycin model choice in a Bayesian software: Lessons from a very large multi-site retrospective study.

Model-informed precision dosing (MIPD) optimizes drug doses based on pharmacokinetic (PK) model predictions, necessitating careful selection of models tailored to patient characteristics. This study evaluates the predictive performance of various vancomycin PK models across diverse age and BMI categories, drawing insights from a large multi-site database. Adults receiving vancomycin intravenous therapy at United States health systems between January 1, 2022, and December 31, 2023, were included. Patient demographics, vancomycin administration records, and therapeutic drug monitoring levels (TDMs) were collected from the InsightRX database. Age and body mass index (BMI)-based subgroups were formed to assess model performance, with predictions made iteratively. The optimal model for each age-BMI subgroup was chosen based on predefined criteria: models were filtered for mean percentage error (MPE) ≤ 20% and normalized root mean squared error (RMSE) < 8 mg/L, and then the most accurate among them was selected. A total of 384,876 treatment courses across 155 US health systems were analyzed, contributing 841,604 TDMs. Eleven models were compared, showing varying accuracy across age-BMI categories (41%-73%), with higher accuracy observed once TDMs were available for Bayesian estimates of individual PK parameters. Models performed more poorly in younger adults compared to older adults, and the optimal model differed depending on age-BMI categories and prediction methods. Notably, in the a priori period, the Colin model performed best in adults aged 18-64 years across most BMI categories; the Goti/Tong model performed best in the older, non-obese adults; and the Hughes model performed best in many of the obese categories. Our study identifies specific vancomycin PK models that demonstrate superior predictions across age-BMI categories in MIPD applications. Our findings underscore the importance of tailored model selection for vancomycin management, especially highlighting the need for improved models in younger adult patients. Further research into the clinical implications of model performance is warranted to enhance patient care outcomes.