Nearest neighbor imputation algorithms: a critical evaluation

Mapping conditional health literacy and digital health literacy in patients with inflammatory bowel disease to optimise availability of digital health information: a cross-sectional study

BACKGROUND AND AIMS

Health literacy and digital health literacy are crucial for spreading information that enhances self-management and health outcomes. IBD patients have called for relevant and reliable information to enable self-management. However, mapping conditional capacities for adapting IBD health information remains unaddressed. This study examines IBD patients' health literacy and digital health literacy covariance with clinical, demographic and patient-reported outcomes.

METHODS

This cross-sectional study recruited patients between April 2023 to February 2024 from a Norwegian university hospital. Canonical correlations identified maximum covariance between health literacy and digital health literacy dimensions against clinical, demographic and patient-reported characteristics. Hierarchical clustering of covariance patterns were compared on external variables using bivariate analyses and logistic regression.

RESULTS

Of 432 consents, 380 (87.96%) IBD patients ≥ 18 years were included. Mean age was 43.6 (14.9) years, 173 (45.5%) had UC, 207 (54.5%) had CD, and 108 (53%) were male. Self-efficacy, illness perception, health status and age correlated with several health literacy and digital health literacy dimensions. Of two identified patient clusters, cluster 1 embodied patients with lowest levels of health literacy, digital health literacy, self-efficacy, health status, illness perception and longest disease duration. Cluster 1 demonstrated significantly lower medication adherence and QoL, higher rates of unemployment, elevated disease activity and fewer receiving biological treatment. Disease activity and biological treatment were the strongest predictors of cluster membership.

CONCLUSIONS

The findings emphasize the necessity of addressing clinical characteristics alongside health literacy and digital health literacy in the dissemination of IBD health information.

Single-cell morphology encodes functional subtypes of senescence in aging human dermal fibroblasts

Cellular senescence, a hallmark of aging, reveals context-dependent phenotypes across multiple biological length scales. Despite its mechanistic importance, identifying and characterizing senescence across cell populations is challenging. Using primary dermal fibroblasts, we combined single-cell imaging, machine learning, several induced senescence conditions, and multiple protein biomarkers to define functional senescence subtypes. Single-cell morphology analysis revealed 11 distinct morphology clusters. Among these, we identified three as bona fide senescence subtypes (C7, C10, and C11), with C10 exhibiting the strongest age dependence within an aging cohort. In addition, we observed that a donor's senescence burden and subtype composition were indicative of susceptibility to doxorubicin-induced senescence. Functional analysis revealed subtype-dependent responses to senotherapies, with C7 being most responsive to the combination of dasatinib and quercetin. Our single-cell analysis framework, SenSCOUT, enables robust identification and classification of senescence subtypes, offering applications in next-generation senotherapy screens, with potential toward explaining heterogeneous senescence phenotypes based on the presence of senescence subtypes.

Unleashing the power of intelligence: revolutionizing malaria outbreak preparedness with an advanced warning system in Benin, West Africa

BACKGROUND

Malaria is a significant vector-borne disease that exhibits high sensitivity to climatic variations within the West African region. In Benin, the effective prevention and mitigation of malaria pose considerable challenges, primarily due to the prevailing conditions of poverty and environmental adversities. This study endeavours to devise an advanced system for early detection and warning of malaria outbreaks in the northern part of Benin, employing monthly time series data pertaining to climatic variables.

METHODS

Monthly climate data were sourced from Meteorological Agency of Benin (METEO-Benin), alongside malaria incidence data procured from the database of the Benin Ministry of Health, that covered the timeframe of 2009-2021. To ascertain the influence of climatic variables on malaria incidence, principal component analysis was applied. Subsequently, an intelligent model for forecasting malaria outbreaks was developed using support vector machine (SVM) algorithm. The developed model for malaria outbreaks was then employed to establish an intelligent system for warning and forecasting malaria incidence on a monthly basis, utilising the Meteostat platform, an online weather data service provider, in conjunction with the Streamlit framework. This application exhibits responsiveness and compatibility across all web browsers.

RESULTS

Relative humidity and maximal temperature significantly influence malaria incidence in the northern region of Benin. SVM regression algorithm forecasts 80% prediction rate for malaria incidence. Consequently, the intelligent malaria outbreak warning system was successfully devised, enabling the automatic and manual prediction of monthly malaria incidence rates within the districts of northern Benin.

CONCLUSIONS

This system serves as a valuable tool for stakeholders and policymakers, facilitating proactive measures to curtail malaria transmission in Benin.

Positive Association Between the Cardiometabolic Index and the Risk of Male Biochemical Androgen Deficiency in Adults

Metabolic disorders are associated with testosterone deficiency, and the cardiometabolic index (CMI) is a recently identified metabolic indicator. The relationship between male biochemical androgen deficiency (MBAD), a precursor to testosterone deficiency, and CMI remains unclear. In this cross-sectional study, we analyzed data from the National Health and Nutrition Examination Survey (NHANES) 2013-2016 to investigate the relationship between MBAD and CMI in men. This study included 1229 participants; among which, 209 participants had MBAD. Machine learning models identified that the importance of CMI on MBAD was in the top three. After adjusting for all covariates, we found a positive association between CMI and MBAD. Restricted cubic spline (RCS) curves validated this association both in age and body mass index subgroups. Trend regression showed that participants with a higher CMI tended to have a higher risk of MBAD. The positive association between CMI and MBAD persisted after multiple interpolations, validating the robustness of the results. Altogether, this study suggests that CMI exhibits a stable positive relationship with MBAD.

Data-Driven Prognostication in Distal Medium Vessel Occlusions Using Explainable Machine Learning

BACKGROUND AND PURPOSE

Distal medium vessel occlusions (DMVOs) are estimated to cause acute ischemic stroke in 25%-40% of cases. Prognostic models can inform patient counseling and research by enabling outcome predictions. However, models designed specifically for DMVOs are lacking.

MATERIALS AND METHODS

This retrospective study developed a machine learning model to predict 90-day unfavorable outcome (defined as an mRS score of 3-6) in 164 patients with primary DMVO. A model developed with the TabPFN algorithm used selected clinical, laboratory, imaging, and treatment data with the least absolute shrinkage and selection operator feature selection. Performance was evaluated via 5-repeat 5-fold cross-validation. Model discrimination and calibration were evaluated. SHapley Additive Explanations (SHAP) identified influential features. A Web application deployed the model for individualized predictions.

RESULTS

The model achieved an area under the receiver operating characteristic curve of 0.815 (95% CI, 0.79-0.841) for predicting unfavorable outcome, demonstrating good discrimination, and a Brier score of 0.19 (95% CI, 0.177-0.202), demonstrating good calibration. SHAP analysis ranked admission NIHSS score, premorbid mRS, type of thrombectomy, modified TICI score, and history of malignancy as top predictors. The Web application enables individualized prognostication.

CONCLUSIONS

Our machine learning model demonstrated good discrimination and calibration for predicting 90-day unfavorable outcomes in primary DMVO strokes. This study demonstrates the potential for personalized prognostic counseling and research to support precision medicine in stroke care and recovery.

scNET: learning context-specific gene and cell embeddings by integrating single-cell gene expression data with protein-protein interactions

Recent advances in single-cell RNA sequencing (scRNA-seq) techniques have provided unprecedented insights into the heterogeneity of various tissues. However, gene expression data alone often fails to capture and identify changes in cellular pathways and complexes, as they are more discernible at the protein level. Moreover, analyzing scRNA-seq data presents further challenges due to inherent characteristics such as high noise levels and zero inflation. In this study, we propose an approach to address these limitations by integrating scRNA-seq datasets with a protein-protein interaction network. Our method utilizes a unique dual-view architecture based on graph neural networks, enabling joint representation of gene expression and protein-protein interaction network data. This approach models gene-to-gene relationships under specific biological contexts and refines cell-cell relations using an attention mechanism. Next, through comprehensive evaluations, we demonstrate that scNET better captures gene annotation, pathway characterization and gene-gene relationship identification, while improving cell clustering and pathway analysis across diverse cell types and biological conditions.

Estimating the Burden of Common Mental Disorders Attributable to Lifestyle Factors: Protocol for the Global Burden of Disease Lifestyle and Mental Disorder (GLAD) Project

BACKGROUND

The Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) collects and calculates risk-outcome data for modifiable lifestyle exposures (eg, dietary intake) and physical health outcomes (eg, cancers). These estimates form a critical digital resource tool, the GBD VizHub data visualization tool, for governments and policy makers to guide local, regional, and global health decisions. Despite evidence showing the contributions of lifestyle exposures to common mental disorders (CMDs), such as depression and anxiety, GBD does not currently generate these lifestyle exposure-mental disorder outcome pairings. This gap is due to a lack of uniformly collected and analyzed data about these exposures as they relate to CMDs. Such data are required to quantify whether, and to what degree, the global burden of CMDs could be reduced by targeting lifestyle factors at regional and global levels. We have established the Global burden of disease Lifestyle And mental Disorder (GLAD) Taskforce to address this gap.

OBJECTIVE

This study aims to generate the necessary estimates to afford the inclusion of lifestyle exposures as risk factors for CMDs in the GBD study and the GBD digital visualization tools, initially focusing on the relationship between dietary intake and CMDs.

METHODS

The GLAD project is a multicenter, collaborative effort to integrate lifestyle exposures as risk factors for CMDs in the GBD study. To achieve this aim, global epidemiological studies will be recruited to conduct harmonized data analyses estimating the risk, odds, or hazards of lifestyle exposures with CMD outcomes. Initially, these models will focus on the relationship between dietary intake, as defined by the GBD, and anxiety and depression.

RESULTS

As of August 2024, 18 longitudinal cohort studies from 9 countries (Australia: n=4; Brazil: n=1; France: n=1; Italy: n=3; The Netherlands: n=3; New Zealand: n=1; South Africa: n=1; Spain: n=1; and United Kingdom: n=3) have agreed to participate in the GLAD project.

CONCLUSIONS

Our comprehensive, collaborative approach allows for the concurrent execution of a harmonized statistical analysis protocol across multiple, internationally renowned epidemiological cohorts. These results will be used to inform the GBD study and incorporate lifestyle risk factors for CMD in the GBD digital platform. Consequently, given the worldwide influence of the GBD study, findings from the GLAD project can offer valuable insights to policy makers worldwide around lifestyle-based mental health care.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/65576.

Associations between cardiac rehabilitation structure and processes and dietary habits after myocardial infarction: a nationwide registry study

AIMS

Improved dietary habits are important for successful secondary prevention after myocardial infarction (MI), with counselling and support on healthy dietary habits constituting a cornerstone of cardiac rehabilitation (CR). However, there is limited knowledge on how to optimize CR organization to motivate patients to adopt healthy dietary habits. We aimed to explore associations between CR programme structure, processes, and self-reported dietary habits 1 year post-MI.

METHODS AND RESULTS

Organizational data from 73 Swedish CR centres and patient-level data from 5248 CR patients were analysed using orthogonal partial least squares discriminant analysis to identify predictors for healthy dietary habits. Variables of importance for the projection (VIP) values exceeding 0.80 were considered meaningful. Key predictors included the CR centre having a medical director [VIP (95% confidence interval)] [1.86 (1.1-2.62)], high self-reported team spirit [1.63 (1.29-1.97)], nurses have formal training in counselling methods [1.20 (0.75-1.65)], providing discharge information on risk factors [2.23 (1.82-2.64)] and lifestyle [1.81 (1.31-2.31)], time dedicated to patient interaction during follow-up [1.60 (0.80-2.40)], and centres aiming for patients to have the same nurse throughout follow-up [1.54 (1.17-1.91)]. The more positive predictors a CR centre reported to follow, the further improvement in patient-level dietary habits, were analysed by multivariable regression analysis [odds ratio for each additional positive predictor reported 1.03 (1.02-1.05), P < 0.001].

CONCLUSION

Several variables related to CR structure and processes were identified as predictors for patients reporting healthier dietary habits. These findings offer guidance for CR centres in resource allocation and optimizing patient benefits of CR attendance.

Feasibility of a Machine Learning Classifier for Predicting Post-Induction Hypotension in Non-Cardiac Surgery

PURPOSE

To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries.

MATERIALS AND METHODS

Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an open-source registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers.

RESULTS

A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767-0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763-0.772), AdaBoost regressor (0.752; 95% CI, 0.743-0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669-0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001).

CONCLUSION

ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Machine learning-based predictive models for perioperative major adverse cardiovascular events in patients with stable coronary artery disease undergoing noncardiac surgery

BACKGROUND AND OBJECTIVE

Accurate prediction of perioperative major adverse cardiovascular events (MACEs) is crucial, as it not only aids clinicians in comprehensively assessing patients' surgical risks and tailoring personalized surgical and perioperative management plans, but also for information-based shared decision-making with patients and efficient allocation of medical resources. This study developed and validated a machine learning (ML) model using accessible preoperative clinical data to predict perioperative MACEs in stable coronary artery disease (SCAD) patients undergoing noncardiac surgery (NCS).

METHODS

We collected data from 9171 adult SCAD patients who underwent NCS and extracted 64 preoperative variables. First, the optimal data imputation, resampling, and feature selection methods were compared and selected to deal with missing data values and imbalances. Then, nine independent machine learning models (logistic regression (LR), support vector machine, Gaussian Naive Bayes (GNB), random forest, gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine, categorical boosting (CatBoost), and deep neural network) and a stacking ensemble model were constructed and compared with the validated Revised Cardiac Risk Index's (RCRI) model for predictive performance, which was evaluated using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC), calibration curve, and decision curve analysis (DCA). To reduce overfitting and enhance robustness, we performed hyperparameter tuning and 5-fold cross-validation. Finally, the Shapley additive interpretation (SHAP) method and a partial dependence plot (PDP) were used to determine the optimal ML model.

RESULTS

Of the 9,171 patients, 514 (5.6 %) developed MACEs. 24 significant preoperative features were selected for model development and evaluation. All ML models performed well, with AUROC above 0.88 and AUPRC above 0.39, outperforming the AUROC (0.716) and AUPRC (0.185) of RCRI (P < 0.001). The best independent model was XGBoost (AUROC = 0.898, AUPRC = 0.479). The calibration curve accurately predicted the risk of MACEs (Brier score = 0.040), and the DCA results showed that XGBoost had a high net benefit for predicting MACEs. The top-ranked stacking ensemble model, consisting of CatBoost, GBDT, GNB, and LR, proved to be the best (AUROC 0.894, AUPRC 0.485). We identified the top 20 most important features using the mean absolute SHAP values and depicted their effects on model predictions using PDP.

CONCLUSIONS

This study combined missing-value imputation, feature screening, unbalanced data processing, and advanced machine learning methods to successfully develop and verify the first ML-based perioperative MACEs prediction model for patients with SCAD, which is more accurate than RCRI and enables effective identification of high-risk patients and implementation of targeted interventions to reduce the incidence of MACEs.

Association between Triglyceride-Glucose Index and Prognosis in Critically Ill Patients with Acute Coronary Syndrome: Evidence from the MIMIC Database

Background: This study aimed to investigate the association between triglyceride-glucose (TyG) index and prognosis in critically ill patients with acute coronary syndrome (ACS), exploring potential heterogeneity of the association among patient subgroups with different characteristics. Methods: Records of patients with ACS were extracted from the MIMIC-IV database. The association between TyG index and mortality was analyzed using Cox proportional-hazard regression model, while potential non-linear associations were assessed using restricted cubic spline (RCS) regression. Meanwhile, linear regression model was used to explore the association between TyG index and length of stay in hospital or ICU. Subpopulation Treatment Effect Pattern Plot (STEPP) was utilized to explore the potential heterogeneous subgroups. Time-dependent Receiver Operating Characteristic (ROC) curve analyses were performed to compare the predictive ability of different Cox proportional-hazard regression models (with or without TyG index). Results: A total of 849 patients were enrolled. Multivariate Cox regression analyses demonstrated that TyG index was significantly associated with 28-day mortality (HR:2.13 [95%CI: 1.23-3.68], P<0.01) and 365-day mortality (HR:1.65 [95%CI: 1.11-2.47], P<0.01). RCS regression analyses revealed an inverted U-shaped association between TyG index and 28-day mortality (P for non-linearity=0.027) and a linear association between TyG index and 365-day mortality (P for non-linearity =0.086). There were subgroups specified by age for 28-day mortality (P for interaction=0.04) and 365-day mortality (P for interaction<0.01), with a cut-off point of 70 years old obtained by STEPP. TyG index was associated with a higher risk of mortality in subgroups aged ≤ 70 years old. Time-dependent ROC curve suggested that TyG index could slightly improve the prediction of mortality. A higher TyG index was associated with longer time of stay in hospital (β: 1.79 [95%CI: 0.06-3.52], P=0.04). Conclusions: A higher TyG index is associated with both short-term and long-term all-cause mortality in critically ill patients with ACS, especially in short-term all-cause mortality. TyG index is associated with higher mortality risk in patient subgroups aged ≤ 70 years old. A higher TyG index is associated with longer time of stay in hospital. TyG index may serve as a useful prognostic marker for patient management and strategic decision-making in clinical settings.

A hybrid machine learning framework for functional annotation of mitochondrial glutathione transport and metabolism proteins in cancers

BACKGROUND

Alterations of metabolism, including changes in mitochondrial metabolism as well as glutathione (GSH) metabolism are a well appreciated hallmark of many cancers. Mitochondrial GSH (mGSH) transport is a poorly characterized aspect of GSH metabolism, which we investigate in the context of cancer. Existing functional annotation approaches from machine (ML) or deep learning (DL) models based only on protein sequences, were unable to annotate functions in biological contexts.

RESULTS

We develop a flexible ML framework for functional annotation from diverse feature data. This hybrid ML framework leverages cancer cell line multi-omics data and other biological knowledge data as features, to uncover potential genes involved in mGSH metabolism and membrane transport in cancers. This framework achieves strong performance across functional annotation tasks and several cell line and primary tumor cancer samples. For our application, classification models predict the known mGSH transporter SLC25A39 but not SLC25A40 as being highly probably related to mGSH metabolism in cancers. SLC25A10, SLC25A50, and orphan SLC25A24, SLC25A43 are predicted to be associated with mGSH metabolism in multiple biological contexts and structural analysis of these proteins reveal similarities in potential substrate binding regions to the binding residues of SLC25A39.

CONCLUSION

These findings have implications for a better understanding of cancer cell metabolism and novel therapeutic targets with respect to GSH metabolism through potential novel functional annotations of genes. The hybrid ML framework proposed here can be applied to other biological function classifications or multi-omics datasets to generate hypotheses in various biological contexts. Code and a tutorial for generating models and predictions in this framework are available at: https://github.com/lkenn012/mGSH_cancerClassifiers .

Predicting Childhood and Adolescence Hypertension: Analysis of Predictors Using Machine Learning

BACKGROUND

There has been a substantial burden of hypertension in children and adolescents. Given the availability of primary prevention strategies, it is important to determine predictors for early identification of children and adolescents at risk of hypertension. This study aims to attempt and validate machine learning (ML) algorithms for accurately predicting blood pressure (BP) status (normal, prehypertension, and hypertension) over 1- and 3-year periods, identifying key predictors without compromising model performance.

METHODS

We included a population-based cohort of primary 1 to secondary 6 students (typically aged 6 to 18 years) during the academic years of 1995 to 1996 and 2019 to 2020 in Hong Kong. Thirty-six easy-assessed predictors were initially model childhood BP status. Multiple ML algorithms, decision tree, random forest, k-nearest neighbor, eXtreme Gradient Boosting (XGBoost), and multinomial logistic regression (MLR), were used. Model evaluation was performed by various accuracy metrics. The Shapley Additive Explanations (SHAP) was used to identify key features for both predictions.

RESULTS

A total of 923 301 and 602 179 visit pairs were used for the 1- and 3-year predictions, respectively. XGBoost demonstrated the highest prediction accuracies for 1-year (macro-area under the receiver operating characteristic curve [AUROC] = 0.92, micro-AUROC = 0.91) and 3-year (macro-AUROC = 0.91, micro-AUROC = 0.90) periods. The traditional MLR approach had the lowest accuracies for 1- (macro-AUROC = 0.70, micro-AUROC = 0.68) and 3-year (macro-AUROC = 0.70, micro-AUROC = 0.68) predictions. The SHAP values identified 17 key predictors without the need for direct BP measurements or laboratory tests.

CONCLUSION

ML prediction models can accurately predict childhood prehypertension and hypertension at 1 and 3 years, independent of BP and laboratory measurements. The identified key predictors may inform areas for personalized prevention in hypertension.

Assessment of machine learning classifiers for predicting intraoperative blood transfusion in non-cardiac surgery

BACKGROUND

This study aimed to develop a machine learning classifier for predicting intraoperative blood transfusion in non-cardiac surgeries.

METHODS

Preoperative data from 6255 patients were extracted from the VitalDB database, an open-source registry. The primary outcome was the area under the receiver operating characteristic (AUROC) curve of ML classifiers in predicting intraoperative blood transfusion, defined as the receipt of at least one unit of packed red blood cells. Five different machine learning algorithms including logistic regression, random forest, adaptive boosting, gradient boosting, and the extremely gradient boosting classifiers were used to construct a binary classifier for intraoperative blood transfusion, and their predictive abilities were compared.

RESULTS

337 (5%) patients received intraoperative blood transfusion. In the test-set, the logistic regression classifier demonstrated the highest AUROC (0.836, 95% CI, 0.795-0.876), followed by the gradient boosting classifier (0.810, 95% CI, 0.750-0.868), AdaBoost classifier (0.776, 95% CI, 0.722-0.829), random forest classifier (0.735, 95% CI, 0.698-0.771), and XGBoost classifier (0.721, 95% CI, 0.695-0.747). The logistic regression classifier showed a higher AUROC compared to that of a multivariable logistic regression model (0.836 vs. 0.623, P < 0.001). Among various parameters used to construct the logistic regression classifier, the top three most important features were operation time (0.999), preoperative serum hemoglobin level (0.785), and open surgery (0.530).

CONCLUSION

We successfully developed various ML classifiers using readily available preoperative data to predict intraoperative transfusion in patients undergoing non-cardiac surgeries. In particular, the logistic regression classifier demonstrated the best performance in predicting intraoperative transfusion.

K-nearest neighbor algorithm for imputing missing longitudinal prenatal alcohol data

Aims

The objective of this study is to illustrate the application of a machine learning algorithm, K Nearest Neighbor (k-NN) to impute missing alcohol data in a prospective study among pregnant women.

Methods

We used data from the Safe Passage study (n = 11,083). Daily alcohol consumption for the last reported drinking day and 30 days prior was recorded using the Timeline Follow back method, which generated a variable amount of missing data per participants. Of the 3.2 million person-days of observation, data were missing for 0.36 million (11.4%). Using the k-NN imputed values were weighted for the distances and matched for the day of the week. Since participants with no missing days were not comparable to those with missing data, segments of non-missing data from all participants were included as a reference. Validation was done after randomly deleting data for 5-15 consecutive days from the first trimester.

Results

We found that data from 5 nearest neighbors (i.e., K = 5) and segments of 55 days provided imputed values with least imputation error. After deleting data segments from the first trimester data set with no missing days, there was no difference between actual and predicted values for 64% of deleted segments. For 31% of the segments, imputed data were within +/-1 drink/day of the actual. Imputation accuracy varied by study site because of the differences in the magnitude of drinking and proportion of missing data.

Conclusion

k-NN can be used to impute missing data from longitudinal studies of alcohol during pregnancy with high accuracy.

Prediction of undernutrition and identification of its influencing predictors among under-five children in Bangladesh using explainable machine learning algorithms

BACKGROUND AND OBJECTIVES

Child undernutrition is a leading global health concern, especially in low and middle-income developing countries, including Bangladesh. Thus, the objectives of this study are to develop an appropriate model for predicting the risk of undernutrition and identify its influencing predictors among under-five children in Bangladesh using explainable machine learning algorithms.

MATERIALS AND METHODS

This study used the latest nationally representative cross-sectional Bangladesh demographic health survey (BDHS), 2017-18 data. The Boruta technique was implemented to identify the important predictors of undernutrition, and logistic regression, artificial neural network, random forest, and extreme gradient boosting (XGB) were adopted to predict undernutrition (stunting, wasting, and underweight) risk. The models' performance was evaluated through accuracy and area under the curve (AUC). Additionally, SHapley Additive exPlanations (SHAP) were employed to illustrate the influencing predictors of undernutrition.

RESULTS

The XGB-based model outperformed the other models, with the accuracy and AUC respectively 81.73% and 0.802 for stunting, 76.15% and 0.622 for wasting, and 79.13% and 0.712 for underweight. Moreover, the SHAP method demonstrated that the father's education, wealth, mother's education, BMI, birth interval, vitamin A, watching television, toilet facility, residence, and water source are the influential predictors of stunting. While, BMI, mother education, and BCG of wasting; and father education, wealth, mother education, BMI, birth interval, toilet facility, breastfeeding, birth order, and residence of underweight.

CONCLUSION

The proposed integrating framework will be supportive as a method for selecting important predictors and predicting children who are at high risk of stunting, wasting, and underweight in Bangladesh.

Causal Explanation from Mild Cognitive Impairment Progression using Graph Neural Networks

Mild Cognitive Impairment (MCI) is a transitional stage between normal cognitive aging and dementia. Some individuals with MCI revert to normal, while others progress to dementia. There are limited studies using explainable artificial intelligence on longitudinal data, particularly including genotypes, biomarkers and chronic diseases, to explore these differences. This study introduces a novel approach to understanding MCI progression using explainable graph neural networks. Utilizing longitudinal temporal data, we constructed a comprehensive graph representation of each individual in the study cohort. Our temporal graph convolutional network achieved 72.4% accuracy in predicting MCI transitions, while our causal explanation method outperformed existing explanation techniques in stability, accuracy, and faithfulness. We identified a causal subgraph with informative variables including hypertension, arrhythmia, congestive heart failure, coronary artery disease, stroke, lipid-related issues, and sex.

Predicting place of delivery choice among childbearing women in East Africa: a comparative analysis of advanced machine learning techniques

Background

Sub-Saharan Africa faces high neonatal and maternal mortality rates due to limited access to skilled healthcare during delivery. This study aims to improve the classification of health facilities and home deliveries using advanced machine learning techniques and to explore factors influencing women's choices of delivery locations in East Africa.

Method

The study focused on 86,009 childbearing women in East Africa. A comparative analysis of 12 advanced machine learning algorithms was conducted, utilizing various data balancing techniques and hyperparameter optimization methods to enhance model performance.

Result

The prevalence of health facility delivery in East Africa was found to be 83.71%. The findings showed that the support vector machine (SVM) algorithm and CatBoost performed best in predicting the place of delivery, in which both of those algorithms scored an accuracy of 95% and an AUC of 0.98 after optimized with Bayesian optimization tuning and insignificant difference between them in all comprehensive analysis of metrics performance. Factors associated with facility-based deliveries were identified using association rule mining, including parental education levels, timing of initial antenatal care (ANC) check-ups, wealth status, marital status, mobile phone ownership, religious affiliation, media accessibility, and birth order.

Conclusion

This study underscores the vital role of machine learning algorithms in predicting health facility deliveries. A slight decline in facility deliveries from previous reports highlights the urgent need for targeted interventions to meet Sustainable Development Goals (SDGs), particularly in maternal health. The study recommends promoting facility-based deliveries. These include raising awareness about skilled birth attendance, encouraging early ANC check-up, addressing financial barriers through targeted support programs, implementing culturally sensitive interventions, utilizing media campaigns, and mobile health initiatives. Design specific interventions tailored to the birth order of the child, recognizing that mothers may have different informational needs depending on whether it is their first or subsequent delivery. Furthermore, we recommended researchers to explore a variety of techniques and validate findings using more recent data.

Predicting poor performance on cognitive tests among older adults using wearable device data and machine learning: a feasibility study

Timely implementation of interventions to slow cognitive decline among older adults requires accurate monitoring to detect changes in cognitive function. Factors known to be associated with cognition that can be gathered from accelerometers, user interfaces, and other sensors within wearable devices could be used to train machine learning models and develop wearable-based cognitive monitoring systems. Using data from over 2400 older adults in the National Health and Nutrition Examination Survey (NHANES) we developed prediction models to differentiate older adults with normal cognition from those with poor cognition based on outcomes from three cognitive tests measuring different domains of cognitive function. During repeated cross-validation CatBoost, XGBoost, and Random Forest models performed best when predicting poor cognition based on tests measuring processing speed, working memory, and attention (median AUCs ≥0.82) compared to immediate and delayed recall (median AUCs ≥0.72) and categorical verbal fluency (median AUC ≥ 0.68). Activity and sleep parameters were also more strongly associated with poor cognition based on tests assessing processing speed, working memory, and attention compared to other cognitive subdomains. Our work provides proof of concept that data collatable through wearable devices such as age, education, sleep parameters, activity summaries, and light exposure metrics could be used to differentiate between older adults with normal versus poor cognition. We further identified metrics that could be targets in future causal studies seeking to better understand how sleep and activity parameters influence cognitive function among older adults.

A novel MissForest-based missing values imputation approach with recursive feature elimination in medical applications

BACKGROUND

Missing values in datasets present significant challenges for data analysis, particularly in the medical field where data accuracy is crucial for patient diagnosis and treatment. Although MissForest (MF) has demonstrated efficacy in imputation research and recursive feature elimination (RFE) has proven effective in feature selection, the potential for enhancing MF through RFE integration remains unexplored.

METHODS

This study introduces a novel imputation method, "recursive feature elimination-MissForest" (RFE-MF), designed to enhance imputation quality by reducing the impact of irrelevant features. A comparative analysis is conducted between RFE-MF and four classical imputation methods: mean/mode, k-nearest neighbors (kNN), multiple imputation by chained equations (MICE), and MF. The comparison is carried out across ten medical datasets containing both numerical and mixed data types. Different missing data rates, ranging from 10 to 50%, are evaluated under the missing completely at random (MCAR) mechanism. The performance of each method is assessed using two evaluation metrics: normalized root mean squared error (NRMSE) and predictive fidelity criterion (PFC). Additionally, paired samples t-tests are employed to analyze the statistical significance of differences among the outcomes.

RESULTS

The findings indicate that RFE-MF demonstrates superior performance across the majority of datasets when compared to four classical imputation methods (mean/mode, kNN, MICE, and MF). Notably, RFE-MF consistently outperforms the original MF, irrespective of variable type (numerical or categorical). Mean/mode imputation exhibits consistent performance across various scenarios. Conversely, the efficacy of kNN imputation fluctuates in relation to varying missing data rates.

CONCLUSION

This study demonstrates that RFE-MF holds promise as an effective imputation method for medical datasets, providing a novel approach to addressing missing data challenges in medical applications.

Systemic Metabolic and Volumetric Assessment via Whole-Body [18F]FDG-PET/CT: Pancreas Size Predicts Cachexia in Head and Neck Squamous Cell Carcinoma

Background/Objectives: Cancer-associated cachexia in head and neck squamous cell carcinoma (HNSCC) is challenging to diagnose due to its complex pathophysiology. This study aimed to identify metabolic biomarkers linked to cachexia and survival in HNSCC patients using [18F]FDG-PET/CT imaging and machine learning (ML) techniques. Methods: We retrospectively analyzed 253 HNSCC patients from Vienna General Hospital and the MD Anderson Cancer Center. Automated organ segmentation was employed to quantify metabolic and volumetric data from [18F]FDG-PET/CT scans across 29 tissues and organs. Patients were categorized into low weight loss (LoWL; grades 0-2) and high weight loss (HiWL; grades 3-4) groups, according to the weight loss grading system (WLGS). Machine learning models, combined with Cox regression, were used to identify survival predictors. Shapley additive explanation (SHAP) analysis was conducted to determine the significance of individual features. Results: The HiWL group exhibited increased glucose metabolism in skeletal muscle and adipose tissue (p = 0.01), while the LoWL group showed higher lung metabolism. The one-year survival rate was 84.1% in the LoWL group compared to 69.2% in the HiWL group (p < 0.01). Pancreatic volume emerged as a key biomarker associated with cachexia, with the ML model achieving an AUC of 0.79 (95% CI: 0.77-0.80) and an accuracy of 0.82 (95% CI: 0.81-0.83). Multivariate Cox regression confirmed pancreatic volume as an independent prognostic factor (HR: 0.66, 95% CI: 0.46-0.95; p < 0.05). Conclusions: The integration of metabolic and volumetric data provided a strong predictive model, highlighting pancreatic volume as a key imaging biomarker in the metabolic assessment of cachexia in HNSCC. This finding enhances our understanding and may improve prognostic evaluations and therapeutic strategies.

Generalizability assessment of AI models across hospitals in a low-middle and high income country

The integration of artificial intelligence (AI) into healthcare systems within low-middle income countries (LMICs) has emerged as a central focus for various initiatives aiming to improve healthcare access and delivery quality. In contrast to high-income countries (HICs), which often possess the resources and infrastructure to adopt innovative healthcare technologies, LMICs confront resource limitations such as insufficient funding, outdated infrastructure, limited digital data, and a shortage of technical expertise. Consequently, many algorithms initially trained on data from non-LMIC settings are now being employed in LMIC contexts. However, the effectiveness of these systems in LMICs can be compromised when the unique local contexts and requirements are not adequately considered. In this study, we evaluate the feasibility of utilizing models developed in the United Kingdom (a HIC) within hospitals in Vietnam (a LMIC). Consequently, we present and discuss practical methodologies aimed at improving model performance, emphasizing the critical importance of tailoring solutions to the distinct healthcare systems found in LMICs. Our findings emphasize the necessity for collaborative initiatives and solutions that are sensitive to the local context in order to effectively tackle the healthcare challenges that are unique to these regions.

Predicting Long-Term Clinical Outcomes of Patients With Recurrent Pericarditis

BACKGROUND

Recurrent pericarditis (RP) is a complex condition associated with significant morbidity. Prior studies have evaluated which variables are associated with clinical remission. However, there is currently no established risk-stratification model for predicting outcomes in these patients.

OBJECTIVES

We developed a risk stratification model that can predict long-term outcomes in patients with RP and enable identification of patients with characteristics that portend poor outcomes.

METHODS

We retrospectively studied a total of 365 consecutive patients with RP from 2012 to 2019. The primary outcome was clinical remission (CR), defined as cessation of all anti-inflammatory therapy with complete resolution of symptoms. Five machine learning survival models were used to calculate the likelihood of CR within 5 years and stratify patients into high-risk, intermediate-risk, and low-risk groups.

RESULTS

Among the cohort, the mean age was 46 ± 15 years, and 205 (56%) were women. CR was achieved in 118 (32%) patients. The final model included steroid dependency, total number of recurrences, pericardial late gadolinium enhancement, age, etiology, sex, ejection fraction, and heart rate as the most important parameters. The model predicted the outcome with a C-index of 0.800 on the test set and exhibited a significant ability in stratification of patients into low-risk, intermediate-risk, and high-risk groups (log-rank test; P < 0.0001).

CONCLUSIONS

We developed a novel risk-stratification model for predicting CR in RP. Our model can also aid in stratifying patients, with high discriminative ability. The use of an explainable machine learning model can aid physicians in making individualized treatment decision in RP patients.

Robust identification key predictors of short- and long-term weight status in children and adolescents by machine learning

Background

Early identification of high-risk individuals for weight problems in children and adolescents is crucial for implementing timely preventive measures. While machine learning (ML) techniques have shown promise in addressing this complex challenge with high-dimensional data, feature selection is vital for identifying the key predictors that can facilitate effective and targeted interventions. This study aims to utilize feature selection process to identify a robust and minimal set of predictors that can aid in the early prediction of short- and long-term weight problems in children and adolescents.

Methods

We utilized demographic, physical, and psychological wellbeing predictors to model weight status (normal, underweight, overweight, and obese) for 1-, 3-, and 5-year periods. To select the most influential features, we employed four feature selection methods: (1) Chi-Square test; (2) Information Gain; (3) Random Forest; (4) eXtreme Gradient Boosting (XGBoost) with six ML approaches. The stability of the feature selection methods was assessed by Jaccard's index, Spearman's rank correlation and Pearson's correlation. Model evaluation was performed by various accuracy metrics.

Results

With 3,862,820 million student-visits were included in this population-based study, the mean age of 11.6 (SD = 3.64) for the training set and 10.8 years (SD = 3.50) for the temporal test set. From the initial set of 38 predictors, we identified 6, 9, and 13 features for 1-, 3-, and 5-year predictions, respectively, by the best performed feature selection method of Chi-Square test in XGBoost models. These feature sets demonstrated excellent stability and achieved prediction accuracies of 0.82, 0.73, and 0.70; macro-AUCs of 0.94, 0.86, and 0.83; micro-AUCs of 0.96, 0.93, and 0.92 for different prediction windows, respectively. Weight, height, sex, total score of self-esteem, and age were consistently the most influential predictors across all prediction windows. Additionally, several psychological and social wellbeing predictors showed relatively high importance in long-term weight status prediction.

Conclusions

We demonstrate the potential of ML in identifying key predictors of weight status in children and adolescents. While traditional anthropometric measures remain important, psychological and social wellbeing factors also emerge as crucial predictors, potentially informing targeted interventions to address childhood and adolescence weight problems.

Modeling the Determinants of Subjective Well-Being in Schizophrenia

BACKGROUND

The ultimate goal of successful schizophrenia treatment is not just to alleviate psychotic symptoms, but also to reduce distress and achieve subjective well-being (SWB). We aimed to identify the determinants of SWB and their interrelationships in schizophrenia.

METHODS

Data were obtained from 637 patients with schizophrenia enrolled in multicenter, open-label, non-comparative clinical trials. The SWB under the Neuroleptic Treatment Scale (SWN) was utilized; a cut-off score of 80 indicated a high level of SWB at baseline and 6 months. Various machine learning (ML) algorithms were employed to identify the determinants of SWB. Furthermore, network analysis and structural equation modeling (SEM) were conducted to explore detailed relationship patterns.

RESULTS

The random forest (RF) model had the highest area under the curve (AUC) of 0.794 at baseline. Obsessive-compulsive symptoms (OCS) had the most significant impact on high levels of SWB, followed by somatization, cognitive deficits, and depression. The network analysis demonstrated robust connections among the SWB, OCS, and somatization. SEM analysis revealed that OCS exerted the strongest direct effect on SWB, and also an indirect effect via the mediation of depression. Furthermore, the contribution of OCS at baseline to SWB was maintained 6 months later.

CONCLUSIONS

OCS, somatization, cognition, and depression, rather than psychotic symptoms, exerted significant impacts on SWB in schizophrenia. Notably, OCS exhibited the most significant contribution not only to the current state of well-being but also to follow-up SWB, implying that OCS was predictive of SWB. The findings demonstrated that OCS management is critical for the treatment of schizophrenia.

Using Machine Learning and Multifaceted Preoperative Measures to Predict Adult Cochlear Implant Outcomes: A Prospective Pilot Study

OBJECTIVES

To use machine learning and a battery of measures for preoperative prediction of speech recognition and quality of life (QOL) outcomes after cochlear implant (CI) surgery.

DESIGN

Demographic, audiologic, cognitive-linguistic, and QOL predictors were collected from 30 postlingually deaf adults before CI surgery. K-means clustering separated patients into groups. Reliable change index scores were computed for speech recognition and QOL from pre-CI to 6 months post-CI, and group differences were determined.

RESULTS

Clustering yielded three groups with differences in reliable change index for sentence recognition. One group demonstrated low baseline sentence recognition and only small improvements post-CI, suggesting a group "at risk" for limited benefits. This group showed lower pre-CI scores on verbal learning and memory and lack of musical training.

CONCLUSIONS

Preoperative assessments can prognosticate CI recipients' postoperative performance and identify individuals at risk for experiencing poor sentence recognition outcomes, which may help guide counseling and rehabilitation.

A machine learning-based approach for individualized prediction of short-term outcomes after anterior cervical corpectomy

STUDY DESIGN

A retrospective machine learning (ML) classification study for prognostic modeling after anterior cervical corpectomy (ACC).

PURPOSE

To evaluate the effectiveness of ML in predicting ACC outcomes and develop an accessible, user-friendly tool for this purpose.

OVERVIEW OF LITERATURE

Based on our literature review, no study has examined the capability of ML algorithms to predict major shortterm ACC outcomes, such as prolonged length of hospital stay (LOS), non-home discharge, and major complications.

METHODS

The American College of Surgeons' National Surgical Quality Improvement Program database was used to identify patients who underwent ACC. Prolonged LOS, non-home discharges, and major complications were assessed as the outcomes of interest. ML models were developed with the TabPFN algorithm and integrated into an open-access website to predict these outcomes.

RESULTS

The models for predicting prolonged LOS, non-home discharges, and major complications demonstrated mean areas under the receiver operating characteristic curve (AUROC) of 0.802, 0.816, and 0.702, respectively. These findings highlight the discriminatory capacities of the models: fair (AUROC >0.7) for differentiating patients with major complications from those without, and good (AUROC >0.8) for distinguishing between those with and without prolonged LOS and non-home discharges. According to the SHapley Additive Explanations analysis, single- versus multiple-level surgery, age, body mass index, preoperative hematocrit, and American Society of Anesthesiologists physical status repetitively emerged as the most important variables for each outcome.

CONCLUSIONS

This study has considerably enhanced the prediction of postoperative results after ACC surgery by implementing advanced ML techniques. A major contribution is the creation of an accessible web application, highlighting the practical value of the developed models. Our findings imply that ML can serve as an invaluable supplementary tool to stratify patient risk for this procedure and can predict diverse postoperative adverse outcomes.

A novel assessment of whole-mount Gleason grading in prostate cancer to identify candidates for radical prostatectomy: a machine learning-based multiomics study

Purpose: This study aims to assess whole-mount Gleason grading (GG) in prostate cancer (PCa) accurately using a multiomics machine learning (ML) model and to compare its performance with biopsy-proven GG (bxGG) assessment. Materials and Methods: A total of 146 patients with PCa recruited in a pilot study of a prospective clinical trial (NCT02659527) were retrospectively included in the side study, all of whom underwent 68Ga-PSMA-11 integrated positron emission tomography (PET) / magnetic resonance (MR) before radical prostatectomy (RP) between May 2014 and April 2020. To establish a multiomics ML model, we quantified PET radiomics features, pathway-level genomics features from whole exome sequencing, and pathomics features derived from immunohistochemical staining of 11 biomarkers. Based on the multiomics dataset, five ML models were established and validated using 100-fold Monte Carlo cross-validation. Results: Among five ML models, the random forest (RF) model performed best in terms of the area under the curve (AUC). Compared to bxGG assessment alone, the RF model was superior in terms of AUC (0.87 vs 0.75), specificity (0.72 vs 0.61), positive predictive value (0.79 vs 0.75), and accuracy (0.78 vs 0.77) and showed slightly decreased sensitivity (0.83 vs 0.89) and negative predictive value (0.80 vs 0.81). Among the feature categories, bxGG was identified as the most important feature, followed by pathomics, clinical, radiomics and genomics features. The three important individual features were bxGG, PSA staining and one intensity-related radiomics feature. Conclusion: The findings demonstrate a superior assessment of the developed multiomics-based ML model in whole-mount GG compared to the current clinical baseline of bxGG. This enables personalized patient management by identifying high-risk PCa patients for RP.

Collaborative Filtering for the Imputation of Patient Reported Outcomes

This study addresses the prevalent issue of missing data in patient-reported outcome datasets, particularly focusing on head and neck cancer patient symptom ratings sourced from the MD Anderson Symptom Inventory. Given that many data mining and machine learning algorithms necessitate complete datasets, the accurate imputation of missing data as an initial step becomes crucial. In this study we propose, for the first time, the use of collaborative filtering for imputing missing head and neck cancer patient symptom ratings. Two configurations of collaborative filtering, namely patient-based and symptom-based, leverage known ratings to infer the missing ones. Additionally, this study compares the performance of collaborative filtering with alternative imputation methods such as Multiple Imputation by Chained Equations, Nearest Neighbor Imputation, and Linear interpolation. Performance is compared using Root Mean Squared Error and Mean Absolute Error metrics. Findings demonstrate that collaborative filtering is a viable and comparatively superior approach for imputing missing patient symptom data.

Social determinants of health and disparities in spine surgery: a 10-year analysis of 8,565 cases using ensemble machine learning and multilayer perceptron

BACKGROUND CONTEXT

The influence of SDOH on spine surgery is poorly understood. Historically, researchers commonly focused on the isolated influences of race, insurance status, or income on healthcare outcomes. However, analysis of SDOH is becoming increasingly more nuanced as viewing social factors in aggregate rather than individually may offer more precise estimates of the impact of SDOH on healthcare delivery.

PURPOSE

The aim of this study was to evaluate the effects of patient social history on length of stay (LOS) and readmission within 90 days following spine surgery using ensemble machine learning and multilayer perceptron.

STUDY DESIGN

Retrospective chart review.

PATIENT SAMPLE

8,565 elective and emergency spine surgery cases performed from 2013 to 2023 using our institution's database of longitudinally collected electronic medical record information.

OUTCOMES MEASURES

Patient LOS, discharge disposition, and rate of 90-day readmission.

METHODS

Ensemble machine learning and multilayer perceptron were employed to predict LOS and readmission within 90 days following spine surgery. All other subsequent statistical analysis was performed using SPSS version 28. To further assess correlations among variables, Pearson's correlation tests and multivariate linear regression models were constructed. Independent sample t-tests, paired sample t-tests, one-way analysis of variance (ANOVA) with post-hoc Bonferroni and Tukey corrections, and Pearson's chi-squared test were applied where appropriate for analysis of continuous and categorical variables.

RESULTS

Black patients demonstrated a greater LOS compared to white patients, but race and ethnicity were not significantly associated with 90-day readmission rates. Insured patients had a shorter LOS and lower readmission rates compared to non-insured patients, as did privately insured patients compared to publicly insured patients. Patients discharged home had lower LOS and lower readmission rates, compared to patients discharged to other facilities. Marriage decreased both LOS and readmission rates, underweight patients showcased increased LOS and readmission rates, and religion was shown to impact LOS and readmission rates. When utilizing patient social history, lab values, and medical history, machine learning determined the top 5 most-important variables for prediction of LOS -along with their respective feature importances-to be insurance status (0.166), religion (0.100), ICU status (0.093), antibiotic use (0.061), and case status: elective or urgent (0.055). The top 5 most-important variables for prediction of 90-day readmission-along with their respective feature importances-were insurance status (0.177), religion (0.123), discharge location (0.096), emergency case status (0.064), and history of diabetes (0.041).

CONCLUSIONS

This study highlights that SDOH is influential in determining patient length of stay, discharge disposition, and likelihood of readmission following spine surgery. Machine learning was utilized to accurately predict LOS and 90-day readmission with patient medical history, lab values, and social history, as well as social history alone.

Post-natal gestational age assessment using targeted metabolites of neonatal heel prick and umbilical cord blood: A GARBH-Ini cohort study from North India

Background

Accurate assessment of gestational age (GA) and identification of preterm birth (PTB) at delivery is essential to guide appropriate post-natal clinical care. Undoubtedly, dating ultrasound sonography (USG) is the gold standard to ascertain GA, but is not accessible to the majority of pregnant women in low- and middle-income countries (LMICs), particularly in rural areas and small secondary care hospitals. Conventional methods of post-natal GA assessment are not reliable at delivery and are further compounded by a lack of trained personnel to conduct them. We aimed to develop a population-specific GA model using integrated clinical and biochemical variables measured at delivery.

Methods

We acquired metabolic profiles on paired neonatal heel prick (nHP) and umbilical cord blood (uCB) dried blood spot (DBS) samples (n = 1278). The master data set consists of 31 predictors from nHP and 24 from uCB after feature selection. These selected predictors including biochemical analytes, birth weight, and placental weight were considered for the development of population-specific GA estimation and birth outcome classification models using eXtreme Gradient Boosting (XGBoost) algorithm.

Results

The nHP and uCB full model revealed root mean square error (RMSE) of 1.14 (95% confidence interval (CI) = 0.82-1.18) and of 1.26 (95% CI = 0.88-1.32) to estimate the GA as compared to actual GA, respectively. In addition, these models correctly estimated 87.9 to 92.5% of the infants within ±2 weeks of the actual GA. The classification models also performed as the best fit to discriminate the PTB from term birth (TB) infants with an area under curve (AUC) of 0.89 (95% CI = 0.84-0.94) for nHP and an AUC of 0.89 (95% CI = 0.85-0.95) for uCB.

Conclusion

The biochemical analytes along with clinical variables in the nHP and uCB data sets provide higher accuracy in predicting GA. These models also performed as the best fit to identify PTB infants at delivery.

Attachment and borderline personality disorder as the dance unfolds: A quantitative analysis of a novel paradigm

Current research on personality disorders strives to identify key behavioural and cognitive facets of patient functioning, to unravel the underlying root causes and maintenance mechanisms. This process often involves the application of social paradigms - however, these often only include momentary affective depictions rather than unfolding interactions. This constitutes a limitation in our capacity to probe core symptoms, and leaves potential findings uncovered which could help those who are in close relationships with affected individuals. Here, we deployed a novel task in which subjects interact with four unknown virtual partners in a turn-taking paradigm akin to a dance, and report on their experience with each. The virtual partners embody four combinations of low/high expressivity of positive/negative mood. Higher scores on our symptomatic measures of attachment anxiety, avoidance, and borderline personality disorder (BPD) were all linked to a general negative appraisal of all the interpersonal experiences. Moreover, the negative appraisal of the partner who displayed a high negative/low positive mood was tied with attachment anxiety and BPD symptoms. The extent to which subjects felt responsible for causing partners' distress was most strongly linked to attachment anxiety. Finally, we provide a fully-fledged exploration of move-by-move action latencies and click distances from partners. This analysis underscored slower movement initiation from anxiously attached individuals throughout all virtual interactions. In summary, we describe a novel paradigm for second-person neuroscience, which allowed both the replication of established results and the capture of new behavioural signatures associated with attachment anxiety, and discuss its limitations.

Identification of risk factors on rabies vaccine efficacy from censored data: Pre-travel tests for dogs and cats from Yaoundé (2005-2015)

Little research is available on acquired immunity to rabies in dogs and cats from Central Africa, particularly regarding the legal movements of pets. Movement of domestic animals from rabies-endemic countries like Cameroon to rabies free areas poses one of the main risks for rabies introduction into rabies-free areas. Thus, the aim of this study was to assess the effect of various risk factors on rabies vaccine efficacy in Cameroonian. Since the dependent variable, rabies neutralizing titres, were censored from above (right-censoring), Generalized Additive Model for Location, Scale and Shape (GAMLSS) was used in the analysis. Overall, 85.7% of dogs and 100% of cats had titres greater than or equal to 0.5 IU/mL, which is considered protective. Additionally, compared to cats, the value of the rabies-neutralizing serum titres in dogs was on average smaller by 2.3 IU/mL. For each additional year of age, the value of the rabies-neutralizing serum titre, on average, increased by approximately 0.14 IU/mL. Finally, for each 30 additional days between the date of the last rabies vaccination and the date of the sampling, the value the rabies neutralizing titre, on average, decreased by approximately 0.10 IU/mL, given the species and age at sampling were equivalent. These results are useful for assessing risk and improving surveillance to prevent the introduction of rabies into a country via the international movement of animals.

Neuropsychological Profiles of Deployment-Related Mild Traumatic Brain Injury: A LIMBIC-CENC Study

BACKGROUND AND OBJECTIVES

Traumatic brain injury (TBI) is a concern for US service members and veterans (SMV), leading to heterogeneous psychological and cognitive outcomes. We sought to identify neuropsychological profiles of mild TBI (mTBI) and posttraumatic stress disorder (PTSD) among the largest SMV sample to date.

METHODS

We analyzed cross-sectional baseline data from SMV with prior combat deployments enrolled in the ongoing Long-term Impact of Military-relevant Brain Injury Consortium-Chronic Effects of Neurotrauma Consortium prospective longitudinal study. Latent profile analysis identified symptom profiles using 35 indicators, including physical symptoms, depression, quality of life, sleep quality, postconcussive symptoms, and cognitive performance. It is important to note that the profiles were determined independently of mTBI and probable PTSD status. After profile identification, we examined associations between demographic variables, mTBI characteristics, and PTSD symptoms with symptom profile membership.

RESULTS

The analytic sample included 1,659 SMV (mean age 41.1 ± 10.0 years; 87% male); among them 29% (n = 480) had a history of non-deployment-related mTBI only, 14% (n = 239) had deployment-related mTBI only, 36% (n = 602) had both non-deployment and deployment-related mTBI, and 30% (n = 497) met criteria for probable PTSD. A 6-profile model had the best fit, with separation on all indicators (p < 0.001). The model revealed distinct neuropsychological profiles, representing a combination of 3 self-reported functioning patterns: high (HS), moderate (MS), and low (LS), and 2 cognitive performance patterns: high (HC) and low (LC). The profiles were (1) HS/HC: n=301, 18.1%; (2) HS/LC: n=294, 17.7%; (3) MS/HC: n=359, 21.6%; (4) MS/LC: n=316, 19.0%; (5) LS/HC: n=228, 13.7%; and (6) LS/LC: n=161, 9.7%. SMV with deployment-related mTBI tended to be grouped into lower functioning profiles and were more likely to meet criteria for probable PTSD. Conversely, SMV with no mTBI exposure or non-deployment-related mTBI were clustered in higher functioning profiles and had a lower likelihood of meeting criteria for probable PTSD.

DISCUSSION

Findings suggest varied symptom and functional profiles in SMV, influenced by injury context and probable PTSD comorbidity. Despite diagnostic challenges, comprehensive assessment of functioning and cognition can detect subtle differences related to mTBI and PTSD, revealing distinct neuropsychological profiles. Prioritizing early treatment based on these profiles may improve prognostication and support efficient recovery.

Opioid Nonadherence Risk Prediction of Patients with Cancer-Related Pain Based on Five Machine Learning Algorithms

Objectives

Opioid nonadherence represents a significant barrier to cancer pain treatment efficacy. However, there is currently no effective prediction method for opioid adherence in patients with cancer pain. We aimed to develop and validate a machine learning (ML) model and evaluate its feasibility to predict opioid nonadherence in patients with cancer pain.

Methods

This was a secondary analysis from a cross-sectional study that included 1195 patients from March 1, 2018, to October 31, 2019. Five ML algorithms, such as logistic regression (LR), random forest, eXtreme Gradient Boosting, multilayer perceptron, and support vector machine, were used to predict opioid nonadherence in patients with cancer pain using 43 demographic and clinical factors as predictors. The predictive effects of the models were compared by the area under the receiver operating characteristic curve (AUC_ROC), accuracy, precision, sensitivity, specificity, and F1 scores. The value of the best model for clinical application was assessed using decision curve analysis (DCA).

Results

The best model obtained in this study, the LR model, had an AUC_ROC of 0.82, accuracy of 0.82, and specificity of 0.71. The DCA showed that clinical interventions for patients at high risk of opioid nonadherence based on the LR model can benefit patients. The strongest predictors for adherence were, in order of importance, beliefs about medicines questionnaire (BMQ)-harm, time since the start of opioid, and BMQ-necessity. Discussion. ML algorithms can be used as an effective means of predicting adherence to opioids in patients with cancer pain, which allows for proactive clinical intervention to optimize cancer pain management. This trial is registered with ChiCTR2000033576.

Machine learning models on a web application to predict short-term postoperative outcomes following anterior cervical discectomy and fusion

BACKGROUND

The frequency of anterior cervical discectomy and fusion (ACDF) has increased up to 400% since 2011, underscoring the need to preoperatively anticipate adverse postoperative outcomes given the procedure's expanding use. Our study aims to accomplish two goals: firstly, to develop a suite of explainable machine learning (ML) models capable of predicting adverse postoperative outcomes following ACDF surgery, and secondly, to embed these models in a user-friendly web application, demonstrating their potential utility.

METHODS

We utilized data from the National Surgical Quality Improvement Program database to identify patients who underwent ACDF surgery. The outcomes of interest were four short-term postoperative adverse outcomes: prolonged length of stay (LOS), non-home discharges, 30-day readmissions, and major complications. We utilized five ML algorithms - TabPFN, TabNET, XGBoost, LightGBM, and Random Forest - coupled with the Optuna optimization library for hyperparameter tuning. To bolster the interpretability of our models, we employed SHapley Additive exPlanations (SHAP) for evaluating predictor variables' relative importance and used partial dependence plots to illustrate the impact of individual variables on the predictions generated by our top-performing models. We visualized model performance using receiver operating characteristic (ROC) curves and precision-recall curves (PRC). Quantitative metrics calculated were the area under the ROC curve (AUROC), balanced accuracy, weighted area under the PRC (AUPRC), weighted precision, and weighted recall. Models with the highest AUROC values were selected for inclusion in a web application.

RESULTS

The analysis included 57,760 patients for prolonged LOS [11.1% with prolonged LOS], 57,780 for non-home discharges [3.3% non-home discharges], 57,790 for 30-day readmissions [2.9% readmitted], and 57,800 for major complications [1.4% with major complications]. The top-performing models, which were the ones built with the Random Forest algorithm, yielded mean AUROCs of 0.776, 0.846, 0.775, and 0.747 for predicting prolonged LOS, non-home discharges, readmissions, and complications, respectively.

CONCLUSIONS

Our study employs advanced ML methodologies to enhance the prediction of adverse postoperative outcomes following ACDF. We designed an accessible web application to integrate these models into clinical practice. Our findings affirm that ML tools serve as vital supplements in risk stratification, facilitating the prediction of diverse outcomes and enhancing patient counseling for ACDF.

Prediction of adolescent weight status by machine learning: a population-based study

BACKGROUND

Adolescent weight problems have become a growing public health concern, making early prediction of non-normal weight status crucial for effective prevention. However, few temporal prediction tools for adolescent four weight status have been developed. This study aimed to predict the short- and long-term weight status of Hong Kong adolescents and assess the importance of predictors.

METHODS

A population-based retrospective cohort study of adolescents was conducted using data from a territory-wide voluntary annual health assessment service provided by the Department of Health in Hong Kong. Using diet habits, physical activity, psychological well-being, and demographics, we generated six prediction models for successive weight status (normal, overweight, obese and underweight) using multiclass Decision Tree, Random Forest, k-Nearest Neighbor, eXtreme gradient boosting, support vector machine, logistic regression. Model performance was evaluated by multiple standard classifier metrics and the overall accuracy. Predictors' importance was assessed using Shapley values.

RESULTS

442,898 Primary 4 (P4, Grade 4 in the US) and 344,186 in Primary 6 (P6, Grade 6 in the US) students, with followed up until their Secondary 6 (Grade 12 in the US) during the academic years 1995/96 to 2014/15 were included. The XG Boosts model consistently outperformed all other model in predicting the long-term weight status at S6 from P4 or P6. It achieved an overall accuracy of 0.72 or 0.74, a micro-averaging AUC of 0.92 or 0.93, and a macro-averaging AUC of 0.83 or 0.86, respectively. XG Boost also demonstrated accurate predictions for each predicted weight status, surpassing the AUC values obtained by other models. Weight, height, sex, age, frequency and hours of aerobic exercise were consistently the most important predictors for both cohorts.

CONCLUSIONS

The machine learning approaches accurately predict adolescent weight status in both short- and long-term. The developed multiclass model that utilizing easy-assessed variables enables accurate long-term prediction on weight status, which can be used by adolescents and parents for self-prediction when applied in health care system. The interpretable models may help to provide the early and individualized interventions suggestions for adolescents with weight problems particularly.

Single-cell morphology encodes functional subtypes of senescence in aging human dermal fibroblasts

Cellular senescence is an established driver of aging, exhibiting context-dependent phenotypes across multiple biological length-scales. Despite its mechanistic importance, profiling senescence within cell populations is challenging. This is in part due to the limitations of current biomarkers to robustly identify senescent cells across biological settings, and the heterogeneous, non-binary phenotypes exhibited by senescent cells. Using a panel of primary dermal fibroblasts, we combined live single-cell imaging, machine learning, multiple senescence induction conditions, and multiple protein-based senescence biomarkers to show the emergence of functional subtypes of senescence. Leveraging single-cell morphologies, we defined eleven distinct morphology clusters, with the abundance of cells in each cluster being dependent on the mode of senescence induction, the time post-induction, and the age of the donor. Of these eleven clusters, we identified three bona-fide senescence subtypes (C7, C10, C11), with C10 showing the strongest age-dependence across a cohort of fifty aging individuals. To determine the functional significance of these senescence subtypes, we profiled their responses to senotherapies, specifically focusing on Dasatinib + Quercetin (D+Q). Results indicated subtype-dependent responses, with senescent cells in C7 being most responsive to D+Q. Altogether, we provide a robust single-cell framework to identify and classify functional senescence subtypes with applications for next-generation senotherapy screens, and the potential to explain heterogeneous senescence phenotypes across biological settings based on the presence and abundance of distinct senescence subtypes.

Machine Learning Model with Computed Tomography Radiomics and Clinicobiochemical Characteristics Predict the Subtypes of Patients with Primary Aldosteronism

RATIONALE AND OBJECTIVES

Adrenal venous sampling (AVS) is the primary method for differentiating between primary aldosterone (PA) subtypes. The aim of study is to develop prediction models for subtyping of patients with PA using computed tomography (CT) radiomics and clinicobiochemical characteristics associated with PA.

MATERIALS AND METHODS

This study retrospectively enrolled 158 patients with PA who underwent AVS between January 2014 and March 2021. Neural network machine learning models were developed using a two-stage analysis of triple-phase abdominal CT and clinicobiochemical characteristics. In the first stage, the models were constructed to classify unilateral or bilateral PA; in the second stage, they were designed to determine the predominant side in patients with unilateral PA. The final proposed model combined the best-performing models from both stages. The model's performance was evaluated using repeated stratified five-fold cross-validation. We employed paired t-tests to compare its performance with the conventional imaging evaluations made by radiologists, which categorize patients as either having bilateral PA or unilateral PA on one side.

RESULTS

In the first stage, the integrated model that combines CT radiomic and clinicobiochemical characteristics exhibited the highest performance, surpassing both the radiomic-alone and clinicobiochemical-alone models. It achieved an accuracy and F1 score of 80.6% ± 3.0% and 74.8% ± 5.2% (area under the receiver operating curve [AUC] = 0.778 ± 0.050). In the second stage, the accuracy and F1 score of the radiomic-based model were 88% ± 4.9% and 81.9% ± 6.2% (AUC=0.831 ± 0.087). The proposed model achieved an accuracy and F1 score of 77.5% ± 3.9% and 70.5% ± 7.1% (AUC=0.771 ± 0.046) in subtype diagnosis and lateralization, surpassing the accuracy and F1 score achieved by radiologists' evaluation (p < .05).

CONCLUSION

The proposed machine learning model can predict the subtypes and lateralization of PA. It yields superior results compared to conventional imaging evaluation and has potential to supplement the diagnostic process in PA.

Optimizing cardiovascular disease mortality prediction: a super learner approach in the tehran lipid and glucose study

BACKGROUND & AIM

Cardiovascular disease (CVD) is the most important cause of death in the world and has a potential impact on health care costs, this study aimed to evaluate the performance of machine learning survival models and determine the optimum model for predicting CVD-related mortality.

METHOD

In this study, the research population was all participants in Tehran Lipid and Glucose Study (TLGS) aged over 30 years. We used the Gradient Boosting model (GBM), Support Vector Machine (SVM), Super Learner (SL), and Cox proportional hazard (Cox-PH) models to predict the CVD-related mortality using 26 features. The dataset was randomly divided into training (80%) and testing (20%). To evaluate the performance of the methods, we used the Brier Score (BS), Prediction Error (PE), Concordance Index (C-index), and time-dependent Area Under the Curve (TD-AUC) criteria. Four different clinical models were also performed to improve the performance of the methods.

RESULTS

Out of 9258 participants with a mean age of (SD; range) 43.74 (15.51; 20-91), 56.60% were female. The CVD death proportion was 2.5% (228 participants). The death proportion was significantly higher in men (67.98% M, 32.02% F). Based on predefined selection criteria, the SL method has the best performance in predicting CVD-related mortality (TD-AUC > 93.50%). Among the machine learning (ML) methods, The SVM has the worst performance (TD-AUC = 90.13%). According to the relative effect, age, fasting blood sugar, systolic blood pressure, smoking, taking aspirin, diastolic blood pressure, Type 2 diabetes mellitus, hip circumference, body mss index (BMI), and triglyceride were identified as the most influential variables in predicting CVD-related mortality.

CONCLUSION

According to the results of our study, compared to the Cox-PH model, Machine Learning models showed promising and sometimes better performance in predicting CVD-related mortality. This finding is based on the analysis of a large and diverse urban population from Tehran, Iran.

An explainable machine learning model for prediction of high-risk nonalcoholic steatohepatitis

Early identification of high-risk metabolic dysfunction-associated steatohepatitis (MASH) can offer patients access to novel therapeutic options and potentially decrease the risk of progression to cirrhosis. This study aimed to develop an explainable machine learning model for high-risk MASH prediction and compare its performance with well-established biomarkers. Data were derived from the National Health and Nutrition Examination Surveys (NHANES) 2017-March 2020, which included a total of 5281 adults with valid elastography measurements. We used a FAST score ≥ 0.35, calculated using liver stiffness measurement and controlled attenuation parameter values and aspartate aminotransferase levels, to identify individuals with high-risk MASH. We developed an ensemble-based machine learning XGBoost model to detect high-risk MASH and explored the model's interpretability using an explainable artificial intelligence SHAP method. The prevalence of high-risk MASH was 6.9%. Our XGBoost model achieved a high level of sensitivity (0.82), specificity (0.91), accuracy (0.90), and AUC (0.95) for identifying high-risk MASH. Our model demonstrated a superior ability to predict high-risk MASH vs. FIB-4, APRI, BARD, and MASLD fibrosis scores (AUC of 0.95 vs. 0.50, 0.50, 0.49 and 0.50, respectively). To explain the high performance of our model, we found that the top 5 predictors of high-risk MASH were ALT, GGT, platelet count, waist circumference, and age. We used an explainable ML approach to develop a clinically applicable model that outperforms commonly used clinical risk indices and could increase the identification of high-risk MASH patients in resource-limited settings.

Exploiting nearest neighbor data and fuzzy membership function to address missing values in classification

The accuracy of most classification methods is significantly affected by missing values. Therefore, this study aimed to propose a data imputation method to handle missing values through the application of nearest neighbor data and fuzzy membership function as well as to compare the results with standard methods. A total of five datasets related to classification problems obtained from the UCI Machine Learning Repository were used. The results showed that the proposed method had higher accuracy than standard imputation methods. Moreover, triangular method performed better than Gaussian fuzzy membership function. This showed that the combination of nearest neighbor data and fuzzy membership function was more effective in handling missing values and improving classification accuracy.

Identifying Predictive Risk Factors for Future Cognitive Impairment Among Chinese Older Adults: Longitudinal Prediction Study

Background

The societal burden of cognitive impairment in China has prompted researchers to develop clinical prediction models aimed at making risk assessments that enable preventative interventions. However, it is unclear what types of risk factors best predict future cognitive impairment, if known risk factors make equally accurate predictions across different socioeconomic groups, and if existing prediction models are equally accurate across different subpopulations.

Objective

This paper aimed to identify which domain of health information best predicts future cognitive impairment among Chinese older adults and to examine if discrepancies exist in predictive ability across different population subsets.

Methods

Using data from the Chinese Longitudinal Healthy Longevity Survey, we quantified the ability of demographics, instrumental activities of daily living, activities of daily living, cognitive tests, social factors and hobbies, psychological factors, diet, exercise and sleep, chronic diseases, and 3 recently published logistic regression-based prediction models to predict 3-year risk of cognitive impairment in the general Chinese population and among male, female, rural-dwelling, urban-dwelling, educated, and not formally educated older adults. Predictive ability was quantified using the area under the receiver operating characteristic curve (AUC) and sensitivity-specificity curves through 20 repeats of 10-fold cross-validation.

Results

A total of 4047 participants were included in the study, of which 337 (8.3%) developed cognitive impairment 3 years after baseline data collection. The risk factor groups with the best predictive ability in the general population were demographics (AUC 0.78, 95% CI 0.77-0.78), cognitive tests (AUC 0.72, 95% CI 0.72-0.73), and instrumental activities of daily living (AUC 0.71, 95% CI 0.70-0.71). Demographics, cognitive tests, instrumental activities of daily living, and all 3 recreated prediction models had significantly higher AUCs when making predictions among female older adults compared to male older adults and among older adults with no formal education compared to those with some education.

Conclusions

This study suggests that demographics, cognitive tests, and instrumental activities of daily living are the most useful risk factors for predicting future cognitive impairment among Chinese older adults. However, the most predictive risk factors and existing models have lower predictive power among male, urban-dwelling, and educated older adults. More efforts are needed to ensure that equally accurate risk assessments can be conducted across different socioeconomic groups in China.

Prediction of cognitive decline in older breast cancer survivors: the Thinking and Living with Cancer study

PURPOSE

Cancer survivors commonly report cognitive declines after cancer therapy. Due to the complex etiology of cancer-related cognitive decline (CRCD), predicting who will be at risk of CRCD remains a clinical challenge. We developed a model to predict breast cancer survivors who would experience CRCD after systematic treatment.

METHODS

We used the Thinking and Living with Cancer study, a large ongoing multisite prospective study of older breast cancer survivors with complete assessments pre-systemic therapy, 12 months and 24 months after initiation of systemic therapy. Cognition was measured using neuropsychological testing of attention, processing speed, and executive function (APE). CRCD was defined as a 0.25 SD (of observed changes from baseline to 12 months in matched controls) decline or greater in APE score from baseline to 12 months (transient) or persistent as a decline 0.25 SD or greater sustained to 24 months. We used machine learning approaches to predict CRCD using baseline demographics, tumor characteristics and treatment, genotypes, comorbidity, and self-reported physical, psychosocial, and cognitive function.

RESULTS

Thirty-two percent of survivors had transient cognitive decline, and 41% of these women experienced persistent decline. Prediction of CRCD was good: yielding an area under the curve of 0.75 and 0.79 for transient and persistent decline, respectively. Variables most informative in predicting CRCD included apolipoprotein E4 positivity, tumor HER2 positivity, obesity, cardiovascular comorbidities, more prescription medications, and higher baseline APE score.

CONCLUSIONS

Our proof-of-concept tool demonstrates our prediction models are potentially useful to predict risk of CRCD. Future research is needed to validate this approach for predicting CRCD in routine practice settings.

Major Depressive Disorder Prediction Based on Sleep-Wake Disorders Symptoms in US Adolescents: A Machine Learning Approach from National Sleep Research Resource

Background

There is substantial evidence from previous studies that abnormalities in sleep parameters associated with depression are demonstrated in almost all stages of sleep architecture. Patients with symptoms of sleep-wake disorders have a much higher risk of developing major depressive disorders (MDD) compared to those without.

Objective

The aim of the present study is to establish and compare the performance of different machine learning models based on sleep-wake disorder symptoms data and to select the optimal model to interpret the importance of sleep-wake disorder symptoms to predict MDD occurrence in adolescents.

Methods

We derived data for this work from 2020 to 2021 Assessing Nocturnal Sleep/Wake Effects on Risk of Suicide Phase I Study from National Sleep Research Resource. Using demographic and sleep-wake disorder symptoms data as predictors and the occurrence of MDD measured base on the center for epidemiologic studies depression scale as an outcome, the following six machine learning predictive models were developed: eXtreme Gradient Boosting model (XGBoost), Light Gradient Boosting mode, AdaBoost, Gaussian Naïve Bayes, Complement Naïve Bayes, and multilayer perceptron. The models' performance was assessed using the AUC and other metrics, and the final model's predictor importance ranking was explained.

Results

XGBoost is the optimal predictive model in comprehensive performance with the AUC of 0.804 in the test set. All sleep-wake disorder symptoms were significantly positively correlated with the occurrence of adolescent MDD. The insomnia severity was the most important predictor compared with the other predictors in this study.

Conclusion

This machine learning predictive model based on sleep-wake disorder symptoms can help to raise the awareness of risk of symptoms between sleep-wake disorders and MDD in adolescents and improve primary care and prevention.

Uncovering the Hidden World of Aqueous Humor Proteins for Discovery of Biomarkers for Marfan Syndrome

Ectopia lentis is a hallmark of Marfan syndrome (MFS), a genetic connective tissue disorder affecting 1/5000 to 1/10 000 individuals worldwide. Early detection in ophthalmology clinics and timely intervention of cardiovascular complications can be lifesaving. In this study, a modified proteomics workflow with liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based data-independent acquisition (DIA) and field asymmetric ion mobility spectrometry (FAIMS) to profile the proteomes of aqueous humor (AH) and lens tissue from MFS children with ectopia lentis is utilized. Over 2300 and 2938 comparable proteins are identified in AH and the lens capsule, respectively. Functional enrichment analyses uncovered dysregulation of complement and coagulation-related pathways, collagen binding, and cell adhesion in MFS. Through weighted correlation network analysis (WGCNA) and machine learning, distinct modules associated with clinical traits are constructed and a unique biomarker panel (Q14376, Q99972, P02760, Q07507; gene names: GALE, MYOC, AMBP, DPT) is defined. These biomarkers are further validated using advanced parallel reaction monitoring (PRM) in an independent patient cohort. The results provide novel insights into the proteome characterization of ectopia lentis and offer a promising approach for developing a valuable biomarker panel to aid in the early diagnosis of Marfan syndrome via AH proteome.

Integrating exploration and prediction in computational psychotherapy science: proof of concept

Introduction

Psychotherapy research has long preferred explanatory over predictive models. As a result, psychotherapy research is currently limited in the variability that can be accounted for in the process and outcome of treatment. The present study is a proof-of-concept approach to psychotherapy science that uses a datadriven approach to achieve robust predictions of the process and outcome of treatment.

Methods

A trial including 65 therapeutic dyads was designed to enable an adequate level of variability in therapist characteristics, overcoming the common problem of restricted range. A mixed-model, data-driven approach with cross-validation machine learning algorithms was used to predict treatment outcome and alliance (within- and between-clients; client- and therapist-rated alliance).

Results and discussion

Based on baseline predictors only, the models explained 52.8% of the variance for out-of-sample prediction in treatment outcome, and 24.1-52.8% in therapeutic alliance. The identified predictors were consistent with previous findings and point to directions for future investigation. Although limited by its sample size, this study serves as proof of the great potential of the presented approach to produce robust predictions regarding the process and outcome of treatment, offering a potential solution to problems such as p-hacking and lack of replicability. Findings should be replicated using larger samples and distinct populations and settings.

Advancing precision prognostication in neuro-oncology: Machine learning models for data-driven personalized survival predictions in IDH-wildtype glioblastoma

Background

Glioblastoma (GBM) remains associated with a dismal prognoses despite standard therapies. While population-level survival statistics are established, generating individualized prognosis remains challenging. We aim to develop machine learning (ML) models that generate personalized survival predictions for GBM patients to enhance prognostication.

Methods

Adult patients with histologically confirmed IDH-wildtype GBM from the National Cancer Database (NCDB) were analyzed. ML models were developed with TabPFN, TabNet, XGBoost, LightGBM, and Random Forest algorithms to predict mortality at 6, 12, 18, and 24 months postdiagnosis. SHapley Additive exPlanations (SHAP) were employed to enhance the interpretability of the models. Models were primarily evaluated using the area under the receiver operating characteristic (AUROC) values, and the top-performing models indicated by the highest AUROCs for each outcome were deployed in a web application that was created for individualized predictions.

Results

A total of 7537 patients were retrieved from the NCDB. Performance evaluation revealed the top-performing models for each outcome were built using the TabPFN algorithm. The TabPFN models yielded mean AUROCs of 0.836, 0.78, 0.732, and 0.724 in predicting 6, 12, 18, and 24 month mortality, respectively.

Conclusions

This study establishes ML models tailored to individual patients to enhance GBM prognostication. Future work should focus on external validation and dynamic updating as new data emerge.

Therapeutic Ultrasound Alone and Associated with Lymphatic Drainage in Women with Breast Engorgement: A Clinical Trial

Introduction: Breast engorgement (BE) is a problem that affects many women, especially in the first days of breastfeeding, producing inflammatory symptoms. Nonpharmacological therapies are inexpensive, safe, and can produce symptom relief. Objective: This study aims to analyze the safety of therapeutic ultrasound regarding possible risks of overheating and the effects of its use alone and associated with lymphatic drainage (LD) in women. Material and Methods: Effectiveness is measured through thermography, visual analog scale, and six-point scale of BE. This is a nonrandomized clinical trial with a sample of 34 in the ultrasound group (G1), 28 in the ultrasound and LD group (G2), and 37 in the control group (G3). Results: The mean reduction for engorgement was 1.3 ± 0.8 to G1, 1.4 ± 1.0 to G2, and 1.2 ± 0.9 to G3 according to the six-point scale. The mean reduction for pain level was 3.6 ± 2.1 to G1, 4.0 ± 3.1 to G2, and 4.0 ± 2.2 to G3 according to the visual analogue scale. Conclusion: It was observed that all therapies were effective in reducing the level of engorgement, according to the six-point scale. However, combined ultrasound and LD therapy has been shown to be more effective in reducing the level of pain. Brazilian Registry of Clinical Trials (RBR-6btb6zz).