Cross-Validatory Choice and Assessment of Statistical Predictions

Multi-Variable Multi-Metric Optimization of Self-Assembled Photocatalytic CO2 Reduction Performance Using Machine Learning Algorithms

The sunlight-driven reduction of CO2 into fuels and platform chemicals is a promising approach to enable a circular economy. However, established optimization approaches are poorly suited to multivariable multimetric photocatalytic systems because they aim to optimize one performance metric while sacrificing the others and thereby limit overall system performance. Herein, we address this multimetric challenge by defining a metric for holistic system performance that takes multiple figures of merit into account, and employ a machine learning algorithm to efficiently guide our experiments through the large parameter matrix to make holistic optimization accessible for human experimentalists. As a test platform, we employ a five-component system that self-assembles into photocatalytic micelles for CO2-to-CO reduction, which we experimentally optimized to simultaneously improve yield, quantum yield, turnover number, and frequency while maintaining high selectivity. Leveraging the data set with machine learning algorithms allows quantification of each parameter's effect on overall system performance. The buffer concentration is unexpectedly revealed as the dominating parameter for optimal photocatalytic activity, and is nearly four times more important than the catalyst concentration. The expanded use and standardization of this methodology to define and optimize holistic performance will accelerate progress in different areas of catalysis by providing unprecedented insights into performance bottlenecks, enhancing comparability, and taking results beyond comparison of subjective figures of merit.

Deep learning classification of EEG-based BCI monitoring of the attempted arm and hand movements

OBJECTIVES

The primary objective of this research is to improve the average classification performance for specific movements in patients with cervical spinal cord injury (SCI).

METHODS

The study utilizes a low-frequency multi-class electroencephalography (EEG) dataset from Graz University of Technology. The research combines convolutional neural network (CNN) and long-short-term memory (LSTM) architectures to uncover neural correlations between temporal and spatial aspects of the EEG signals associated with attempted arm and hand movements. To achieve this, three different methods are used to select relevant features, and the proposed model's robustness against variations in the data is validated using 10-fold cross-validation (CV). The research also investigates subject-specific adaptation in an online paradigm, extending movement classification proof-of-concept.

RESULTS

The combined CNN-LSTM model, enhanced by three feature selection methods, demonstrates robustness with a mean accuracy of 75.75 % and low standard deviation (+/- 0.74 %) in 10-fold cross-validation, confirming its reliability.

CONCLUSIONS

In summary, this research aims to make valuable contributions to the field of neuro-technology by developing EEG-controlled assistive devices using a generalized brain-computer interface (BCI) and deep learning (DL) framework. The focus is on capturing high-level spatiotemporal features and latent dependencies to enhance the performance and usability of EEG-based assistive technologies.

Prognostic subgroups of chronic pain patients using latent variable mixture modeling within a supervised machine learning framework

The present study combined a supervised machine learning framework with an unsupervised method, finite mixture modeling, to identify prognostically meaningful subgroups of diverse chronic pain patients undergoing interdisciplinary treatment. Questionnaire data collected at pre-treatment and 1-year follow up from 11,995 patients from the Swedish Quality Registry for Pain Rehabilitation were used. Indicators measuring pain characteristics, psychological aspects, and social functioning and general health status were used to form subgroups, and pain interference at follow-up was used for the selection and the performance evaluation of models. A nested cross-validation procedure was used for determining the number of classes (inner cross-validation) and the prediction accuracy of the selected model among unseen cases (outer cross-validation). A four-class solution was identified as the optimal model. Identified subgroups were separable on indicators, predictive of long-term outcomes, and related to background characteristics. Results are discussed in relation to previous clustering attempts of patients with diverse chronic pain conditions. Our analytical approach, as the first to combine mixture modeling with supervised, targeted learning, provides a promising framework that can be further extended and optimized for improving accurate prognosis in pain treatment and identifying clinically meaningful subgroups among chronic pain patients.

MicroRNA classification and discovery for major depressive disorder diagnosis: Towards a robust and interpretable machine learning approach

BACKGROUND

Major depressive disorder (MDD) is notably underdiagnosed and undertreated due to its complex nature and subjective diagnostic methods. Biomarker identification would help provide a clearer understanding of MDD aetiology. Although machine learning (ML) has been implemented in previous studies to study the alteration of microRNA (miRNA) levels in MDD cases, clinical translation has not been feasible due to the lack of interpretability (i.e. too many miRNAs for consideration) and stability.

METHODS

This study applied logistic regression (LR) model to the blood miRNA expression profile to differentiate patients with MDD (n = 60) from healthy controls (HCs, n = 60). Embedded (L1-regularised logistic regression) feature selector was utilised to extract clinically relevant miRNAs, and optimized for clinical application.

RESULTS

Patients with MDD could be differentiated from HCs with the area under the receiver operating characteristic curve (AUC) of 0.81 on testing data when all available miRNAs were considered (which served as a benchmark). Our LR model selected miRNAs up to 5 (known as LR-5 model) emerged as the best model because it achieved a moderate classification ability (AUC = 0.75), relatively high interpretability (feature number = 5) and stability (ϕ̂Z=0.55) compared to the benchmark. The top-ranking miRNAs identified by our model have demonstrated associations with MDD pathways involving cytokine signalling in the immune system, the reelin signalling pathway, programmed cell death and cellular responses to stress.

CONCLUSION

The LR-5 model, which is optimised based on ML design factors, may lead to a robust and clinically usable MDD diagnostic tool.

Utility of Single Items within the Suicidal Behaviors Questionnaire-Revised (SBQ-R): A Bayesian Network Approach and Relative Importance Analysis

The Suicidal Behaviors Questionnaire-Revised (SBQ-R) comprises four content-specific items widely used to assess the history of suicide-related thoughts, plans or attempts, frequency of suicidal ideation, communication of intent to die by suicide and self-reported likelihood of a suicide attempt. Each item focuses on a specific parameter of the suicide-related thoughts and behaviors construct. Past research has primarily focused on the total score. This study used Bayesian network modeling and relative importance analyses on SBQ-R data from 1160 U.S. and 1141 Chinese undergraduate students. The Bayesian network analysis results showed that Item 1 is suitable for identifying other parameters of the suicide-related thoughts and behaviors construct. The results of the relative importance analysis further highlighted the relevancy of each SBQ-R item score when examining evidence for suicide-related thoughts and behaviors. These findings provided empirical support for using the SBQ-R item scores to understand the performances of different suicide-related behavior parameters. Further, they demonstrated the potential value of examining individual item-level responses to offer clinically meaningful insights. To conclude, the SBQ-R allows for the evaluation of each critical suicide-related thought and behavior parameter and the overall suicide risk.

A non-linear partial least squares based on monotonic inner relation

A novel regression model, monotonic inner relation-based non-linear partial least squares (MIR-PLS), is proposed to address complex issues like limited observations, multicollinearity, and nonlinearity in Chinese Medicine (CM) dose-effect relationship experimental data. MIR-PLS uses a piecewise mapping function based on monotonic cubic splines to model the non-linear inner relations between input and output score vectors. Additionally, a new weight updating strategy (WUS) is developed by leveraging the properties of monotonic functions. The proposed MIR-PLS method was compared with five well-known PLS variants: standard PLS, quadratic PLS (QPLS), error-based QPLS (EB-QPLS), neural network PLS (NNPLS), and spline PLS (SPL-PLS), using CM dose-effect relationship datasets and near-infrared (NIR) spectroscopy datasets. Experimental results demonstrate that MIR-PLS exhibits general applicability, achieving excellent predictive performances in the presence or absence of significant non-linear relationships. Furthermore, the model is not limited to CM dose-effect relationship research and can be applied to other regression tasks.

Deep neural network uncertainty estimation for early oral cancer diagnosis

BACKGROUND

Early diagnosis in oral cancer is essential to reduce both morbidity and mortality. This study explores the use of uncertainty estimation in deep learning for early oral cancer diagnosis.

METHODS

We develop a Bayesian deep learning model termed 'Probabilistic HRNet', which utilizes the ensemble MC dropout method on HRNet. Additionally, two oral lesion datasets with distinct distributions are created. We conduct a retrospective study to assess the predictive performance and uncertainty of Probabilistic HRNet across these datasets.

RESULTS

Probabilistic HRNet performs optimally on the In-domain test set, achieving an F1 score of 95.3% and an AUC of 96.9% by excluding the top 30% high-uncertainty samples. For evaluations on the Domain-shift test set, the results show an F1 score of 64.9% and an AUC of 80.3%. After excluding 30% of the high-uncertainty samples, these metrics improve to an F1 score of 74.4% and an AUC of 85.6%.

CONCLUSION

Redirecting samples with high uncertainty to experts for subsequent diagnosis significantly decreases the rates of misdiagnosis, which highlights that uncertainty estimation is vital to ensure safe decision making for computer-aided early oral cancer diagnosis.

Bayesian learners in gradient boosting for linear mixed models

Selection of relevant fixed and random effects without prior choices made from possibly insufficient theory is important in mixed models. Inference with current boosting techniques suffers from biased estimates of random effects and the inflexibility of random effects selection. This paper proposes a new inference method "BayesBoost" that integrates a Bayesian learner into gradient boosting with simultaneous estimation and selection of fixed and random effects in linear mixed models. The method introduces a novel selection strategy for random effects, which allows for computationally fast selection of random slopes even in high-dimensional data structures. Additionally, the new method not only overcomes the shortcomings of Bayesian inference in giving precise and unambiguous guidelines for the selection of covariates by benefiting from boosting techniques, but also provides Bayesian ways to construct estimators for the precision of parameters such as variance components or credible intervals, which are not available in conventional boosting frameworks. The effectiveness of the new approach can be observed via simulation and in a real-world application.

Status Consumption as Coping With Fear of Death: The Mediating Role of Death Avoidance and the Moderating Role of Materialism

Individuals employ various coping mechanisms to deal with the fear of death. While materialism and status consumption are commonly recognized in the literature as such strategies, no study has yet empirically tested this premise. Accordingly, this study examined the mediating role of death avoidance in the link between the fear of death and death-related status consumption (DRSC). Data obtained from 346 participants were analyzed using structural equation modeling. The results showed that fear of death significantly and positively influences DRSC and that death avoidance partially and positively mediates this relationship. Results also revealed that materialism strengthens the relationship between fear of death and DRSC, while it does not significantly moderate the relationship between death avoidance and DRSC. These results support the conclusion that death-related status consumption may play a critical role as an avoidance mechanism in coping with the fear of death. This study, being among the few that investigate death-related consumer behaviors, enriches both terror management theory and the literature on consumer behavior in crises.

Practical approaches in evaluating validation and biases of machine learning applied to mobile health studies

BACKGROUND

Machine learning (ML) models are evaluated in a test set to estimate model performance after deployment. The design of the test set is therefore of importance because if the data distribution after deployment differs too much, the model performance decreases. At the same time, the data often contains undetected groups. For example, multiple assessments from one user may constitute a group, which is usually the case in mHealth scenarios.

METHODS

In this work, we evaluate a model's performance using several cross-validation train-test-split approaches, in some cases deliberately ignoring the groups. By sorting the groups (in our case: Users) by time, we additionally simulate a concept drift scenario for better external validity. For this evaluation, we use 7 longitudinal mHealth datasets, all containing Ecological Momentary Assessments (EMA). Further, we compared the model performance with baseline heuristics, questioning the essential utility of a complex ML model.

RESULTS

Hidden groups in the dataset leads to overestimation of ML performance after deployment. For prediction, a user's last completed questionnaire is a reasonable heuristic for the next response, and potentially outperforms a complex ML model. Because we included 7 studies, low variance appears to be a more fundamental phenomenon of mHealth datasets.

CONCLUSIONS

The way mHealth-based data are generated by EMA leads to questions of user and assessment level and appropriate validation of ML models. Our analysis shows that further research needs to follow to obtain robust ML models. In addition, simple heuristics can be considered as an alternative for ML. Domain experts should be consulted to find potentially hidden groups in the data.

What are the roles of positive psychological construct in blended learning contexts? Integrating academic buoyancy into the Community of Inquiry framework

Introduction

In the post-epidemic era, blended learning has become a social trend for the future of higher education, and scholars have endeavored to understand the factors that influence student learning in these blended communities. Communities of Inquiry is a conceptual framework that describes the components of blended learning environments, indicating teaching presence, social presence, and cognitive presence. However, the framework fails to adequately explore how individual learning motivational factors influence student learning. Therefore, this study extends the Community of Inquiry framework by drawing on a positive psychological construct-academic buoyancy to reveal the relationship between academic buoyancy and the three presences through empirical research.

Methods

The theoretical model was validated by SPSS 26.0 and smartPLS4.0. To evaluate the measurement and structural models, structural equation modeling (SEM) was carried out using the partial least squares (PLS) method.

Findings

(a) Teaching presence positively predicts academic buoyancy, and academic buoyancy positively predicts social presence and cognitive presence; (b) academic buoyancy mediates teaching presence and social presence, as well as teaching presence and cognitive presence; and (c) academic buoyancy acts as a chain mediator between teaching presence and cognitive presence through social presence.

Discussion

The results of this study fill a gap in the multiple roles of individual positive psychological construct-academic buoyancy in blended learning communities, extend the Community of Inquiry theoretical framework, and provide empirical evidence for blended learning quality and practical improvement strategies.

Predicting anxiety, depression, and insomnia among Bangladeshi university students using tree-based machine learning models

Background and Aims

Mental health problem is a rising public health concern. People of all ages, specially Bangladeshi university students, are more affected by this burden. Thus, the objective of the study was to use tree-based machine learning (ML) models to identify major risk factors and predict anxiety, depression, and insomnia in university students.

Methods

A social media-based cross-sectional survey was employed for data collection. We used Generalized Anxiety Disorder (GAD-7), Patient Health Questionnaire (PHQ-9) and Insomnia Severity Index (ISI-7) scale for measuring students' anxiety, depression and insomnia problems. The tree-based supervised decision tree (DT), random forest (RF) and robust eXtreme Gradient Boosting (XGBoost) ML algorithms were used to build the prediction models and their predictive performance was evaluated using confusion matrix and receiver operating characteristic (ROC) curves.

Results

Of the 1250 students surveyed, 64.7% were male and 35.3% were female. The students' ages ranged from 18 to 26 years old, with an average age of 22.24 years (SD = 1.30). Majority of the students (72.6%) were from rural areas and social media addicted (56.6%). Almost 83.3% of the students had moderate to severe anxiety, 84.7% had moderate to severe depression and 76.5% had moderate to severe insomnia problems. Students' social media addiction, age, academic performance, smoking status, monthly family income and morningness-eveningness are the main risk factors of anxiety, depression and insomnia. The highest predictive performance was observed from the XGBoost model for anxiety, depression and insomnia.

Conclusion

The study findings offer valuable insights for stakeholders, families and policymakers enabling a more profound comprehension of the pressing mental health disorders. This understanding can guide the formulation of improved policy strategies, initiatives for mental health promotion, and the development of effective counseling services within university campus. Additionally, our proposed model might play a critical role in diagnosing and predicting mental health problems among Bangladeshi university students and similar settings.

Magnetic resonance imaging radiomic features stability in brain metastases: Impact of image preprocessing, image-, and feature-level harmonization

Background and purpose

Magnetic resonance imaging (MRI) scans are highly sensitive to acquisition and reconstruction parameters which affect feature stability and model generalizability in radiomic research. This work aims to investigate the effect of image pre-processing and harmonization methods on the stability of brain MRI radiomic features and the prediction performance of radiomic models in patients with brain metastases (BMs).

Materials and methods

Two T1 contrast enhanced brain MRI data-sets were used in this study. The first contained 25 BMs patients with scans at two different time points and was used for features stability analysis. The effect of gray level discretization (GLD), intensity normalization (Z-score, Nyul, WhiteStripe, and in house-developed method named N-Peaks), and ComBat harmonization on features stability was investigated and features with intraclass correlation coefficient >0.8 were considered as stable. The second data-set containing 64 BMs patients was used for a classification task to investigate the informativeness of stable features and the effects of harmonization methods on radiomic model performance.

Results

Applying fixed bin number (FBN) GLD, resulted in higher number of stable features compare to fixed bin size (FBS) discretization (10 ± 5.5 % higher). `Harmonization in feature domain improved the stability for non-normalized and normalized images with Z-score and WhiteStripe methods. For the classification task, keeping the stable features resulted in good performance only for normalized images with N-Peaks along with FBS discretization.

Conclusions

To develop a robust MRI based radiomic model we recommend using an intensity normalization method based on a reference tissue (e.g N-Peaks) and then using FBS discretization.

Advanced detection of coronary artery disease via deep learning analysis of plasma cytokine data

The 2017 World Health Organization Fact Sheet highlights that coronary artery disease is the leading cause of death globally, responsible for approximately 30% of all deaths. In this context, machine learning (ML) technology is crucial in identifying coronary artery disease, thereby saving lives. ML algorithms can potentially analyze complex patterns and correlations within medical data, enabling early detection and accurate diagnosis of CAD. By leveraging ML technology, healthcare professionals can make informed decisions and implement timely interventions, ultimately leading to improved outcomes and potentially reducing the mortality rate associated with coronary artery disease. Machine learning algorithms create non-invasive, quick, accurate, and economical diagnoses. As a result, machine learning algorithms can be employed to supplement existing approaches or as a forerunner to them. This study shows how to use the CNN classifier and RNN based on the LSTM classifier in deep learning to attain targeted "risk" CAD categorization utilizing an evolving set of 450 cytokine biomarkers that could be used as suggestive solid predictive variables for treatment. The two used classifiers are based on these "45" different cytokine prediction characteristics. The best Area Under the Receiver Operating Characteristic curve (AUROC) score achieved is (0.98) for a confidence interval (CI) of 95; the classifier RNN-LSTM used "450" cytokine biomarkers had a great (AUROC) score of 0.99 with a confidence interval of 0.95 the percentage 95, the CNN model containing cytokines received the second best AUROC score (0.92). The RNN-LSTM classifier considerably beats the CNN classifier regarding AUROC scores, as evidenced by a p-value smaller than 7.48 obtained via an independent t-test. As large-scale initiatives to achieve early, rapid, reliable, inexpensive, and accessible individual identification of CAD risk gain traction, robust machine learning algorithms can now augment older methods such as angiography. Incorporating 65 new sensitive cytokine biomarkers can increase early detection even more. Investigating the novel involvement of cytokines in CAD could lead to better risk detection, disease mechanism discovery, and new therapy options.

Long-Term Risk of Arterial Thrombosis After Intracerebral Hemorrhage: MUCH-Italy

BACKGROUND

The identification of patients surviving an acute intracerebral hemorrhage who are at a long-term risk of arterial thrombosis is a poorly defined, crucial issue for clinicians.

METHODS

In the setting of the MUCH-Italy (Multicenter Study on Cerebral Haemorrhage in Italy) prospective observational cohort, we enrolled and followed up consecutive 30-day intracerebral hemorrhage survivors to assess the long-term incidence of arterial thrombotic events, to assess the impact of clinical and radiological variables on the risk of these events, and to develop a tool for estimating such a risk at the individual level. Primary end point was a composite of ischemic stroke, myocardial infarction, or other arterial thrombotic events. A point-scoring system was generated by the β-coefficients of the variables independently associated with the long-term risk of arterial thrombosis, and the predictive MUCH score was calculated as the sum of the weighted scores.

RESULTS

Overall, 1729 patients (median follow-up time, 43 months [25th to 75th percentile, 69.0]) qualified for inclusion. Arterial thrombotic events occurred in 169 (9.7%) patients. Male sex, diabetes, hypercholesterolemia, atrial fibrillation, and personal history of coronary artery disease were associated with increased long-term risk of arterial thrombosis, whereas the use of statins and antithrombotic medications after the acute intracerebral hemorrhage was associated with a reduced risk. The area under the receiver operating characteristic curve of the MUCH score predictive validity was 0.716 (95% CI, 0.56-0.81) for the 0- to 1-year score, 0.672 (95% CI, 0.58-0.73) for the 0- to 5-year score, and 0.744 (95% CI, 0.65-0.81) for the 0- to 10-year score. C statistic for the prediction of events that occur from 0 to 10 years was 0.69 (95% CI, 0.64-0.74).

CONCLUSIONS

Intracerebral hemorrhage survivors are at high long-term risk of arterial thrombosis. The MUCH score may serve as a simple tool for risk estimation.

Collectivism, face concern and Chinese-style lurking among university students: the moderating role of trait mindfulness

Introduction

This study focuses on understanding the unique causes and mechanisms of "Chinese-style lurking" on WeChat among university students, within a cultural context that emphasizes collectivism and face concern. The research also looks into the moderating role of trait mindfulness.

Methods

For the confirmation of these phenomena and to validate the theories, a structural equation model was constructed using the Stress-Strain-Outcome (SSO) theory and mindfulness buffering theory. The model was then tested and validated with data from 1,453 valid online surveys. These data were analyzed using the SmartPLS 4.0 software.

Results

The results indicate that collectivism increases face concern, which in turn escalates online social anxiety. Face concern completely mediates between collectivism and online social anxiety, creating a serial mediation effect between face concern, online social anxiety, and lurking behavior. Additionally, trait mindfulness was found to negatively modulate the pathways from collectivism to face concern and from online social anxiety to lurking.

Discussion

The findings underscore the influence of traditional Chinese culture on contemporary students' online behavior and provide a new perspective for understanding social media lurking in an Eastern context. The results suggest that a mindfulness-based approach could be used to mitigate the associated silence and anxiety.

Strengths and limitations of non-disclosive data analysis: a comparison of breast cancer survival classifiers using VisualSHIELD

Preserving data privacy is an important concern in the research use of patient data. The DataSHIELD suite enables privacy-aware advanced statistical analysis in a federated setting. Despite its many applications, it has a few open practical issues: the complexity of hosting a federated infrastructure, the performance penalty imposed by the privacy-preserving constraints, and the ease of use by non-technical users. In this work, we describe a case study in which we review different breast cancer classifiers and report our findings about the limits and advantages of such non-disclosive suite of tools in a realistic setting. Five independent gene expression datasets of breast cancer survival were downloaded from Gene Expression Omnibus (GEO) and pooled together through the federated infrastructure. Three previously published and two newly proposed 5-year cancer-free survival risk score classifiers were trained in a federated environment, and an additional reference classifier was trained with unconstrained data access. The performance of these six classifiers was systematically evaluated, and the results show that i) the published classifiers do not generalize well when applied to patient cohorts that differ from those used to develop them; ii) among the methods we tried, the classification using logistic regression worked better on average, closely followed by random forest; iii) the unconstrained version of the logistic regression classifier outperformed the federated version by 4% on average. Reproducibility of our experiments is ensured through the use of VisualSHIELD, an open-source tool that augments DataSHIELD with new functions, a standardized deployment procedure, and a simple graphical user interface.

Psychophysical measurement of perceived motion flow of naturalistic scenes

The neural and computational mechanisms underlying visual motion perception have been extensively investigated over several decades, but little attempt has been made to measure and analyze, how human observers perceive the map of motion vectors, or optical flow, in complex naturalistic scenes. Here, we developed a psychophysical method to assess human-perceived motion flows using local vector matching and a flash probe. The estimated perceived flow for naturalistic movies agreed with the physically correct flow (ground truth) at many points, but also showed consistent deviations from the ground truth (flow illusions) at other points. Comparisons with the predictions of various computational models, including cutting-edge computer vision algorithms and coordinate transformation models, indicated that some flow illusions are attributable to lower-level factors such as spatiotemporal pooling and signal loss, while others reflect higher-level computations, including vector decomposition. Our study demonstrates a promising data-driven psychophysical paradigm for an advanced understanding of visual motion perception.

Small groups in multidimensional feature space: Two examples of supervised two-group classification from biomedicine

Some biomedical datasets contain a small number of samples which have large numbers of features. This can make analysis challenging and prone to errors such as overfitting and misinterpretation. To improve the accuracy and reliability of analysis in such cases, we present a tutorial that demonstrates a mathematical approach for a supervised two-group classification problem using two medical datasets. A tutorial provides insights on effectively addressing uncertainties and handling missing values without the need for removing or inputting additional data. We describe a method that considers the size and shape of feature distributions, as well as the pairwise relations between measured features as separate derived features and prognostic factors. Additionally, we explain how to perform similarity calculations that account for the variation in feature values within groups and inaccuracies in individual value measurements. By following these steps, a more accurate and reliable analysis can be achieved when working with biomedical datasets that have a small sample size and multiple features.

Weakly supervised deep learning for diagnosis of multiple vertebral compression fractures in CT

OBJECTIVE

This study aims to develop a weakly supervised deep learning (DL) model for vertebral-level vertebral compression fracture (VCF) classification using image-level labelled data.

METHODS

The training set included 815 patients with normal (n = 507, 62%) or VCFs (n = 308, 38%). Our proposed model was trained on image-level labelled data for vertebral-level classification. Another supervised DL model was trained with vertebral-level labelled data to compare the performance of the proposed model.

RESULTS

The test set included 227 patients with normal (n = 117, 52%) or VCFs (n = 110, 48%). For a fair comparison of the two models, we compared sensitivities with the same specificities of the proposed model and the vertebral-level supervised model. The specificity for overall L1-L5 performance was 0.981. The proposed model may outperform the vertebral-level supervised model with sensitivities of 0.770 vs 0.705 (p = 0.080), respectively. For vertebral-level analysis, the specificities for each L1-L5 were 0.974, 0.973, 0.970, 0.991, and 0.995, respectively. The proposed model yielded the same or better sensitivity than the vertebral-level supervised model in L1 (0.750 vs 0.694, p = 0.480), L3 (0.793 vs 0.586, p < 0.05), L4 (0.833 vs 0.667, p = 0.480), and L5 (0.600 vs 0.600, p = 1.000), respectively. The proposed model showed lower sensitivity than the vertebral-level supervised model for L2, but there was no significant difference (0.775 vs 0.825, p = 0.617).

CONCLUSIONS

The proposed model may have a comparable or better performance than the supervised model in vertebral-level VCF classification.

CLINICAL RELEVANCE STATEMENT

Vertebral-level vertebral compression fracture classification aids in devising patient-specific treatment plans by identifying the precise vertebrae affected by compression fractures.

KEY POINTS

• Our proposed weakly supervised method may have comparable or better performance than the supervised method for vertebral-level vertebral compression fracture classification. • The weakly supervised model could have classified cases with multiple vertebral compression fractures at the vertebral-level, even if the model was trained with image-level labels. • Our proposed method could help reduce radiologists' labour because it enables vertebral-level classification from image-level labels.

Exploring the Spatial Pattern of Urban Forest Ecosystem Services based on i-Tree Eco and Spatial Interpolation: A Case Study of Kyoto City, Japan

Urban forest, as an essential urban green infrastructure, is critical in providing ecosystem services to cities. To enhance the mainstreaming of ecosystem services in urban planning, it is necessary to explore the spatial pattern of urban forest ecosystem services in cities. This study provides a workflow for urban forest planning based on field investigation, i-Tree Eco, and geostatistical interpolation. Firstly, trees across an array of land use types were investigated using a sampling method. Then i-Tree Eco was applied to quantify ecosystem services and ecosystem service value in each plot. Based on the ecosystem services estimates for plots, four interpolation methods were applied and compared by cross-validation. The Empirical Bayesian Kriging was determined as the best interpolation method with higher prediction accuracy. With the results of Empirical Bayesian Kriging, this study compared urban forest ecosystem services and ecosystem service value across land use types. The spatial correlations between ecosystem service value and four types of point of interest in urban places were explored using the bivariate Moran's I statistic and the bivariate local indicators of spatial association. Our results show that the residential area in the built-up area of Kyoto city had higher species richness, tree density, ecosystem services, and total ecosystem service value. Positive spatial correlations were found between ecosystem service value and the distribution of urban space types including the tourist attraction distribution, urban park distribution, and school distribution. This study provides a specific ecosystem service-oriented reference for urban forest planning based on land use and urban space types.

PortPred: Exploiting deep learning embeddings of amino acid sequences for the identification of transporter proteins and their substrates

The physiology of every living cell is regulated at some level by transporter proteins which constitute a relevant portion of membrane-bound proteins and are involved in the movement of ions, small and macromolecules across bio-membranes. The importance of transporter proteins is unquestionable. The prediction and study of previously unknown transporters can lead to the discovery of new biological pathways, drugs and treatments. Here we present PortPred, a tool to accurately identify transporter proteins and their substrate starting from the protein amino acid sequence. PortPred successfully combines pre-trained deep learning-based protein embeddings and machine learning classification approaches and outperforms other state-of-the-art methods. In addition, we present a comparison of the most promising protein sequence embeddings (Unirep, SeqVec, ProteinBERT, ESM-1b) and their performances for this specific task.

Simulation and optimization of nutrient uptake and biomass formation using a multi-parameter Monod-type model of tobacco BY-2 cell suspension cultures in a stirred-tank bioreactor

Introduction

Tobacco (Nicotiana tabacum) cv Bright Yellow-2 (BY-2) cell suspension cultures enable the rapid production of complex protein-based biopharmaceuticals but currently achieve low volumetric productivity due to slow biomass formation. The biomass yield can be improved with tailored media, which can be designed either by laborious trial-and-error experiments or systematic, rational design using mechanistic models, linking nutrient consumption and biomass formation.

Methods

Here we developed an iterative experiment-modeling-optimization workflow to gradually refine such a model and its predictions, based on collected data concerning BY-2 cell macronutrient consumption (sucrose, ammonium, nitrate and phosphate) and biomass formation.

Results and discussion

The biomass formation was well predicted by an unstructured segregated mechanistic Monod-type model as long as the nutrient concentrations did not approach zero (we omitted phosphate, which was completely depleted). Multi-criteria optimization for sucrose and biomass formation indicated the best tradeoff (in a Paretian sense) between maximum biomass yield and minimum process time by reducing the initial sucrose concentration, whereas the inoculation biomass could be increased to maximize the biomass yield or minimize the process time, which we confirmed in calibration experiments. The model became inaccurate at biomass densities > 8 g L-1 dry mass when sucrose was almost depleted. We compensated for this limitation by including glucose and fructose as sucrose hydrolysis products in the model. The remaining offset between the simulation and experimental data might be resolved by including intracellular pools of sucrose, ammonium, nitrate and phosphate. Overall, we demonstrated that iterative models can be used to systematically optimize conditions for bioreactor-based processes.

Validation of a Smartphone Pupillometry Application in Diagnosing Severe Traumatic Brain Injury

The pupillary light reflex (PLR) is an important biomarker for the detection and management of traumatic brain injury (TBI). We investigated the performance of PupilScreen, a smartphone-based pupillometry app, in classifying healthy control subjects and subjects with severe TBI in comparison to the current gold standard NeurOptics pupillometer (NPi-200 model with proprietary Neurological Pupil Index [NPi] TBI severity score). A total of 230 PLR video recordings taken using both the PupilScreen smartphone pupillometer and NeurOptics handheld device (NPi-200) pupillometer were collected from 33 subjects with severe TBI (sTBI) and 132 subjects who were healthy without self-reported neurological disease. Severe TBI status was determined by Glasgow Coma Scale (GCS) at the time of recording. The proprietary NPi score was collected from the NPi-200 pupillometer for each subject. Seven PLR curve morphological parameters were collected from the PupilScreen app for each subject. A comparison via t-test and via binary classification algorithm performance using NPi scores from the NPi-200 and PLR parameter data from the PupilScreen app was completed. This was used to determine how the frequently used NPi-200 proprietary NPi TBI severity score compares to the PupilScreen app in ability to distinguish between healthy and sTBI subjects. Binary classification models for this task were trained for the diagnosis of healthy or severe TBI using logistic regression, k-nearest neighbors, support vector machine, and random forest machine learning classification models. Overall classification accuracy, sensitivity, specificity, area under the curve, and F1 score values were calculated. Median GCS was 15 for the healthy cohort and 6 (interquartile range 2) for the severe TBI cohort. Smartphone app PLR parameters as well as NPi from the digital infrared pupillometer were significantly different between healthy and severe TBI cohorts; 33% of the study cohort had dark eye colors defined as brown eyes of varying shades. Across all classification models, the top performing PLR parameter combination for classifying subjects as healthy or sTBI for PupilScreen was maximum diameter, constriction velocity, maximum constriction velocity, and dilation velocity with accuracy, sensitivity, specificity, area under the curve (AUC), and F1 score of 87%, 85.9%, 88%, 0.869, and 0.85, respectively, in a random forest model. The proprietary NPi TBI severity score demonstrated greatest AUC value, F1 score, and sensitivity of 0.648, 0.567, and 50.9% respectively using a random forest classifier and greatest overall accuracy and specificity of 67.4% and 92.4% using a logistic regression model in the same classification task on the same dataset. The PupilScreen smartphone pupillometry app demonstrated binary healthy versus severe TBI classification ability greater than that of the NPi-200 proprietary NPi TBI severity score. These results may indicate the potential benefit of future study of this PupilScreen smartphone pupillometry application in comparison to the NPi-200 digital infrared pupillometer across the broader TBI spectrum, as well as in other neurological diseases.

Modeling the significance of green orientation and culture on green innovation performance: moderating effect of firm size and green implementation

The current global trend in sustainable business practices is to optimize green innovation performance. To protect the environment and maintain their own survival, organizations must strengthen their green innovation capabilities. Drawing on the recourse-based view and ecology modernization theory (EMT), this study examines the direct effect of green strategic orientations, green entrepreneurial orientation, green market orientation, green innovation orientation, and green organizational culture on the firm's green innovation capability, as well as the mediating effect of green innovation capability on the relationship of these four factors and green innovation performance. Besides, this study also explored the moderating effects of green management system implementation and firm size on the association between green innovation capability and green innovation performance. To test the hypothesized model, a questionnaire survey was administered to gather responses from 293 medium-sized and large manufacturing firms operating in Pakistan. The partial least squares method was used for data analysis. The results revealed that green entrepreneurial orientation, green market orientation, green innovation orientation, and green organizational culture positively impacted green innovation capability, which subsequently positively influenced green innovation performance. Moreover, effective implementation of green management systems can bolster the effect of green innovation capability on green innovation performance, and the mediating effect of green innovation capability has also been confirmed. These indicated that the management of medium and large manufacturing firms operating in Pakistan should focus on encouraging green innovation and training employees regarding the latest eco-friendly technologies to attain performance and sustainable development goals. Policymakers should implement green business development programs and offer rewards or penalties for promoting compliance. The present study contributes greatly to the literature by applying EMT as an alternative to address the mediating role of green innovation capability and the moderating effect of green management system implementation in enhancing firms' green innovation performance.

Sparse Bayesian Learning With Weakly Informative Hyperprior and Extended Predictive Information Criterion

This article considers the regression problem with sparse Bayesian learning (SBL) when the number of weights P is larger than the data size N , i.e., P >> N . The situation induces overfitting and makes regression tasks, such as prediction and basis selection, challenging. We show a strategy to address this problem. Our strategy consists of two steps. The first is to apply an inverse gamma hyperprior with a shape parameter close to zero over the noise precision of automatic relevance determination (ARD) prior. This hyperprior is associated with the concept of a weakly informative prior in terms of enhancing sparsity. The model sparsity can be controlled by adjusting a scale parameter of inverse gamma hyperprior, leading to the prevention of overfitting. The second is to select an optimal scale parameter. We develop an extended predictive information criterion (EPIC) for optimal selection. We investigate the strategy through relevance vector machine (RVM) with a multiple-kernel scheme dealing with highly nonlinear data, including smooth and less smooth regions. This setting is one form of the regression task with SBL in the P >> N situation. As an empirical evaluation, regression analyses on four artificial datasets and eight real datasets are performed. We see that the overfitting is prevented, while predictive performance may be not drastically superior to comparative methods. Our methods allow us to select a small number of nonzero weights while keeping the model sparse. Thus, the methods are expected to be useful for basis and variable selection.

3D Finite Element Models Reconstructed From 2D Dual-Energy X-Ray Absorptiometry (DXA) Images Improve Hip Fracture Prediction Compared to Areal BMD in Osteoporotic Fractures in Men (MrOS) Sweden Cohort

Bone strength is an important contributor to fracture risk. Areal bone mineral density (aBMD) derived from dual-energy X-ray absorptiometry (DXA) is used as a surrogate for bone strength in fracture risk prediction tools. 3D finite element (FE) models predict bone strength better than aBMD, but their clinical use is limited by the need for 3D computed tomography and lack of automation. We have earlier developed a method to reconstruct the 3D hip anatomy from a 2D DXA image, followed by subject-specific FE-based prediction of proximal femoral strength. In the current study, we aim to evaluate the method's ability to predict incident hip fractures in a population-based cohort (Osteoporotic Fractures in Men [MrOS] Sweden). We defined two subcohorts: (i) hip fracture cases and controls cohort: 120 men with a hip fracture (<10 years from baseline) and two controls to each hip fracture case, matched by age, height, and body mass index; and (ii) fallers cohort: 86 men who had fallen the year before their hip DXA scan was acquired, 15 of which sustained a hip fracture during the following 10 years. For each participant, we reconstructed the 3D hip anatomy and predicted proximal femoral strength in 10 sideways fall configurations using FE analysis. The FE-predicted proximal femoral strength was a better predictor of incident hip fractures than aBMD for both hip fracture cases and controls (difference in area under the receiver operating characteristics curve, ΔAUROC = 0.06) and fallers (ΔAUROC = 0.22) cohorts. This is the first time that FE models outperformed aBMD in predicting incident hip fractures in a population-based prospectively followed cohort based on 3D FE models obtained from a 2D DXA scan. Our approach has potential to notably improve the accuracy of fracture risk predictions in a clinically feasible manner (only one single DXA image is needed) and without additional costs compared to the current clinical approach. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

HPOSS: A hierarchical portfolio optimization stacking strategy to reduce the generalization error of ensembles of models

Surrogate models are frequently used to replace costly engineering simulations. A single surrogate is frequently chosen based on previous experience or by fitting multiple surrogates and selecting one based on mean cross-validation errors. A novel stacking strategy will be presented in this paper. This new strategy results from reinterpreting the model selection process based on the generalization error. For the first time, this problem is proposed to be translated into a well-studied financial problem: portfolio management and optimization. In short, it is demonstrated that the individual residues calculated by leave-one-out procedures are samples from a given random variable ϵi, whose second non-central moment is the i-th model's generalization error. Thus, a stacking methodology based solely on evaluating the behavior of the linear combination of the random variables ϵi is proposed. At first, several surrogate models are calibrated. The Directed Bubble Hierarchical Tree (DBHT) clustering algorithm is then used to determine which models are worth stacking. The stacking weights can be calculated using any financial approach to the portfolio optimization problem. This alternative understanding of the problem enables practitioners to use established financial methodologies to calculate the models' weights, significantly improving the ensemble of models' out-of-sample performance. A study case is carried out to demonstrate the applicability of the new methodology. Overall, a total of 124 models were trained using a specific dataset: 40 Machine Learning models and 84 Polynomial Chaos Expansion models (which considered 3 types of base random variables, 7 least square algorithms for fitting the up to fourth order expansion's coefficients). Among those, 99 models could be fitted without convergence and other numerical issues. The DBHT algorithm with Pearson correlation distance and generalization error similarity was able to select a subgroup of 23 models from the 99 fitted ones, implying a reduction of about 77% in the total number of models, representing a good filtering scheme which still preserves diversity. Finally, it has been demonstrated that the weights obtained by building a Hierarchical Risk Parity (HPR) portfolio perform better for various input random variables, indicating better out-of-sample performance. In this way, an economic stacking strategy has demonstrated its worth in improving the out-of-sample capabilities of stacked models, which illustrates how the new understanding of model stacking methodologies may be useful.

Validity and Reliability of Work Ability Index in Turkish Context: Inter-Level, Direct, and Indirect Relations with Job Satisfaction and Task Performance

Descriptive research on work ability is well advanced. However, literature is still far from explaining its consequences. Besides, Turkish literature has been quite limited in terms of considering the work ability concept. In the research, the work ability index, and task performance scale were adapted to Turkish. The research was patterned with quantitative method. According to findings, the improvement of work ability levels and increase in job satisfaction and task performance are related in the same direction. Moreover, work ability positively contributes to directly predicting job satisfaction and task performance. Furthermore, work ability positively contributes to indirectly predicting task performance through job satisfaction. This mediation effect, determined for the first time, is an important research finding regarding its contribution to literature. Apart from these, there are significant differences in work ability among age groups, and certain age group categories have a moderating effect on the relationship between work ability and task performance. This study demonstrated that work ability index is a valid and reliable tool for the Turkish sample. Besides, the study provides holistic findings thanks to work ability levels and direct and indirect effect analysis. The theoretical and practical implications were discussed, and directions were made to further research.

Single-molecule tracking to determine the abundances and stoichiometries of freely-diffusing protein complexes in living cells: Past applications and future prospects

Most biological processes in living cells rely on interactions between proteins. Live-cell compatible approaches that can quantify to what extent a given protein participates in homo- and hetero-oligomeric complexes of different size and subunit composition are therefore critical to advance our understanding of how cellular physiology is governed by these molecular interactions. Biomolecular complex formation changes the diffusion coefficient of constituent proteins, and these changes can be measured using fluorescence microscopy-based approaches, such as single-molecule tracking, fluorescence correlation spectroscopy, and fluorescence recovery after photobleaching. In this review, we focus on the use of single-molecule tracking to identify, resolve, and quantify the presence of freely-diffusing proteins and protein complexes in living cells. We compare and contrast different data analysis methods that are currently employed in the field and discuss experimental designs that can aid the interpretation of the obtained results. Comparisons of diffusion rates for different proteins and protein complexes in intracellular aqueous environments reported in the recent literature reveal a clear and systematic deviation from the Stokes-Einstein diffusion theory. While a complete and quantitative theoretical explanation of why such deviations manifest is missing, the available data suggest the possibility of weighing freely-diffusing proteins and protein complexes in living cells by measuring their diffusion coefficients. Mapping individual diffusive states to protein complexes of defined molecular weight, subunit stoichiometry, and structure promises to provide key new insights into how protein-protein interactions regulate protein conformational, translational, and rotational dynamics, and ultimately protein function.

Predictive overfitting in immunological applications: Pitfalls and solutions

Overfitting describes the phenomenon where a highly predictive model on the training data generalizes poorly to future observations. It is a common concern when applying machine learning techniques to contemporary medical applications, such as predicting vaccination response and disease status in infectious disease or cancer studies. This review examines the causes of overfitting and offers strategies to counteract it, focusing on model complexity reduction, reliable model evaluation, and harnessing data diversity. Through discussion of the underlying mathematical models and illustrative examples using both synthetic data and published real datasets, our objective is to equip analysts and bioinformaticians with the knowledge and tools necessary to detect and mitigate overfitting in their research.

Interface State Density Prediction between an Insulator and a Semiconductor by Gaussian Process Regression Models for a Modified Process

During data-driven process condition optimization on a laboratory scale, only a small-size data set is accessible and should be effectively utilized. On the other hand, during process development, new operations are frequently inserted or current operations are modified. These accessible data sets are somewhat related but not exactly the same type. In this study, we focus on the prediction of the quality of the interface between an insulator and GaN as a semiconductor for the potential application of GaN power semiconductor devices. The quality of the interface was represented as the interface state density, Dit, and the inserted operation to the process was the ultraviolet (UV)/O3-gas treatment. Our retrospective evaluation of model-building approaches for Dit prediction from a process condition revealed that for the UV/O3-treated interfaces, data of interfaces without the treatment contributed to performance improvement. Such performance improvement was not observed when using a data set of Si as the semiconductor. As a modeling method, the automatic relevance vector-based Gaussian process regression with the prior distribution of the length-scale parameters exhibited a relatively high predictive performance and represented a reasonable uncertainty of prediction as reflected by the distance to the training data set. This feature is a prerequisite for a potential application of Bayesian optimization. Furthermore, hyperparameters in the prior distribution of the length-scales could be optimized by leave-one-out cross-validation.

Machine Learning Analysis of Microtensile Bond Strength of Dental Adhesives

Dental adhesives provide retention to composite fillings in dental restorations. Microtensile bond strength (µTBS) test is the most used laboratory test to evaluate bonding performance of dental adhesives. The traditional approach for developing dental adhesives involves repetitive laboratory measurements, which consumes enormous time and resources. Machine learning (ML) is a promising tool for accelerating this process. This study aimed to develop ML models to predict the µTBS of dental adhesives using their chemical features and to identify important contributing factors for µTBS. Specifically, the chemical composition and µTBS information of 81 dental adhesives were collected from the manufacturers and the literature. The average µTBS value of each adhesive was labeled as either 0 (if <36 MPa) or 1 (if ≥36 MPa) to denote the low and high µTBS classes. The initial 9-feature data set comprised pH, HEMA, BisGMA, UDMA, MDP, PENTA, filler, fluoride, and organic solvent (OS) as input features. Nine ML algorithms, including logistic regression, k-nearest neighbor, support vector machine, decision trees and tree-based ensembles, and multilayer perceptron, were implemented for model development. Feature importance analysis identified MDP, pH, OS, and HEMA as the top 4 contributing features, which were used to construct a 4-feature data set. Grid search with stratified 10-fold cross-validation (CV) was employed for hyperparameter tunning and model performance evaluation using 2 metrics, the area under the receiver operating characteristic curve (AUC) and accuracy. The 4-feature data set generated slightly better performance than the 9-feature data set, with the highest AUC score of 0.90 and accuracy of 0.81 based on stratified CV. In conclusion, ML is an effective tool for predicting dental adhesives with low and high µTBS values and for identifying important chemical features contributing to the µTBS. The ML-based data-driven approach has great potential to accelerate the discovery of new dental adhesives and other dental materials.

Strategy inference during learning via cognitive activity-based credit assignment models

We develop a method for selecting meaningful learning strategies based solely on the behavioral data of a single individual in a learning experiment. We use simple Activity-Credit Assignment algorithms to model the different strategies and couple them with a novel hold-out statistical selection method. Application on rat behavioral data in a continuous T-maze task reveals a particular learning strategy that consists in chunking the paths used by the animal. Neuronal data collected in the dorsomedial striatum confirm this strategy.

Estimation of rice seedling growth traits with an end-to-end multi-objective deep learning framework

In recent years, rice seedling raising factories have gradually been promoted in China. The seedlings bred in the factory need to be selected manually and then transplanted to the field. Growth-related traits such as height and biomass are important indicators for quantifying the growth of rice seedlings. Nowadays, the development of image-based plant phenotyping has received increasing attention, however, there is still room for improvement in plant phenotyping methods to meet the demand for rapid, robust and low-cost extraction of phenotypic measurements from images in environmentally-controlled plant factories. In this study, a method based on convolutional neural networks (CNNs) and digital images was applied to estimate the growth of rice seedlings in a controlled environment. Specifically, an end-to-end framework consisting of hybrid CNNs took color images, scaling factor and image acquisition distance as input and directly predicted the shoot height (SH) and shoot fresh weight (SFW) after image segmentation. The results on the rice seedlings dataset collected by different optical sensors demonstrated that the proposed model outperformed compared random forest (RF) and regression CNN models (RCNN). The model achieved R² values of 0.980 and 0.717, and normalized root mean square error (NRMSE) values of 2.64% and 17.23%, respectively. The hybrid CNNs method can learn the relationship between digital images and seedling growth traits, promising to provide a convenient and flexible estimation tool for the non-destructive monitoring of seedling growth in controlled environments.

iDHS-FFLG: Identifying DNase I Hypersensitive Sites by Feature Fusion and Local-Global Feature Extraction Network

The DNase I hypersensitive sites (DHSs) are active regions on chromatin that have been found to be highly sensitive to DNase I. These regions contain various cis-regulatory elements, including promoters, enhancers and silencers. Accurate identification of DHSs helps researchers better understand the transcriptional machinery of DNA and deepen the knowledge of functional DNA elements in non-coding sequences. Researchers have developed many methods based on traditional experiments and machine learning to identify DHSs. However, low prediction accuracy and robustness limit their application in genetics research. In this paper, a novel computational approach based on deep learning is proposed by feature fusion and local-global feature extraction network to identify DHSs in mouse, named iDHS-FFLG. First of all, multiple binary features of nucleotides are fused to better express sequence information. Then, a network consisting of the convolutional neural network (CNN), bidirectional long short-term memory (BiLSTM) and self-attention mechanism is designed to extract local features and global contextual associations. In the end, the prediction module is applied to distinguish between DHSs and non-DHSs. The results of several experiments demonstrate the superior performances of iDHS-FFLG compared to the latest methods.

Examining the Acceptance of Blockchain by Real Estate Buyers and Sellers

Buying and selling real estate is time consuming and labor intensive, requires many intermediaries, and incurs high fees. Blockchain technology provides the real estate industry with a reliable means of tracking transactions and increases trust between the parties involved. Despite the benefits of blockchain, its adoption in the real estate industry is still in its infancy. Therefore, we investigate the factors that influence the acceptance of blockchain technology by buyers and sellers of real estate. A research model was designed based on the combined strengths of the unified theory of technology acceptance and use model and the technology readiness index model. Data were collected from 301 real estate buyers and sellers and analyzed using the partial least squares method. The study found that real estate stakeholders should focus on psychological factors rather than technological factors when adopting blockchain. This study adds to the existing body of knowledge and provides valuable insights to real estate stakeholders on how to implement blockchain technology.

Prediction of enzymatic function with high efficiency and a reduced number of features using genetic algorithm

The post-genomic era has raised a growing demand for efficient procedures to identify protein functions, which can be accomplished by applying machine learning to the characteristics set extracted from the protein. This approach is feature-based and has been the focus of several works in bioinformatics. In this work, we investigated the characteristics of proteins, representing the primary, secondary, tertiary, and quaternary structures of the protein, that improve the model's quality by applying dimensionality reduction techniques and using the Support Vector Machine classifier for predicting the enzymes' classes. During the investigation, two approaches were evaluated: feature extraction/transformation, which was performed using the statistical technique Factor Analysis, and feature selection methods. For feature selection, we proposed an approach based on a genetic algorithm to face the optimization conflict between the simplicity and reliability of an ideal representation of the characteristics of the enzymes and also compared and employed other methods for this purpose. The best result was accomplished using a feature subset generated by our implementation of a multi-objective genetic algorithm enriched with features that this work identified as relevant to represent the enzymes. This subset representation reduced the dataset by about 87% and reached 85.78% of F-measure performance, improving the overall quality of the model classification. In addition, we verified in this work a subset addressed with only 28 features out of a total of 424 that reached a performance above 80% of F-measure for four of the six evaluated classes, showing that satisfactory classification performance can be achieved with a reduced number of enzymes's characteristics. The datasets and implementations are openly available.

Impact of supply chain digitalization on supply chain resilience and performance: A multi-mediation model

The outbreak of COVID-19 has accelerated the building of resilient supply chains, and supply chain digitalization is gradually being recognized as an enabling means to this end. Nevertheless, scholars generally agree that more empirical studies will need to be conducted on how digitalization can facilitate supply chain resilience at various stages and enhance supply chain performance in a highly uncertain environment. To echo the call, this study develops a theoretical influence mechanism of "supply chain digitalization → supply chain resilience → supply chain performance" based on dynamic capability theory. The proposed relationships are validated using survey data collected from 210 Chinese manufacturing companies. The results help identify the paths digitalization and supply chain resilience can take to improve supply chain performance in a turbulent environment. The different roles of three supply chain resilience capabilities, namely absorptive capability (before the disruption), response capability (during the disruption), and recovery capability (after the disruption), which impact on supply chain performance differently, are highlighted. In addition, it is found that digitalization can bring a differential impact on these three supply chain resilience capabilities through different aspects of resource and structural adjustment measures. The findings also confirm the mediating role of absorptive capability, response capability, and recovery capability between digitalization and supply chain performance. During crisis, supply chain digitalization can increase cost-effectiveness, enhance information and communication efficiency, and promote supply chain resilience to achieve better performance. For theoretical contribution, this study enriches the research on supply chain digitalization and resilience by underpinning the relationships between the two with dynamic capability theory. For practical contribution, the research findings provide insights for enterprises to leverage digitalization to strengthen resilience in supply chain.

Discovering functional connectivity features characterizing multiple sclerosis phenotypes using explainable artificial intelligence

Multiple sclerosis (MS) is a neurological condition characterized by severe structural brain damage and by functional reorganization of the main brain networks that try to limit the clinical consequences of structural burden. Resting-state (RS) functional connectivity (FC) abnormalities found in this condition were shown to be variable across different MS phases, according to the severity of clinical manifestations. The article describes a system exploiting machine learning on RS FC matrices to discriminate different MS phenotypes and to identify relevant functional connections for MS stage characterization. To this end, the system exploits some mathematical properties of covariance-based RS FC representation, which can be described by a Riemannian manifold. The classification performance of the proposed framework was significantly above the chance level for all MS phenotypes. Moreover, the proposed system was successful in identifying relevant RS FC alterations contributing to an accurate phenotype classification.

Meta-Analysis of Candidate Surrogate End Points in Advanced Prostate Cancer

BACKGROUND: The Intermediate Clinical Endpoints in Cancer of the Prostate (ICECaP) working group identified metastasis-free survival as a valid surrogate end point for overall survival (OS) for patients with localized prostate cancer. No comparably validated surrogate end points exist in advanced prostate cancer. METHODS: We searched for trials in advanced prostate cancer, defined as node-positive, metastatic castration-sensitive, nonmetastatic, or metastatic castration-resistant prostate cancer. Eligible randomized trials reported OS and one or more intermediate clinical end points, including biochemical failure (BF), clinical failure, biochemical failure–free survival (BFS), progression-free survival (PFS), and radiographic PFS. Candidacy for surrogacy was assessed by using the second condition of the meta-analytic approach; R2 was weighted by the inverse variance of the log intermediate clinical end point hazard ratio and defined as R2>0.70. RESULTS: A total of 143 randomized trials (n=75,601 patients) were included. No candidate end points met the criteria for surrogacy (R2 BF [n=28,922], 0.42 [95% confidence interval (CI), 0.18 to 0.64]; BFS [n=25,741], 0.57 [95% CI, 0.37 to 0.73]; clinical failure [n=22,616], 0.31 [95% CI, 0.075 to 0.56]; PFS [n=52,639], 0.50 [95% CI, 0.35 to 0.63]; and radiographic PFS [n=52,548], 0.50 [95% CI, 0.35 to 0.63]). Within preplanned subgroups according to castration-sensitive or castration-resistant disease or according to treatment type, neither BFS nor PFS consistently met criteria for surrogacy. Sensitivity analyses showed that candidacy for surrogacy of all end points tested did not change over time. CONCLUSIONS: Our aggregate screening method for surrogate end points in advanced prostate cancer showed that commonly used clinical end points are not clear valid surrogate end points for OS. (Funded by the Prostate Cancer Foundation and the National Cancer Institute.)

Revisiting predictions of movie economic success: random Forest applied to profits

Previous studies have employed machine learning tools to classify films according to success to guide a reduction in the degree of uncertainty of film production. We revisited the literature to contribute to three relevant issues in classifying films according to economic success. First, we explored the differences between the results of the shortest or longest samples in terms of time to study possible changes in patterns of consumption mainly due to technological changes and between total and wide-released films. Second, we used profits free of price inflation as measures of economic success instead of the usual box office nominal revenues. Third, we employed a smaller set of features, only the ones available at the time of production, to help producers maneuver contingencies since little or nothing can be done by the time a film is in the theaters. We followed the literature to choose the classifiers - Random Forest, Support Vector Machine, and Neural Network - and designed sub-datasets to model and compare the performance of our results. Our dataset includes all films with budgets disclosed at the Box Office Mojo website, resulting in 3167 movies released at theaters worldwide between 1980 and 2019. The Random Forest results outperform previous similar studies with different sampling in time, including results for a less usual larger sample, with the best data sample about 97% both in accuracy and F1-score.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11042-023-15169-4.

Virtual clinical trial based on outcome modeling with iteratively redistributed extrapolation data

Virtual clinical trials (VCTs) can potentially simulate clinical trials on a computer, but their application with a limited number of past clinical cases is challenging due to the biased estimation of the statistical population. In this study, we developed ExMixup, a novel training technique based on machine learning, using iteratively redistributed extrapolated data. Information obtained from 100 patients with prostate cancer and 385 patients with oropharyngeal cancer was used to predict the recurrence after radiotherapy. Model performance was evaluated by developing outcome prediction models based on three types of training methods: training with original data (baseline), interpolation data (Mixup), and interpolation + extrapolation data (ExMixup). Two types of VCTs were conducted to predict the treatment response of patients with distinct characteristics compared to the training data obtained from patient cohorts categorized under risk classification or cancer stage. The prediction models developed with ExMixup yielded concordance indices (95% confidence intervals) of 0.751 (0.719-0.818) and 0.752 (0.734-0.785) for VCTs on the prostate and oropharyngeal cancer datasets, respectively, which significantly outperformed the baseline and Mixup models (P < 0.01). The proposed approach could enhance the ability of VCTs to predict treatment results in patients excluded from past clinical trials.

Predicting Future Depressive Episodes from Resting-State fMRI with Generative Embedding

After a first episode of major depressive disorder (MDD), there is substantial risk for a long-term remitting-relapsing course. Prevention and early interventions are thus critically important. Various studies have examined the feasibility of detecting at-risk individuals based on out-of-sample predictions about the future occurrence of depression. However, functional magnetic resonance imaging (MRI) has received very little attention for this purpose so far. Here, we explored the utility of generative models (i.e. different dynamic causal models, DCMs) as well as functional connectivity (FC) for predicting future episodes of depression in never-depressed adults, using a large dataset (N=906) of task-free ("resting state") fMRI data from the UK Biobank. Connectivity analyses were conducted using timeseries from pre-computed spatially independent components of different dimensionalities. Over a three year period, 50% of participants showed indications of at least one depressive episode, while the other 50% did not. Using nested cross-validation for training and a held-out test set (80/20 split), we systematically examined the combination of 8 connectivity feature sets and 17 classifiers. We found that a generative embedding procedure based on combining regression DCM (rDCM) with a support vector machine (SVM) enabled the best predictions, both on the training set (0.63 accuracy, 0.66 area under the curve, AUC) and the test set (0.62 accuracy, 0.64 AUC; p<0.001). However, on the test set, rDCM was only slightly superior to predictions based on FC (0.59 accuracy, 0.61 AUC). Interpreting model predictions based on SHAP (SHapley Additive exPlanations) values suggested that the most predictive connections were widely distributed and not confined to specific networks. Overall, our analyses suggest (i) ways of improving future fMRI-based generative embedding approaches for the early detection of individuals at-risk for depression and that (ii) achieving accuracies of clinical utility may require combination of fMRI with other data modalities.

How Do COVID-19 Risk, Life-Safety Risk, Job Insecurity, and Work-Family Conflict Affect Miner Performance? Health-Anxiety and Job-Anxiety Perspectives

The coronavirus pandemic (COVID-19) has created challenging working conditions in coal-production activities. In addition to the massive loss of resources for miners, it has had a devastating impact on these individuals' mental health. Based on the conservation of resources (COR) theory and a resource-loss perspective, this study examined the impact of COVID-19 risk, life-safety risk, perceived job insecurity, and work-family conflict on miners' job performance. Moreover, this study investigated the mediating role of job anxiety (JA) and health anxiety (HA). The study data were collected through online structured questionnaires disseminated to 629 employees working in a coal mine in China. The data analysis and hypothesis generation were conducted using the structural equation modeling (partial least squares) method. The results demonstrated that the perception of COVID-19 risk, life-safety risk, job insecurity, and work-family conflict negatively and significantly impacted miners' job performance. In addition, JA and HA negatively mediated the relationships between the perception of COVID-19 risk, life-safety risk, perceived job insecurity, work-family conflict, and job performance. The findings of this study can give coal-mining companies and their staff useful insights into how to minimize the pandemic's effects on their operations.

Evaluation of Chinese populational exposure to environmental electromagnetic field based on stochastic dosimetry and parametric human modelling

This study aimed to estimate the distribution of the whole-body averaged specific absorption rate (WBSAR) using several measurable physique parameters for Chinese adult population exposed to environmental electromagnetic fields (EMFs) of current wireless communication frequencies, and to discuss the effects of these physique parameters in the frequency-dependent dosimetric results. The physique distribution of Chinese adults was obtained from the National Physical Fitness and Health Database comprising 81,490 adult samples. The number of physique parameters used to construct the surrogate model was reduced to three via mutual information analysis. A stochastic method with 40 deterministic simulations was used to generate frequency-dependent and gender-specific surrogate models for WBSAR via polynomial chaos expansion. In the simulations, we constructed anatomically correct models conforming to the targeted physique parameters via deformable human modelling technique, which was based on deep learning from the image database including 767 Chinese adults. Thereafter, we analysed the sensitivity of the physique parameters to WBSAR by covariance-based Sobol decomposition. The results indicated that the generated models were consistent with the targeted physique parameters. The estimated dosimetric results were validated using finite-difference time-domain simulations (the error was < 6% across all the investigated frequencies for WBSAR). The novelty of the study included that it demonstrated the feasibility of estimating the individual WBSAR using a limited number of physique parameters with the aid of surrogate modelling. In addition, the population-based distribution of the WBSAR in Chinese adults was firstly presented in the manuscript. The results also indicated that the different combinations of physique parameter, dependent on genders and frequencies, significantly influenced the WBSAR, although the general conservativeness of the guidelines of the International Commission on Non-Ionizing Radiation and Protection can be confirmed in the surveyed population.

Machine Learning for Predicting Micro- and Macrovascular Complications in Individuals With Prediabetes or Diabetes: Retrospective Cohort Study

BACKGROUND

Micro- and macrovascular complications are a major burden for individuals with diabetes and can already arise in a prediabetic state. To allocate effective treatments and to possibly prevent these complications, identification of those at risk is essential.

OBJECTIVE

This study aimed to build machine learning (ML) models that predict the risk of developing a micro- or macrovascular complication in individuals with prediabetes or diabetes.

METHODS

In this study, we used electronic health records from Israel that contain information about demographics, biomarkers, medications, and disease codes; span from 2003 to 2013; and were queried to identify individuals with prediabetes or diabetes in 2008. Subsequently, we aimed to predict which of these individuals developed a micro- or macrovascular complication within the next 5 years. We included 3 microvascular complications: retinopathy, nephropathy, and neuropathy. In addition, we considered 3 macrovascular complications: peripheral vascular disease (PVD), cerebrovascular disease (CeVD), and cardiovascular disease (CVD). Complications were identified via disease codes, and, for nephropathy, the estimated glomerular filtration rate and albuminuria were considered additionally. Inclusion criteria were complete information on age and sex and on disease codes (or measurements of estimated glomerular filtration rate and albuminuria for nephropathy) until 2013 to account for patient dropout. Exclusion criteria for predicting a complication were diagnosis of this specific complication before or in 2008. In total, 105 predictors from demographics, biomarkers, medications, and disease codes were used to build the ML models. We compared 2 ML models: logistic regression and gradient-boosted decision trees (GBDTs). To explain the predictions of the GBDTs, we calculated Shapley additive explanations values.

RESULTS

Overall, 13,904 and 4259 individuals with prediabetes and diabetes, respectively, were identified in our underlying data set. For individuals with prediabetes, the areas under the receiver operating characteristic curve for logistic regression and GBDTs were, respectively, 0.657 and 0.681 (retinopathy), 0.807 and 0.815 (nephropathy), 0.727 and 0.706 (neuropathy), 0.730 and 0.727 (PVD), 0.687 and 0.693 (CeVD), and 0.707 and 0.705 (CVD); for individuals with diabetes, the areas under the receiver operating characteristic curve were, respectively, 0.673 and 0.726 (retinopathy), 0.763 and 0.775 (nephropathy), 0.745 and 0.771 (neuropathy), 0.698 and 0.715 (PVD), 0.651 and 0.646 (CeVD), and 0.686 and 0.680 (CVD). Overall, the prediction performance is comparable for logistic regression and GBDTs. The Shapley additive explanations values showed that increased levels of blood glucose, glycated hemoglobin, and serum creatinine are risk factors for microvascular complications. Age and hypertension were associated with an elevated risk for macrovascular complications.

CONCLUSIONS

Our ML models allow for an identification of individuals with prediabetes or diabetes who are at increased risk of developing micro- or macrovascular complications. The prediction performance varied across complications and target populations but was in an acceptable range for most prediction tasks.

Chemical Descriptors for a Large-Scale Study on Drop-Weight Impact Sensitivity of High Explosives

The drop-weight impact test is an experiment that has been used for nearly 80 years to evaluate handling sensitivity of high explosives. Although the results of this test are known to have large statistical uncertainties, it is one of the most common tests due to its accessibility and modest material requirements. In this paper, we compile a large data set of drop-weight impact sensitivity test results (mainly performed at Los Alamos National Laboratory), along with a compendium of molecular and chemical descriptors for the explosives under test. These data consist of over 500 unique explosives, over 1000 repeat tests, and over 100 descriptors, for a total of about 1500 observations. We use random forest methods to estimate a model of explosive handling sensitivity as a function of chemical and molecular properties of the explosives under test. Our model predicts well across a wide range of explosive types, spanning a broad range of explosive performance and sensitivity. We find that properties related to explosive performance, such as heat of explosion, oxygen balance, and functional group, are highly predictive of explosive handling sensitivity. Yet, models that omit many of these properties still perform well. Our results suggest that there is not one or even several factors that explain explosive handling sensitivity, but that there are many complex, interrelated effects at play.