Search

The rapid advancement of machine learning techniques has led to their widespread application in various domains, including water resources. However, snowmelt modeling remains an area that has not been extensively explored. In this study, we propose a state-of-the-art (SOTA) deep learning sequential model, leveraging a Temporal Convolutional Network (TCN), for snowmelt forecasting of the Hindu Kush Himalayan (HKH) region. To evaluate the performance of our proposed model, we conducted a comparative analysis with other popular models, including Support Vector Regression (SVR), Long Short-Term Memory (LSTM), and Transformer models. Furthermore, nested cross-validation (CV) was used with five outer folds and three inner folds, and hyperparameter tuning was performed on the inner folds. To evaluate the performance of the model, the Mean Absolute Error (MAE), Root-Mean-Square Error (RMSE), R square ( R 2 ), Kling–Gupta Efficiency (KGE), and Nash–Sutcliffe Efficiency (NSE) were computed for each outer fold. The average metrics revealed that the TCN outperformed the other models, with an average MAE of 0.011, RMSE of 0.023, R 2 of 0.991, KGE of 0.992, and NSE of 0.991 for one-day forecasts of streamflow. The findings of this study demonstrate the effectiveness of the proposed deep learning model as compared to traditional machine learning approaches for snowmelt-driven streamflow forecasting. Moreover, the superior performance of this TCN highlights its potential as a promising deep learning model for similar hydrological applications. [ABSTRACT FROM AUTHOR]

Abstract Social difficulties during interactions with others are central to autism spectrum disorder (ASD). Understanding the links between these social difficulties and their underlying neural processes is a primary aim focused on improved diagnosis and treatment. In keeping with this goal, we have developed a multivariate classification method based on neural data acquired by functional near infrared spectroscopy, fNIRS, during live eye-to-eye contact with adults who were either typically developed (TD) or individuals with ASD. The ASD diagnosis was based on the gold-standard Autism Diagnostic Observation Schedule (ADOS) which also provides an index of symptom severity. Using a nested cross-validation method, a support vector machine (SVM) was trained to discriminate between ASD and TD groups based on the neural responses during eye-to-eye contact. ADOS scores were not applied in the classification training. To test the hypothesis that SVM identifies neural activity patterns related to one of the neural mechanisms underlying the behavioral symptoms of ASD, we determined the correlation coefficient between the SVM scores and the individual ADOS scores. Consistent with the hypothesis, the correlation between observed and predicted ADOS scores was 0.72 (p

This study aimed to evaluate the utility and applicability of electrooculography (EOG) when studying ocular activity during complex motor behavior. Due to its lower spatial resolution relative to eye tracking (ET), it is unclear whether EOG can provide valid and accurate temporal measurements such as the duration of the Quiet Eye (QE), that is the uninterrupted dwell time on the visual target prior to and during action. However, because of its greater temporal resolution, EOG is better suited for temporal‐spectral decomposition, a technique that allows us to distinguish between lower and higher frequency activity as a function of time. Sixteen golfers of varying expertise (novices to experts) putted 60 balls to a 4‐m distant target on a flat surface while we recorded EOG, ET, performance accuracy, and putter kinematics. Correlational and discrepancy analyses confirmed that EOG yielded valid and accurate QE measurements, but only when using certain processing parameters. Nested cross‐validation indicated that, among a set of ET and EOG temporal and spectral oculomotor features, EOG power was the most useful when predicting performance accuracy through robust regression. Follow‐up cross‐validation and correlational analyses revealed that more accurate performance was preceded by diminished lower‐frequency activity immediately before movement initiation and elevated higher‐frequency activity during movement recorded from the horizontal channel. This higher‐frequency activity was also found to accompany a smoother movement execution. This study validates EOG algorithms (code provided) for measuring temporal parameters and presents a novel approach to extracting temporal and spectral oculomotor features during complex motor behavior. This study validates electrooculography for computing temporal indices of oculomotor activity typically obtained through mobile eye tracking. It also indicates that using EOG indices that embed both temporal and spectral features can improve predictions of movement accuracy compared to those that rely on temporal features only. [ABSTRACT FROM AUTHOR]

Social difficulties during interactions with others are central to autism spectrum disorder (ASD). Understanding the links between these social difficulties and their underlying neural processes is a primary aim focused on improved diagnosis and treatment. In keeping with this goal, we have developed a multivariate classification method based on neural data acquired by functional near infrared spectroscopy, fNIRS, during live eye-to-eye contact with adults who were either typically developed (TD) or individuals with ASD. The ASD diagnosis was based on the gold-standard Autism Diagnostic Observation Schedule (ADOS) which also provides an index of symptom severity. Using a nested cross-validation method, a support vector machine (SVM) was trained to discriminate between ASD and TD groups based on the neural responses during eye-to-eye contact. ADOS scores were not applied in the classification training. To test the hypothesis that SVM identifies neural activity patterns related to one of the neural mechanisms underlying the behavioral symptoms of ASD, we determined the correlation coefficient between the SVM scores and the individual ADOS scores. Consistent with the hypothesis, the correlation between observed and predicted ADOS scores was 0.72 (p < 0.002). Findings suggest that multivariate classification methods combined with the live interaction paradigm of eye-to-eye contact provide a promising approach to link neural processes and social difficulties in individuals with ASD. [ABSTRACT FROM AUTHOR]

Summary: Cross‐validation (CV) is one of the most widely used techniques in statistical learning for estimating the test error of a model, but its behavior is not yet fully understood. It has been shown that standard confidence intervals for test error using estimates from CV may have coverage below nominal levels. This phenomenon occurs because each sample is used in both the training and testing procedures during CV and as a result, the CV estimates of the errors become correlated. Without accounting for this correlation, the estimate of the variance is smaller than it should be. One way to mitigate this issue is by estimating the mean squared error of the prediction error instead using nested CV. This approach has been shown to achieve superior coverage compared to intervals derived from standard CV. In this work, we generalize the nested CV idea to the Cox proportional hazards model and explore various choices of test error for this setting. [ABSTRACT FROM AUTHOR]

The rapid advancement of machine learning techniques has led to their widespread application in various domains, including water resources. However, snowmelt modeling remains an area that has not been extensively explored. In this study, we propose a state-of-the-art (SOTA) deep learning sequential model, leveraging a Temporal Convolutional Network (TCN), for snowmelt forecasting of the Hindu Kush Himalayan (HKH) region. To evaluate the performance of our proposed model, we conducted a comparative analysis with other popular models, including Support Vector Regression (SVR), Long Short-Term Memory (LSTM), and Transformer models. Furthermore, nested cross-validation (CV) was used with five outer folds and three inner folds, and hyperparameter tuning was performed on the inner folds. To evaluate the performance of the model, the Mean Absolute Error (MAE), Root-Mean-Square Error (RMSE), R square (R2), Kling–Gupta Efficiency (KGE), and Nash–Sutcliffe Efficiency (NSE) were computed for each outer fold. The average metrics revealed that the TCN outperformed the other models, with an average MAE of 0.011, RMSE of 0.023, R2 of 0.991, KGE of 0.992, and NSE of 0.991 for one-day forecasts of streamflow. The findings of this study demonstrate the effectiveness of the proposed deep learning model as compared to traditional machine learning approaches for snowmelt-driven streamflow forecasting. Moreover, the superior performance of this TCN highlights its potential as a promising deep learning model for similar hydrological applications.

Background: Machine learning models trained with multiparametric quantitative MRIs (qMRIs) have the potential to provide valuable information about the structural composition of articular cartilage. Purpose: To study the performance and feasibility of machine learning models combined with qMRIs for noninvasive assessment of collagen fiber orientation and proteoglycan content. Study Type: Retrospective, animal model. Animal Model: An open‐source single slice MRI dataset obtained from 20 samples of 10 Shetland ponies (seven with surgically induced cartilage lesions followed by treatment and three healthy controls) yielded to 1600 data points, including 10% for test and 90% for train validation. Field Strength/Sequence: A 9.4 T MRI scanner/qMRI sequences: T1, T2, adiabatic T1ρ and T2ρ, continuous‐wave T1ρ and relaxation along a fictitious field (TRAFF) maps. Assessment: Five machine learning regression models were developed: random forest (RF), support vector regression (SVR), gradient boosting (GB), multilayer perceptron (MLP), and Gaussian process regression (GPR). A nested cross‐validation was used for performance evaluation. For reference, proteoglycan content and collagen fiber orientation were determined by quantitative histology from digital densitometry (DD) and polarized light microscopy (PLM), respectively. Statistical Tests: Normality was tested using Shapiro–Wilk test, and association between predicted and measured values was evaluated using Spearman's Rho test. A P‐value of 0.05 was considered as the limit of statistical significance. Results: Four out of the five models (RF, GB, MLP, and GPR) yielded high accuracy (R2 = 0.68–0.75 for PLM and 0.62–0.66 for DD), and strong significant correlations between the reference measurements and predicted cartilage matrix properties (Spearman's Rho = 0.72–0.88 for PLM and 0.61–0.83 for DD). GPR algorithm had the highest accuracy (R2 = 0.75 and 0.66) and lowest prediction‐error (root mean squared [RMSE] = 1.34 and 2.55) for PLM and DD, respectively. Data Conclusion: Multiparametric qMRIs in combination with regression models can determine cartilage compositional and structural features, with higher accuracy for collagen fiber orientation than proteoglycan content. Evidence Level: 2 Technical Efficacy: Stage 2 [ABSTRACT FROM AUTHOR]

To overcome the problem that the point prediction method of slope safety factor has uncertainty in its prediction and hence cannot a reliable slope safety factor, a quantile-based ensemble learning regression method is proposed that quantifies the uncertainty of point prediction models by predicting the value interval of the slope safety factor. Firstly, the slope safety factor and its influencing factors are preprocessed to obtain an effective dataset. Secondly, with minimizing the quantile loss as the model optimization goal, three commonly used ensemble learning models (Random Forest, LightGBM, and CatBoost) are optimized by grid search and nested 10-fold cross-validation to obtain three quantile-based ensemble learning models (QRF, QLightGBM, and QCatBoost) that can directly construct prediction intervals. Finally, the corrected predictive interval accuracy (CPIA) index is introduced to compare the prediction performance of the different models under different quantile conditions. The results show that the three proposed models can eliminate the influence of the uncertainty of the traditional point prediction model. Among them, the point prediction performance of the models is 7.9%, 4.1%, and 0.5% higher than that of the traditional model in terms of R2, and CPIA is 0.713, 0.729, and 0.703, respectively. The proposed method can provide an effective value interval for the prediction of slope safety factors and provide a new idea for the uncertainty analysis of safety factors. [ABSTRACT FROM AUTHOR]

New methods to perform time series classification arise frequently and multiple state-of-the-art approaches achieve high performance on benchmark datasets with respect to accuracy and computation time. However, often the modeling procedures do not include proper validation but rather rely only on either external test dataset or one-level cross-validation. ATSC-NEX is an automated procedure that employs sequential model-based optimization together with nested cross-validation to build an accurate and properly validated time series classification model. It aims to find an optimal pipeline configuration that includes the selection of input type and settings, as well as model type and hyperparameters. The results of a case study in which a model for the identification of diesel engine type is developed, show that the algorithm can efficiently find a well-performing pipeline configuration. The comparison between ATSC-NEX and some state-of-the-art methods on several benchmark datasets shows that similar accuracy can be achieved.

Blood-brain barrier (BBB) is a major barrier to drug delivery into the brain in the treatment of central nervous system (CNS) diseases. Blood-brain barrier penetrating peptides (BBPs), a class of peptides that can cross BBB through various mechanisms without damaging BBB, are effective drug candidates for CNS diseases. However, identification of BBPs by experimental methods is time-consuming and laborious. To discover more BBPs as drugs for CNS disease, it is urgent to develop computational methods that can quickly and accurately identify BBPs and non-BBPs. In the present study, we created a training dataset that consists of 326 BBPs derived from previous databases and published manuscripts and 326 non-BBPs collected from UniProt, to construct a BBP predictor based on sequence information. We also constructed an independent testing dataset with 99 BBPs and 99 non-BBPs. Multiple machine learning methods were compared based on the training dataset via a nested cross-validation. The final BBP predictor was constructed based on the training dataset and the results showed that random forest (RF) method outperformed other classification algorithms on the training and independent testing dataset. Compared with previous BBP prediction tools, the RF-based predictor, named BBPpredict, performs considerably better than state-of-the-art BBP predictors. BBPpredict is expected to contribute to the discovery of novel BBPs, or at least can be a useful complement to the existing methods in this area. BBPpredict is freely available at http://i.uestc.edu.cn/BBPpredict/cgi-bin/BBPpredict.pl. [ABSTRACT FROM AUTHOR]

Blood-brain barrier (BBB) is a major barrier to drug delivery into the brain in the treatment of central nervous system (CNS) diseases. Blood-brain barrier penetrating peptides (BBPs), a class of peptides that can cross BBB through various mechanisms without damaging BBB, are effective drug candidates for CNS diseases. However, identification of BBPs by experimental methods is time-consuming and laborious. To discover more BBPs as drugs for CNS disease, it is urgent to develop computational methods that can quickly and accurately identify BBPs and non-BBPs. In the present study, we created a training dataset that consists of 326 BBPs derived from previous databases and published manuscripts and 326 non-BBPs collected from UniProt, to construct a BBP predictor based on sequence information. We also constructed an independent testing dataset with 99 BBPs and 99 non-BBPs. Multiple machine learning methods were compared based on the training dataset via a nested cross-validation. The final BBP predictor was constructed based on the training dataset and the results showed that random forest (RF) method outperformed other classification algorithms on the training and independent testing dataset. Compared with previous BBP prediction tools, the RF-based predictor, named BBPpredict, performs considerably better than state-of-the-art BBP predictors. BBPpredict is expected to contribute to the discovery of novel BBPs, or at least can be a useful complement to the existing methods in this area. BBPpredict is freely available at http://i.uestc.edu.cn/BBPpredict/cgi-bin/BBPpredict.pl.

Volatile organic compounds (VOCs) indicative of pork microbial spoilage can be quantified rapidly at trace levels using selected-ion flow-tube mass spectrometry (SIFT-MS). Packaging atmosphere is one of the factors influencing VOC production patterns during storage. On this basis, machine learning would help to process complex volatolomic data and predict pork microbial quality efficiently. This study focused on (1) investigating model generalizability based on different nested cross-validation settings, and (2) comparing the predictive power and feature importance of nine algorithms, including Artificial Neural Network (ANN), k -Nearest Neighbors, Support Vector Regression, Decision Tree, Partial Least Squares Regression, and four ensemble learning models. The datasets used contain 37 VOCs' concentrations (input) and total plate counts (TPC, output) of 350 pork samples with different storage times, including 225 pork loin samples stored under three high- O 2 and three low- O 2 conditions, and 125 commercially packaged products. An appropriate choice of cross-validation strategies resulted in trustworthy and relevant predictions. When trained on all possible selections of two high- O 2 and two low- O 2 conditions, ANNs produced satisfactory TPC predictions of unseen test scenarios (one high- O 2 condition, one low- O 2 condition, and the commercial products). ANN-based bagging outperformed other employed models, when TPC exceeded ca. 6 log CFU/g. VOCs including benzaldehyde, 3-methyl-1-butanol, ethanol and methyl mercaptan were identified with high feature importance. This elaborated case study illustrates great prospects of real-time detection techniques and machine learning in meat quality prediction. Further investigations on handling low VOC levels would enhance the model performance and decision making in commercial meat quality control. [Display omitted] • Real-time SIFT-MS data collected from pork stored under modified atmospheres. • Prediction of pork microbial quality using volatolome-based machine learning. • A sensible cross-validation setting selection improved model generalizability. • ANN-based bagging delivered satisfactory predictions of total microbial counts. • Several volatile organic compounds were identified with high feature importance. [ABSTRACT FROM AUTHOR]

Applications of machine-learning-based approaches in the geosciences have witnessed a substantial increase over the past few years. Here we present an approach that accounts for spatial autocorrelation by introducing spatial features to the models. In particular, we explore two types of spatial features, namely spatial lag and eigenvector spatial filtering (ESF). These features are used within the widely used random forest (RF) method, and their effect is illustrated on two public datasets of varying sizes (Meuse and California housing datasets). The least absolute shrinkage and selection operator (LASSO) is used to determine the best subset of spatial features, and nested cross-validation is used for hyper-parameter tuning and performance evaluation. We utilize Moran's I and local indicators of spatial association (LISA) to assess how spatial autocorrelation is captured at both global and local scales. Our results show that RF models combined with either spatial lag or ESF features yield lower errors (up to 33% different) and reduce the global spatial autocorrelation of the residuals (up to 95% decrease in Moran's I) compared to the RF model with no spatial features. The local autocorrelation patterns of the residuals are weakened as well. Compared to benchmark geographically weighted regression (GWR) models, the RF models with spatial features yielded more accurate models with similar levels of global and local autocorrelation in the prediction residuals. This study reveals the effectiveness of spatial features in capturing spatial autocorrelation and provides a generic machine-learning modelling workflow for spatial prediction. [ABSTRACT FROM AUTHOR]

The sharp slope deformation which often contains seasonal patterns is the major source of the landslide hazard with respect to the local community, which it is a serious geological environment problem. In this paper, a long short-term memory-based deep learning framework has been proposed to model the deformation behaviors especially the sharp deformation of the landslide. First, the Box–Cox transformation is applied to normalize the dataset that includes time-series deformation, precipitation, and reservoir water level. Then, an elastic net (EN)-based ensemble of long short-term memory recurrent neural networks (LSTM-RNNs) is applied to forecast landslide deformation by using month forward-chaining nested cross-validation. This method is performed on the time-series data as our training strategy. Last, the Hurst exponent is formulated to identify incoming sharp deformation. The computational results demonstrated that this approach can accurately identify future sharp deformation. The Hurst exponent illustrates the abnormal patterns in the prediction errors which indicate sharp deformation. As a result, the proposed framework would assist the on-site risk analysis and decision-making process for geological engineers to prevent the landslide hazard in the future.

PurposeThe present study aimed to preoperatively predict the status of 1p/19q based on radiomics analysis in patients with World Health Organization (WHO) grade II gliomas.MethodsThis retrospective study enrolled 157 patients with WHO grade II gliomas (76 patients with astrocytomas with mutant IDH, 16 patients with astrocytomas with wild-type IDH, and 65 patients with oligodendrogliomas with mutant IDH and 1p/19q codeletion). Radiomic features were extracted from magnetic resonance images, including T1-weighted, T2-weighted, and contrast T1-weighted images. Elastic net and support vector machines with radial basis function kernel were applied in nested 10-fold cross-validation loops to predict the 1p/19q status. Receiver operating characteristic analysis and precision-recall analysis were used to evaluate the model performance. Student’s t-tests were then used to compare the posterior probabilities of 1p/19q co-deletion prediction in the group with different 1p/19q status.ResultsSix valuable radiomic features, along with age, were selected with the nested 10-fold cross-validation loops. Five features showed significant difference in patients with different 1p/19q status. The area under curve and accuracy of the predictive model were 0.8079 (95% confidence interval, 0.733–0.8755) and 0.758 (0.6879–0.8217), respectively, and the F1-score of the precision-recall curve achieved 0.6667 (0.5201–0.7705). The posterior probabilities in the 1p/19q co-deletion group were significantly different from the non-deletion group.ConclusionCombined radiomics analysis and machine learning showed potential clinical utility in the preoperative prediction of 1p/19q status, which can aid in making customized neurosurgery plans and glioma management strategies before postoperative pathology.

Purpose: The present study aimed to preoperatively predict the status of 1p/19q based on radiomics analysis in patients with World Health Organization (WHO) grade II gliomas. Methods: This retrospective study enrolled 157 patients with WHO grade II gliomas (76 patients with astrocytomas with mutant IDH, 16 patients with astrocytomas with wild-type IDH, and 65 patients with oligodendrogliomas with mutant IDH and 1p/19q codeletion). Radiomic features were extracted from magnetic resonance images, including T1-weighted, T2-weighted, and contrast T1-weighted images. Elastic net and support vector machines with radial basis function kernel were applied in nested 10-fold cross-validation loops to predict the 1p/19q status. Receiver operating characteristic analysis and precision-recall analysis were used to evaluate the model performance. Student's t -tests were then used to compare the posterior probabilities of 1p/19q co-deletion prediction in the group with different 1p/19q status. Results: Six valuable radiomic features, along with age, were selected with the nested 10-fold cross-validation loops. Five features showed significant difference in patients with different 1p/19q status. The area under curve and accuracy of the predictive model were 0.8079 (95% confidence interval, 0.733–0.8755) and 0.758 (0.6879–0.8217), respectively, and the F1-score of the precision-recall curve achieved 0.6667 (0.5201–0.7705). The posterior probabilities in the 1p/19q co-deletion group were significantly different from the non-deletion group. Conclusion: Combined radiomics analysis and machine learning showed potential clinical utility in the preoperative prediction of 1p/19q status, which can aid in making customized neurosurgery plans and glioma management strategies before postoperative pathology. [ABSTRACT FROM AUTHOR]

Both wrapper and hybrid methods in feature selection need the intervention of learning algorithm to train parameters. The preset parameters and dataset are used to construct several sub-optimal models, from which the final model is selected. The question is how to evaluate the performance of these sub-optimal models? What are the effects of different evaluation methods of sub-optimal model on the result of feature selection? Aiming at the evaluation problem of predictive models in feature selection, we chose a hybrid feature selection algorithm, FDHSFFS, and conducted comparative experiments on four UCI datasets with large differences in feature dimension and sample size by using five different cross-validation (CV) methods. The experimental results show that in the process of feature selection, twofold CV and leave-one-out-CV are more suitable for the model evaluation of low-dimensional and small sample datasets, tenfold nested CV and tenfold CV are more suitable for the model evaluation of high-dimensional datasets; tenfold nested CV is close to the unbiased estimation, and different optimal models may choose the same approximate optimal feature subset.

The detection of mutations in telomerase reverse transcriptase promoter (pTERT) is important since preoperative diagnosis of pTERT status helps with evaluating prognosis and determining the surgical strategy. Here, we aimed to establish a radiomics-based machine-learning algorithm and evaluated its performance with regard to the prediction of mutations in pTERT in patients with World Health Organization (WHO) grade II gliomas. In total, 164 patients with WHO grade II gliomas were enrolled in this retrospective study. We extracted a total of 1,293 radiomics features from multi-parametric magnetic resonance imaging scans. Elastic net (used for feature selection) and support vector machine with linear kernel were applied in nested 10-fold cross-validation loops. The predictive model was evaluated by receiver operating characteristic and precision-recall analyses. We performed an unpaired t-test to compare the posterior predictive probabilities among patients with differing pTERT statuses. We selected 12 valuable radiomics features using nested 10-fold cross-validation loops. The area under the curve (AUC) was 0.8446 (95% confidence interval [CI], 0.7735–0.9065) with an optimal summed value of sensitivity of 0.9355 (95% CI, 0.8802–0.9788) and specificity of 0.6197 (95% CI, 0.5071–0.7371). The overall accuracy was 0.7988 (95% CI, 0.7378–0.8598). The F1-score was 0.8406 (95% CI, 0.7684–0.902) with an optimal precision of 0.7632 (95% CI, 0.6818–0.8364) and recall of 0.9355 (95% CI, 0.8802–0.9788). Posterior probabilities of pTERT mutations were significantly different between patients with wild-type and mutant TERT promoters. Our findings suggest that a radiomics analysis with a machine-learning algorithm can be useful for predicting pTERT status in patients with WHO grade II glioma and may aid in glioma management.

The detection of mutations in telomerase reverse transcriptase promoter (p TERT) is important since preoperative diagnosis of p TERT status helps with evaluating prognosis and determining the surgical strategy. Here, we aimed to establish a radiomics-based machine-learning algorithm and evaluated its performance with regard to the prediction of mutations in p TERT in patients with World Health Organization (WHO) grade II gliomas. In total, 164 patients with WHO grade II gliomas were enrolled in this retrospective study. We extracted a total of 1,293 radiomics features from multi-parametric magnetic resonance imaging scans. Elastic net (used for feature selection) and support vector machine with linear kernel were applied in nested 10-fold cross-validation loops. The predictive model was evaluated by receiver operating characteristic and precision-recall analyses. We performed an unpaired t-test to compare the posterior predictive probabilities among patients with differing p TERT statuses. We selected 12 valuable radiomics features using nested 10-fold cross-validation loops. The area under the curve (AUC) was 0.8446 (95% confidence interval [CI], 0.7735–0.9065) with an optimal summed value of sensitivity of 0.9355 (95% CI, 0.8802–0.9788) and specificity of 0.6197 (95% CI, 0.5071–0.7371). The overall accuracy was 0.7988 (95% CI, 0.7378–0.8598). The F1-score was 0.8406 (95% CI, 0.7684–0.902) with an optimal precision of 0.7632 (95% CI, 0.6818–0.8364) and recall of 0.9355 (95% CI, 0.8802–0.9788). Posterior probabilities of p TERT mutations were significantly different between patients with wild-type and mutant TERT promoters. Our findings suggest that a radiomics analysis with a machine-learning algorithm can be useful for predicting p TERT status in patients with WHO grade II glioma and may aid in glioma management. [ABSTRACT FROM AUTHOR]

We constructed an optimal machine learning (ML) method for predicting intravenous immunoglobulin (IVIG) resistance in children with Kawasaki disease (KD) using commonly available clinical and laboratory variables. We retrospectively collected 98 clinical records of hospitalized children with KD (2–109 months of age). We found that 20 (20%) children were resistant to initial IVIG therapy. We trained three ML techniques, including logistic regression, linear support vector machine, and eXtreme gradient boosting with 10 variables against IVIG resistance. Moreover, we estimated the predictive performance based on nested 5-fold cross-validation (CV). We also selected variables using the recursive feature elimination method and performed the nested 5-fold CV with selected variables in a similar manner. We compared ML models with the existing system regardless of their predictive performance. Results of the area under the receiver operator characteristic curve were in the range of 0.58–0.60 in the all-variable model and 0.60–0.75 in the select model. The specificities were more than 0.90 and higher than those in existing scoring systems, but the sensitivities were lower. Three ML models based on demographics and routine laboratory variables did not provide reliable performance. This is possibly the first study that has attempted to establish a better predictive model. Additional biomarkers are probably needed to generate an effective prediction model.

Applications of machine-learning-based approaches in the geosciences have witnessed a substantial increase over the past few years. Here we present an approach that accounts for spatial autocorrelation by introducing spatial features to the models. In particular, we explore two types of spatial features, namely spatial lag and eigenvector spatial filtering (ESF). These features are used within the widely used random forest (RF) method, and their effect is illustrated on two public datasets of varying sizes (Meuse and California housing datasets). The least absolute shrinkage and selection operator (LASSO) is used to determine the best subset of spatial features, and nested cross-validation is used for hyper-parameter tuning and performance evaluation. We utilize Moran’s I and local indicators of spatial association (LISA) to assess how spatial autocorrelation is captured at both global and local scales. Our results show that RF models combined with either spatial lag or ESF features yield lower errors (up to 33% different) and reduce the global spatial autocorrelation of the residuals (up to 95% decrease in Moran’s I) compared to the RF model with no spatial features. The local autocorrelation patterns of the residuals are weakened as well. Compared to benchmark geographically weighted regression (GWR) models, the RF models with spatial features yielded more accurate models with similar levels of global and local autocorrelation in the prediction residuals. This study reveals the effectiveness of spatial features in capturing spatial autocorrelation and provides a generic machine-learning modelling workflow for spatial prediction.

With the rapid spreading of in-vehicle information systems such as smartphones, navigation systems, and radios, the number of traffic accidents caused by driver distractions shows an increasing trend. Timely identification and warning are deemed to be crucial for distracted driving and the establishment of driver assistance systems is of great value. However, almost all research on the recognition of the driver’s distracted actions using computer vision methods neglected the importance of temporal information for action recognition. This paper proposes a hybrid deep learning model for recognizing the actions of distracted drivers. Specifically, we used OpenPose to obtain skeleton information of the human body and then constructed the vector angle and modulus ratio of the human body structure as features to describe the driver’s actions, thereby realizing the fusion of deep network features and artificial features, which improve the information density of spatial features. The K-means clustering algorithm was used to preselect the original frames, and the method of inter-frame comparison was used to obtain the final keyframe sequence by comparing the Euclidean distance between manually constructed vectors representing frames and the vector representing the cluster center. Finally, we constructed a two-layer long short-term memory neural network to obtain more effective spatiotemporal features, and one softmax layer to identify the distracted driver’s action. The experimental results based on the collected dataset prove the effectiveness of this framework, and it can provide a theoretical basis for the establishment of vehicle distraction warning systems.

Rheumatic heart disease (RHD) is one of the most common causes of cardiovascular complications in developing countries. It is a heart valve disease that typically affects children. Impaired heart valves stop functioning properly, resulting in a turbulent blood flow within the heart known as a murmur. This murmur can be detected by cardiac auscultation. However, the specificity and sensitivity of manual auscultation were reported to be low. The other alternative is echocardiography, which is costly and requires a highly qualified physician. Given the disease’s current high prevalence rate (the latest reported rate in the study area (Ethiopia) was 5.65%), there is a pressing need for early detection of the disease through mass screening programs. This paper proposes an automated RHD screening approach using machine learning that can be used by non-medically trained persons outside of a clinical setting. Heart sound data was collected from 124 persons with RHD (PwRHD) and 46 healthy controls (HC) in Ethiopia with an additional 81 HC records from an open-access dataset. Thirty-one distinct features were extracted to correctly represent RHD. A support vector machine (SVM) classifier was evaluated using two nested cross-validation approaches to quantitatively assess the generalization of the system to previously unseen subjects. For regular nested 10-fold cross-validation, an f1-score of 96.0 ± 0.9%, recall 95.8 ± 1.5%, precision 96.2 ± 0.6% and a specificity of 96.0 ± 0.6% were achieved. In the imbalanced nested cross-validation at a prevalence rate of 5%, it achieved an f1-score of 72.2 ± 0.8%, recall 92.3 ± 0.4%, precision 59.2 ± 3.6%, and a specificity of 94.8 ± 0.6%. In screening tasks where the prevalence of the disease is small, recall is more important than precision. The findings are encouraging, and the proposed screening tool can be inexpensive, easy to deploy, and has an excellent detection rate. As a result, it has the potential for mass screening and early detection of RHD in developing countries.

Keywords: area under curve; egg allergy; logistic regression; machine learning; nested cross-validation; oral food challenge; predictive model; support vector machine EN area under curve egg allergy logistic regression machine learning nested cross-validation oral food challenge predictive model support vector machine 776 778 3 05/04/21 20210501 NES 210501 To the Editor. We identified a patient as OFC positive if he or she experienced an immediate allergic reaction, which included either one clear objective sign or multiple persistent subjective symptoms4 in response to ingestion of less than one half of one egg or 16 g of EW. Serum levels of allergen-specific IgE (sIgE) are among the tools used to diagnose hen's egg allergy. Area under curve, egg allergy, logistic regression, machine learning, nested cross-validation, oral food challenge, predictive model, support vector machine. [Extracted from the article]

Drug-induced liver injury (DILI) remains one of the challenges in the safety profile of both authorized and candidate drugs, and predicting hepatotoxicity from the chemical structure of a substance remains a task worth pursuing. Such an approach is coherent with the current tendency for replacing non-clinical tests with in vitro or in silico alternatives. In 2016, a group of researchers from the FDA published an improved annotated list of drugs with respect to their DILI risk, constituting “the largest reference drug list ranked by the risk for developing drug-induced liver injury in humans” (DILIrank). This paper is one of the few attempting to predict liver toxicity using the DILIrank dataset. Molecular descriptors were computed with the Dragon 7.0 software, and a variety of feature selection and machine learning algorithms were implemented in the R computing environment. Nested (double) cross-validation was used to externally validate the models selected. A total of 78 models with reasonable performance were selected and stacked through several approaches, including the building of multiple meta-models. The performance of the stacked models was slightly superior to other models published. The models were applied in a virtual screening exercise on over 100,000 compounds from the ZINC database and about 20% of them were predicted to be non-hepatotoxic.

An approach for using lasso (Least Absolute Shrinkage and Selection Operator) regression in creating sparse 3D models of soil properties for spatial prediction at multiple depths is presented. Modeling soil properties in 3D benefits from interactions of spatial predictors with soil depth and its polynomial expansion, which yields a large number of model variables (and corresponding model parameters). Lasso is able to perform variable selection, hence reducing the number of model parameters and making the model more easily interpretable. This also prevents overfitting, which makes the model more accurate. The presented approach was tested using four variable selection approaches – none, stepwise, lasso and hierarchical lasso, on four kinds of models – standard linear model, linear model with polynomial expansion of depth, linear model with interactions of covariates with depth and linear model with interactions of covariates with depth and its polynomial expansion. This framework was used to predict Soil Organic Carbon (SOC) in three contrasting study areas: Bor (Serbia), Edgeroi (Australia) and the Netherlands. Results show that lasso yields substantial improvements in accuracy over standard and stepwise regression — up to 50 % of total variance. It yields models which contain up to five times less nonzero parameters than the full models and that are usually more sparse than models obtained by stepwise regression, up to three times. Extension of the standard linear model by including interactions typically improves the accuracy of models produced by lasso, but is detrimental to standard and stepwise regression. Regarding computation time, it was demonstrated that lasso is several orders of magnitude more efficient than stepwise regression for models with tens or hundreds of variables (including interactions). Proper model evaluation is emphasized. Considering the fact that lasso requires meta-parameter tuning, standard cross-validation does not suffice for adequate model evaluation, hence a nested cross-validation was employed. The presented approach is implemented as publicly available sparsereg3D R package. [ABSTRACT FROM AUTHOR]

Vaccination is an effective prevention of influenza infection. However, certain individuals develop a lower antibody response after vaccination, which may lead to susceptibility to subsequent infection. An important challenge in human health is to find baseline gene signatures to help identify individuals who are at higher risk for infection despite influenza vaccination. We developed a multi-level machine learning strategy to build a predictive model of vaccine response using pre−vaccination antibody titers and network interactions between pre−vaccination gene expression levels. The first-level baseline−antibody model explains a significant amount of variation in post-vaccination response, especially for subjects with large pre−existing antibody titers. In the second level, we clustered individuals based on pre−vaccination antibody titers to focus gene−based modeling on individuals with lower baseline HAI where additional response variation may be predicted by baseline gene expression levels. In the third level, we used a gene−association interaction network (GAIN) feature selection algorithm to find the best pairs of genes that interact to influence antibody response within each baseline titer cluster. We used ratios of the top interacting genes as predictors to stabilize machine learning model generalizability. We trained and tested the multi-level approach on data with young and older individuals immunized against influenza vaccine in multiple cohorts. Our results indicate that the GAIN feature selection approach improves model generalizability and identifies genes enriched for immunologically relevant pathways, including B Cell Receptor signaling and antigen processing. Using a multi-level approach, starting with a baseline HAI model and stratifying on baseline HAI, allows for more targeted gene−based modeling. We provide an interactive tool that may be extended to other vaccine studies.

Objective: There is a need to monitor the heart rhythm in resuscitation to improve treatment quality. Resuscitation rhythms are categorized into: ventricular tachycardia (VT), ventricular fibrillation (VF), pulseless electrical activity (PEA), asystole (AS), and pulse-generating rhythm (PR). Manual annotation of rhythms is time-consuming and infeasible for large datasets. Our objective was to develop ECG-based algorithms for the retrospective and automatic classification of resuscitation cardiac rhythms. Methods: The dataset consisted of 1631 3-s ECG segments with clinical rhythm annotations, obtained from 298 out-of-hospital cardiac arrest patients. In total, 47 wavelet- and time-domain-based features were computed from the ECG. Features were selected using a wrapper-based feature selection architecture. Classifiers based on Bayesian decision theory, k-nearest neighbor, k-local hyperplane distance nearest neighbor, artificial neural network (ANN), and ensemble of decision trees were studied. Results: The best results were obtained for ANN classifier with Bayesian regularization backpropagation training algorithm with 14 features, which forms the proposed algorithm. The overall accuracy for the proposed algorithm was 78.5%. The sensitivities (and positive-predictive-values) for AS, PEA, PR, VF, and VT were 88.7% (91.0%), 68.9% (70.4%), 65.9% (69.0%), 86.2% (83.8%), and 78.8% (72.9%), respectively. Conclusions: The results demonstrate that it is possible to classify resuscitation cardiac rhythms automatically, but the accuracy for the organized rhythms (PEA and PR) is low. Significance: We have made an important step toward making classification of resuscitation rhythms more efficient in the sense of minimal feedback from human experts. [ABSTRACT FROM AUTHOR]

Prediction problems in biomedical sciences, including protein function prediction (PFP), are generally quite difficult. This is due in part to incomplete knowledge of the cellular phenomenon of interest, the appropriateness and data quality of the variables and measurements used for prediction, as well as a lack of consensus regarding the ideal predictor for specific problems. In such scenarios, a powerful approach to improving prediction performance is to construct heterogeneous ensemble predictors that combine the output of diverse individual predictors that capture complementary aspects of the problems and/or datasets. In this paper, we demonstrate the potential of such heterogeneous ensembles, derived from stacking and ensemble selection methods, for addressing PFP and other similar biomedical prediction problems. Deeper analysis of these results shows that the superior predictive ability of these methods, especially stacking, can be attributed to their attention to the following aspects of the ensemble learning process: (i) better balance of diversity and performance, (ii) more effective calibration of outputs and (iii) more robust incorporation of additional base predictors. Finally, to make the effective application of heterogeneous ensembles to large complex datasets ( big data ) feasible, we present DataSink , a distributed ensemble learning framework, and demonstrate its sound scalability using the examined datasets. DataSink is publicly available from https://github.com/shwhalen/datasink . [ABSTRACT FROM AUTHOR]

A number of supervised machine learning models have recently been introduced for the prediction of drug-target interactions based on chemical structure and genomic sequence information. Although these models could offer improved means for many network pharmacology applications, such as repositioning of drugs for new therapeutic uses, the prediction models are often being constructed and evaluated under overly simplified settings that do not reflect the real-life problem in practical applications. Using quantitative drug-target bioactivity assays for kinase inhibitors, as well as a popular benchmarking data set of binary drug-target interactions for enzyme, ion channel, nuclear receptor and G protein-coupled receptor targets, we illustrate here the effects of four factors that may lead to dramatic differences in the prediction results: (i) problem formulation (standard binary classification or more realistic regression formulation), (ii) evaluation data set (drug and target families in the application use case), (iii) evaluation procedure (simple or nested cross-validation) and (iv) experimental setting (whether training and test sets share common drugs and targets, only drugs or targets or neither). Each of these factors should be taken into consideration to avoid reporting overoptimistic drug-target interaction prediction results. We also suggest guidelines on how to make the supervised drug-target interaction prediction studies more realistic in terms of such model formulations and evaluation setups that better address the inherent complexity of the prediction task in the practical applications, as well as novel benchmarking data sets that capture the continuous nature of the drug-target interactions for kinase inhibitors. [ABSTRACT FROM AUTHOR]

Objective. The aim of the present study was to elucidate the brain dynamics underlying the maintenance of a constant force level exerted during a visually guided isometric contraction task by optimizing a predictive multivariate model based on global and spectral brain dynamics features. Approach. Electroencephalography (EEG) was acquired in 18 subjects who were asked to press a bulb and maintain a constant force level, indicated by a bar on a screen. For intervals of 500 ms, we calculated an index of force stability as well as indices of brain dynamics: microstate metrics (duration, occurrence, global explained variance, directional predominance) and EEG spectral amplitudes in the theta, low alpha, high alpha and beta bands. We optimized a multivariate regression model (partial least square (PLS)) where the microstate features and the spectral amplitudes were the input variables and the indexes of force stability were the output variables. The issues related to the collinearity among the input variables and to the generalizability of the model were addressed using PLS in a nested cross-validation approach. Main results. The optimized PLS regression model reached a good generalizability and succeeded to show the predictive value of microstates and spectral features in inferring the stability of the exerted force. Longer duration and higher occurrence of microstates, associated with visual and executive control networks, corresponded to better contraction performances, in agreement with the role played by the visual system and executive control network for visuo-motor integration. Significance. A combination of microstate metrics and brain rhythm amplitudes could be considered as biomarkers of a stable visually guided motor output not only at a group level, but also at an individual level. Our results may play an important role for a better understanding of the motor control in single trials or in real-time applications as well as in the study of motor control., (Creative Commons Attribution license.)

When selecting a classification algorithm to be applied to a particular problem, one has to simultaneously select the best algorithm for that dataset and the best set of hyperparameters for the chosen model. The usual approach is to apply a nested cross-validation procedure: hyperparameter selection is performed in the inner cross-validation, while the outer cross-validation computes an unbiased estimate of the expected accuracy of the algorithm with cross-validation based hyperparameter tuning. The alternative approach, which we shall call "flat cross-validation", uses a single cross-validation step both to select the optimal hyperparameter values and to provide an estimate of the expected accuracy of the algorithm that, while biased, may nevertheless still be used to select the best learning algorithm. We tested both procedures using 12 different algorithms on 115 real-life binary datasets and conclude that using the less computationally costly flat cross-validation procedure will generally result in the selection of an algorithm that is, for all practical purposes, of similar quality to that selected via nested cross-validation, provided the learning algorithms have relatively few hyperparameters to be optimised. • Flat cross validation computes both the best hyperparameter and the expected accuracy. • Nested cross validation separated both computations and it is more costly. • Nested cross validation does not incur on biased estimation of the accuracy. • Algorithm selection using flat cross validation does not incur in worse selections. [ABSTRACT FROM AUTHOR]

With the rapid spreading of in-vehicle information systems such as smartphones, navigation systems, and radios, the number of traffic accidents caused by driver distractions shows an increasing trend. Timely identification and warning are deemed to be crucial for distracted driving and the establishment of driver assistance systems is of great value. However, almost all research on the recognition of the driver's distracted actions using computer vision methods neglected the importance of temporal information for action recognition. This paper proposes a hybrid deep learning model for recognizing the actions of distracted drivers. Specifically, we used OpenPose to obtain skeleton information of the human body and then constructed the vector angle and modulus ratio of the human body structure as features to describe the driver's actions, thereby realizing the fusion of deep network features and artificial features, which improve the information density of spatial features. The K-means clustering algorithm was used to preselect the original frames, and the method of inter-frame comparison was used to obtain the final keyframe sequence by comparing the Euclidean distance between manually constructed vectors representing frames and the vector representing the cluster center. Finally, we constructed a two-layer long short-term memory neural network to obtain more effective spatiotemporal features, and one softmax layer to identify the distracted driver's action. The experimental results based on the collected dataset prove the effectiveness of this framework, and it can provide a theoretical basis for the establishment of vehicle distraction warning systems. [ABSTRACT FROM AUTHOR]

Rheumatic heart disease (RHD) is one of the most common causes of cardiovascular complications in developing countries. It is a heart valve disease that typically affects children. Impaired heart valves stop functioning properly, resulting in a turbulent blood flow within the heart known as a murmur. This murmur can be detected by cardiac auscultation. However, the specificity and sensitivity of manual auscultation were reported to be low. The other alternative is echocardiography, which is costly and requires a highly qualified physician. Given the disease's current high prevalence rate (the latest reported rate in the study area (Ethiopia) was 5.65%), there is a pressing need for early detection of the disease through mass screening programs. This paper proposes an automated RHD screening approach using machine learning that can be used by non-medically trained persons outside of a clinical setting. Heart sound data was collected from 124 persons with RHD (PwRHD) and 46 healthy controls (HC) in Ethiopia with an additional 81 HC records from an open-access dataset. Thirty-one distinct features were extracted to correctly represent RHD. A support vector machine (SVM) classifier was evaluated using two nested cross-validation approaches to quantitatively assess the generalization of the system to previously unseen subjects. For regular nested 10-fold cross-validation, an f1-score of 96.0 ± 0.9%, recall 95.8 ± 1.5%, precision 96.2 ± 0.6% and a specificity of 96.0 ± 0.6% were achieved. In the imbalanced nested cross-validation at a prevalence rate of 5%, it achieved an f1-score of 72.2 ± 0.8%, recall 92.3 ± 0.4%, precision 59.2 ± 3.6%, and a specificity of 94.8 ± 0.6%. In screening tasks where the prevalence of the disease is small, recall is more important than precision. The findings are encouraging, and the proposed screening tool can be inexpensive, easy to deploy, and has an excellent detection rate. As a result, it has the potential for mass screening and early detection of RHD in developing countries. [ABSTRACT FROM AUTHOR]