Fast Gaussian Naïve Bayes for searchlight classification analysis

Industrial Fault Detection Employing Meta Ensemble Model Based on Contact Sensor Ultrasonic Signal

Ultrasonic diagnostics is the earliest way to predict industrial faults. Usually, a contact microphone is employed for detection, but the recording will be contaminated with noise. In this paper, a dataset that contains 10 main faults of pipelines and motors is analyzed from which 30 different features in the time and frequency domains are extracted. Afterward, for dimensionality reduction, principal component analysis (PCA), linear discriminant analysis (LDA), and t-distributed stochastic neighbor embedding (t-SNE) are performed. In the subsequent phase, recursive feature elimination (RFE) is employed as a strategic method to analyze and select the most relevant features for the classifiers. Next, predictive models consisting of k-Nearest Neighbor (KNN), Logistic Regression (LR), Decision Tree (DT), Gaussian Naive Bayes (GNB), and Support Vector Machine (SVM) are employed. Then, in order to solve the classification problem, a stacking classifier based on a meta-classifier which combines multiple classification models is introduced. Furthermore, the k-fold cross-validation technique is employed to assess the effectiveness of the model in handling new data for the evaluation of experimental results in ultrasonic fault detection. With the proposed method, the accuracy is around 5% higher over five cross folds with the least amount of variation. The timing evaluation of the meta model on the 64 MHz Cortex M4 microcontroller unit (MCU) revealed an execution time of 11 ms, indicating it could be a promising solution for real-time monitoring.

Approaching Personalized Medicine: The Use of Machine Learning to Determine Predictors of Mortality in a Population with SARS-CoV-2 Infection

The COVID-19 pandemic demonstrated the need to develop strategies to control a new viral infection. However, the different characteristics of the health system and population of each country and hospital would require the implementation of self-systems adapted to their characteristics. The objective of this work was to determine predictors that should identify the most severe patients with COVID-19 infection. Given the poor situation of the hospitals in the first wave, the analysis of the data from that period with an accurate and fast technique can be an important contribution. In this regard, machine learning is able to objectively analyze data in hourly sets and is used in many fields. This study included 291 patients admitted to a hospital in Spain during the first three months of the pandemic. After screening seventy-one features with machine learning methods, the variables with the greatest influence on predicting mortality in this population were lymphocyte count, urea, FiO2, potassium, and serum pH. The XGB method achieved the highest accuracy, with a precision of >95%. Our study shows that the machine learning-based system can identify patterns and, thus, create a tool to help hospitals classify patients according to their severity of illness in order to optimize admission.

A machine learning-based prediction model for postoperative delirium in cardiac valve surgery using electronic health records

BACKGROUND

Previous models for predicting delirium after cardiac surgery remained inadequate. This study aimed to develop and validate a machine learning-based prediction model for postoperative delirium (POD) in cardiac valve surgery patients.

METHODS

The electronic medical information of the cardiac surgical intensive care unit (CSICU) was extracted from a tertiary and major referral hospital in southern China over 1 year, from June 2019 to June 2020. A total of 507 patients admitted to the CSICU after cardiac valve surgery were included in this study. Seven classical machine learning algorithms (Random Forest Classifier, Logistic Regression, Support Vector Machine Classifier, K-nearest Neighbors Classifier, Gaussian Naive Bayes, Gradient Boosting Decision Tree, and Perceptron.) were used to develop delirium prediction models under full (q = 31) and selected (q = 19) feature sets, respectively.

RESULT

The Random Forest classifier performs exceptionally well in both feature datasets, with an Area Under the Curve (AUC) of 0.92 for the full feature dataset and an AUC of 0.86 for the selected feature dataset. Additionally, it achieves a relatively lower Expected Calibration Error (ECE) and the highest Average Precision (AP), with an AP of 0.80 for the full feature dataset and an AP of 0.73 for the selected feature dataset. To further evaluate the best-performing Random Forest classifier, SHAP (Shapley Additive Explanations) was used, and the importance matrix plot, scatter plots, and summary plots were generated.

CONCLUSIONS

We established machine learning-based prediction models to predict POD in patients undergoing cardiac valve surgery. The random forest model has the best predictive performance in prediction and can help improve the prognosis of patients with POD.

Analyzing patients satisfaction level for medical services using twitter data

Public concern regarding health systems has experienced a rapid surge during the last two years due to the COVID-19 outbreak. Accordingly, medical professionals and health-related institutions reach out to patients and seek feedback to analyze, monitor, and uplift medical services. Such views and perceptions are often shared on social media platforms like Facebook, Instagram, Twitter, etc. Twitter is the most popular and commonly used by the researcher as an online platform for instant access to real-time news, opinions, and discussion. Its trending hashtags (#) and viral content make it an ideal hub for monitoring public opinion on a variety of topics. The tweets are extracted using three hashtags #healthcare, #healthcare services, and #medical facilities. Also, location and tweet sentiment analysis are considered in this study. Several recent studies deployed Twitter datasets using ML and DL models, but the results show lower accuracy. In addition, the studies did not perform extensive comparative analysis and lack validation. This study addresses two research questions: first, what are the sentiments of people toward medical services worldwide? and second, how effective are the machine learning and deep learning approaches for the classification of sentiment on healthcare tweets? Experiments are performed using several well-known machine learning models including support vector machine, logistic regression, Gaussian naive Bayes, extra tree classifier, k nearest neighbor, random forest, decision tree, and AdaBoost. In addition, this study proposes a transfer learning-based LSTM-ETC model that effectively predicts the customer's satisfaction level from the healthcare dataset. Results indicate that despite the best performance by the ETC model with an 0.88 accuracy score, the proposed model outperforms with a 0.95 accuracy score. Predominantly, the people are happy about the provided medical services as the ratio of the positive sentiments is substantially higher than the negative sentiments. The sentiments, either positive or negative, play a crucial role in making important decisions through customer feedback and enhancing quality.

Familiarity Processing through Faces and Names: Insights from Multivoxel Pattern Analysis

The way our brain processes personal familiarity is still debatable. We used searchlight multivoxel pattern analysis (MVPA) to identify areas where local fMRI patterns could contribute to familiarity detection for both faces and name categories. Significantly, we identified cortical areas in frontal, temporal, cingulate, and insular areas, where it is possible to accurately cross-classify familiar stimuli from one category using a classifier trained with the stimulus from the other (i.e., abstract familiarity) based on local fMRI patterns. We also discovered several areas in the fusiform gyrus, frontal, and temporal regions-primarily lateralized to the right hemisphere-supporting the classification of familiar faces but failing to do so for names. Also, responses to familiar names (compared to unfamiliar names) consistently showed less activation strength than responses to familiar faces (compared to unfamiliar faces). The results evinced a set of abstract familiarity areas (independent of the stimulus type) and regions specifically related only to face familiarity, contributing to recognizing familiar individuals.

Machine learning algorithms for integrating clinical features to predict intracranial hemorrhage in patients with acute leukemia

BACKGROUND

Intracranial hemorrhage (ICH) in acute leukemia (AL) patients leads to high morbidity and mortality, treatment approaches for ICH are generally ineffective. Thus, early identification of which subjects are at high risk of ICH is of key importance. Currently, machine learning can achieve well predictive capability through constructing algorithms that simultaneously exploit the information coming from clinical features.

METHODS

After rigid data preprocessing, 42 different clinical features from 948 AL patients were used to train different machine learning algorithms. We used the feature selection algorithms to select the top 10 features from 42 clinical features. To test the performance of the machine learning algorithms, we calculated area under the curve (AUC) values from receiver operating characteristic (ROC) curves along with 95% confidence intervals (CIs) by cross-validation.

RESULTS

With the 42 features, RF exhibited the best predictive power. After feature selection, the top 10 features were international normalized ratio (INR), prothrombin time (PT), creatinine (Cr), indirect bilirubin (IBIL), albumin (ALB), monocyte (MONO), platelet (PLT), lactic dehydrogenase (LDH), fibrinogen (FIB) and prealbumin (PA). Among the top 10 features, INR, PT, Cr, IBIL and ALB had high predictive performance with an AUC higher than 0.8 respectively.

CONCLUSIONS

The RF algorithm exhibited a higher cross-validated performance compared with the classical algorithms, and the selected important risk features should help in individualizing aggressive treatment in AL patients to prevent ICH. Efforts that will be made to test and optimize in independent samples will warrant the application of such algorithm and predictors in the future.

Machine learning-based models for the prediction of breast cancer recurrence risk

Breast cancer is the most common malignancy diagnosed in women worldwide. The prevalence and incidence of breast cancer is increasing every year; therefore, early diagnosis along with suitable relapse detection is an important strategy for prognosis improvement. This study aimed to compare different machine algorithms to select the best model for predicting breast cancer recurrence. The prediction model was developed by using eleven different machine learning (ML) algorithms, including logistic regression (LR), random forest (RF), support vector classification (SVC), extreme gradient boosting (XGBoost), gradient boosting decision tree (GBDT), decision tree, multilayer perceptron (MLP), linear discriminant analysis (LDA), adaptive boosting (AdaBoost), Gaussian naive Bayes (GaussianNB), and light gradient boosting machine (LightGBM), to predict breast cancer recurrence. The area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and F1 score were used to evaluate the performance of the prognostic model. Based on performance, the optimal ML was selected, and feature importance was ranked by Shapley Additive Explanation (SHAP) values. Compared to the other 10 algorithms, the results showed that the AdaBoost algorithm had the best prediction performance for successfully predicting breast cancer recurrence and was adopted in the establishment of the prediction model. Moreover, CA125, CEA, Fbg, and tumor diameter were found to be the most important features in our dataset to predict breast cancer recurrence. More importantly, our study is the first to use the SHAP method to improve the interpretability of clinicians to predict the recurrence model of breast cancer based on the AdaBoost algorithm. The AdaBoost algorithm offers a clinical decision support model and successfully identifies the recurrence of breast cancer.

A clinical prediction model based on interpretable machine learning algorithms for prolonged hospital stay in acute ischemic stroke patients: a real-world study

Objective

Acute ischemic stroke (AIS) brings an increasingly heavier economic burden nowadays. Prolonged length of stay (LOS) is a vital factor in healthcare expenditures. The aim of this study was to predict prolonged LOS in AIS patients based on an interpretable machine learning algorithm.

Methods

We enrolled AIS patients in our hospital from August 2017 to July 2019, and divided them into the "prolonged LOS" group and the "no prolonged LOS" group. Prolonged LOS was defined as hospitalization for more than 7 days. The least absolute shrinkage and selection operator (LASSO) regression was applied to reduce the dimensionality of the data. We compared the predictive capacity of extended LOS in eight different machine learning algorithms. SHapley Additive exPlanations (SHAP) values were used to interpret the outcome, and the most optimal model was assessed by discrimination, calibration, and clinical utility.

Results

Prolonged LOS developed in 149 (22.0%) of the 677 eligible patients. In eight machine learning algorithms, prolonged LOS was best predicted by the Gaussian naive Bayes (GNB) model, which had a striking area under the curve (AUC) of 0.878 ± 0.007 in the training set and 0.857 ± 0.039 in the validation set. The variables sorted by the gap values showed that the strongest predictors were pneumonia, dysphagia, thrombectomy, and stroke severity. High net benefits were observed at 0%-76% threshold probabilities, while good agreement was found between the observed and predicted probabilities.

Conclusions

The model using the GNB algorithm proved excellent for predicting prolonged LOS in AIS patients. This simple model of prolonged hospitalization could help adjust policies and better utilize resources.

Modality-specific brain representations during automatic processing of face, voice and body expressions

A central question in affective science and one that is relevant for its clinical applications is how emotions provided by different stimuli are experienced and represented in the brain. Following the traditional view emotional signals are recognized with the help of emotion concepts that are typically used in descriptions of mental states and emotional experiences, irrespective of the sensory modality. This perspective motivated the search for abstract representations of emotions in the brain, shared across variations in stimulus type (face, body, voice) and sensory origin (visual, auditory). On the other hand, emotion signals like for example an aggressive gesture, trigger rapid automatic behavioral responses and this may take place before or independently of full abstract representation of the emotion. This pleads in favor specific emotion signals that may trigger rapid adaptative behavior only by mobilizing modality and stimulus specific brain representations without relying on higher order abstract emotion categories. To test this hypothesis, we presented participants with naturalistic dynamic emotion expressions of the face, the whole body, or the voice in a functional magnetic resonance (fMRI) study. To focus on automatic emotion processing and sidestep explicit concept-based emotion recognition, participants performed an unrelated target detection task presented in a different sensory modality than the stimulus. By using multivariate analyses to assess neural activity patterns in response to the different stimulus types, we reveal a stimulus category and modality specific brain organization of affective signals. Our findings are consistent with the notion that under ecological conditions emotion expressions of the face, body and voice may have different functional roles in triggering rapid adaptive behavior, even if when viewed from an abstract conceptual vantage point, they may all exemplify the same emotion. This has implications for a neuroethologically grounded emotion research program that should start from detailed behavioral observations of how face, body, and voice expressions function in naturalistic contexts.

SoyDNGP: a web-accessible deep learning framework for genomic prediction in soybean breeding

Soybean is a globally significant crop, playing a vital role in human nutrition and agriculture. Its complex genetic structure and wide trait variation, however, pose challenges for breeders and researchers aiming to optimize its yield and quality. Addressing this biological complexity requires innovative and accurate tools for trait prediction. In response to this challenge, we have developed SoyDNGP, a deep learning-based model that offers significant advancements in the field of soybean trait prediction. Compared to existing methods, such as DeepGS and DNNGP, SoyDNGP boasts a distinct advantage due to its minimal increase in parameter volume and superior predictive accuracy. Through rigorous performance comparison, including prediction accuracy and model complexity, SoyDNGP represents improved performance to its counterparts. Furthermore, it effectively predicted complex traits with remarkable precision, demonstrating robust performance across different sample sizes and trait complexities. We also tested the versatility of SoyDNGP across multiple crop species, including cotton, maize, rice and tomato. Our results showed its consistent and comparable performance, emphasizing SoyDNGP's potential as a versatile tool for genomic prediction across a broad range of crops. To enhance its accessibility to users without extensive programming experience, we designed a user-friendly web server, available at http://xtlab.hzau.edu.cn/SoyDNGP. The server provides two features: 'Trait Lookup', offering users the ability to access pre-existing trait predictions for over 500 soybean accessions, and 'Trait Prediction', allowing for the upload of VCF files for trait estimation. By providing a high-performing, accessible tool for trait prediction, SoyDNGP opens up new possibilities in the quest for optimized soybean breeding.

Prediction-Inspired Intelligent Training for the Development of Classification Read-across Structure-Activity Relationship (c-RASAR) Models for Organic Skin Sensitizers: Assessment of Classification Error Rate from Novel Similarity Coefficients

The advancements in the field of cheminformatics have led to a reduction in animal testing to estimate the activity, property, and toxicity of query chemicals. Read-across structure-activity relationship (RASAR) is an emerging concept that utilizes various similarity functions derived from chemical information to develop highly predictive models. Unlike quantitative structure-activity relationship (QSAR) models, RASAR descriptors of a query compound are computed from its close congeners instead of the compound itself, thus targeting predictions in the model training phase. The objective of the present study is not to propose new QSAR models for skin sensitization but to demonstrate the enhancement in the quality of predictions of the skin-sensitizing potential of organic compounds by developing classification-based RASAR (c-RASAR) models. A diverse, previously curated data set was collected from the literature for which 2D descriptors were computed. The extracted essential features were then used to develop a classification-based linear discriminant analysis (LDA) QSAR model. Furthermore, from the read-across-based predictions, RASAR descriptors were calculated using the basic settings of the hyperparameters for the Laplacian Kernel-based optimum similarity measure. After feature selection, an LDA c-RASAR model was developed, which superseded the prediction quality of the LDA-QSAR model. Various other combinations of RASAR descriptors were also taken to develop additional c-RASAR models, all showing better prediction quality than the LDA QSAR model while using a lower number of descriptors. Various other machine learning c-RASAR models were also developed for comparison purposes. In this work, we have proposed and analyzed three new similarity metrics: gm_class, sm1, and sm2. The first one is an indicator variable used to generate a simple univariate c-RASAR model with good prediction ability, while the remaining two are similarity indices used to analyze possible activity cliffs in the training and test sets and are believed to play an important role in the modelability analysis of data sets.

Body composition predicts hypertension using machine learning methods: a cohort study

We used machine learning methods to investigate if body composition indices predict hypertension. Data from a cohort study was used, and 4663 records were included (2156 were male, 1099 with hypertension, with the age range of 35-70 years old). Body composition analysis was done using bioelectrical impedance analysis (BIA); weight, basal metabolic rate, total and regional fat percentage (FATP), and total and regional fat-free mass (FFM) were measured. We used machine learning methods such as Support Vector Classifier, Decision Tree, Stochastic Gradient Descend Classifier, Logistic Regression, Gaussian Naïve Bayes, K-Nearest Neighbor, Multi-Layer Perceptron, Random Forest, Gradient Boosting, Histogram-based Gradient Boosting, Bagging, Extra Tree, Ada Boost, Voting, and Stacking to classify the investigated cases and find the most relevant features to hypertension. FATP, AFFM, BMR, FFM, TRFFM, AFATP, LFATP, and older age were the top features in hypertension prediction. Arm FFM, basal metabolic rate, total FFM, Trunk FFM, leg FFM, and male gender were inversely associated with hypertension, but total FATP, arm FATP, leg FATP, older age, trunk FATP, and female gender were directly associated with hypertension. AutoMLP, stacking and voting methods had the best performance for hypertension prediction achieving an accuracy rate of 90%, 84% and 83%, respectively. By using machine learning methods, we found that BIA-derived body composition indices predict hypertension with acceptable accuracy.

A Deep Learning Approach for Predicting Multiple Sclerosis

This paper proposes a deep learning model based on an artificial neural network with a single hidden layer for predicting the diagnosis of multiple sclerosis. The hidden layer includes a regularization term that prevents overfitting and reduces the model complexity. The purposed learning model achieved higher prediction accuracy and lower loss than four conventional machine learning techniques. A dimensionality reduction method was used to select the most relevant features from 74 gene expression profiles for training the learning models. The analysis of variance test was performed to identify the statistical difference between the mean of the proposed model and the compared classifiers. The experimental results show the effectiveness of the proposed artificial neural network.

Robust Classification Using Posterior Probability Threshold Computation Followed by Voronoi Cell Based Class Assignment Circumventing Pitfalls of Bayesian Analysis of Biomedical Data

Bayesian inference is ubiquitous in science and widely used in biomedical research such as cell sorting or “omics” approaches, as well as in machine learning (ML), artificial neural networks, and “big data” applications. However, the calculation is not robust in regions of low evidence. In cases where one group has a lower mean but a higher variance than another group, new cases with larger values are implausibly assigned to the group with typically smaller values. An approach for a robust extension of Bayesian inference is proposed that proceeds in two main steps starting from the Bayesian posterior probabilities. First, cases with low evidence are labeled as “uncertain” class membership. The boundary for low probabilities of class assignment (threshold ε) is calculated using a computed ABC analysis as a data-based technique for item categorization. This leaves a number of cases with uncertain classification (p < ε). Second, cases with uncertain class membership are relabeled based on the distance to neighboring classified cases based on Voronoi cells. The approach is demonstrated on biomedical data typically analyzed with Bayesian statistics, such as flow cytometric data sets or biomarkers used in medical diagnostics, where it increased the class assignment accuracy by 1−10% depending on the data set. The proposed extension of the Bayesian inference of class membership can be used to obtain robust and plausible class assignments even for data at the extremes of the distribution and/or for which evidence is weak.

Pleasantness Recognition Induced by Different Odor Concentrations Using Olfactory Electroencephalogram Signals

Olfactory-induced emotion plays an important role in communication, decision-making, multimedia, and disorder treatment. Using electroencephalogram (EEG) technology, this paper focuses on (1) exploring the possibility of recognizing pleasantness induced by different concentrations of odors, (2) finding the EEG rhythm wave that is most suitable for the recognition of different odor concentrations, (3) analyzing recognition accuracies with concentration changes, and (4) selecting a suitable classifier for this classification task. To explore these issues, first, emotions induced by five different concentrations of rose or rotten odors are divided into five kinds of pleasantness by averaging subjective evaluation scores. Then, the power spectral density features of EEG signals and support vector machine (SVM) are used for classification tasks. Classification results on the EEG signals collected from 13 participants show that for pleasantness recognition induced by pleasant or disgusting odor concentrations, considerable average classification accuracies of 93.5% or 92.2% are obtained, respectively. The results indicate that (1) using EEG technology, pleasantness recognition induced by different odor concentrations is possible; (2) gamma frequency band outperformed other EEG rhythm-based frequency bands in terms of classification accuracy, and as the maximum frequency of the EEG spectrum increases, the pleasantness classification accuracy gradually increases; (3) for both rose and rotten odors, the highest concentration obtains the best classification accuracy, followed by the lowest concentration.

Forecast and analysis of aircraft passenger satisfaction based on RF-RFE-LR model

Airplanes have always been one of the first choices for people to travel because of their convenience and safety. However, due to the outbreak of the new coronavirus epidemic in 2020, the civil aviation industry of various countries in the world has encountered severe challenges. Predicting aircraft passenger satisfaction and excavating the main influencing factors can help airlines improve their services and gain advantages in difficult situations and competition. This paper proposes a RF-RFE-Logistic feature selection model to extract the influencing factors of passenger satisfaction. First, preliminary feature selection is performed using recursive feature elimination based on random forest (RF-RFE). Second, based on different classification models, KNN, logistic regression, random forest, Gaussian Naive Bayes, and BP neural network, the classification performance of the models before and after feature selection is compared, and the prediction model with the best classification performance is selected. Finally, based on the RF-RFE feature selection, combined with the logistic model, the factors affecting customer satisfaction are further extracted. The experimental results show that the RF-RFE model selects a feature subset containing 17 variables. In the classification prediction model, the random forest after RF-RFE feature selection shows the best classification performance. Finally, combined with the four important variables extracted by RF-RFE and logistic regression, further discussion is carried out, and suggestions are given for airlines to improve passenger satisfaction.

Evading obscure communication from spam emails

Spam is any form of annoying and unsought digital communication sent in bulk and may contain offensive content feasting viruses and cyber-attacks. The voluminous increase in spam has necessitated developing more reliable and vigorous artificial intelligence-based anti-spam filters. Besides text, an email sometimes contains multimedia content such as audio, video, and images. However, text-centric email spam filtering employing text classification techniques remains today's preferred choice. In this paper, we show that text pre-processing techniques nullify the detection of malicious contents in an obscure communication framework. We use Spamassassin corpus with and without text pre-processing and examined it using machine learning (ML) and deep learning (DL) algorithms to classify these as ham or spam emails. The proposed DL-based approach consistently outperforms ML models. In the first stage, using pre-processing techniques, the long-short-term memory (LSTM) model achieves the highest results of 93.46% precision, 96.81% recall, and 95% F1-score. In the second stage, without using pre-processing techniques, LSTM achieves the best results of 95.26% precision, 97.18% recall, and 96% F1-score. Results show the supremacy of DL algorithms over the standard ones in filtering spam. However, the effects are unsatisfactory for detecting encrypted communication for both forms of ML algorithms.

A Novel Feature Selection Technique to Better Predict Climate Change Stage of Change

Indications of people’s environmental concern are linked to transport decisions and can provide great support for policymaking on climate change. This study aims to better predict individual climate change stage of change (CC-SoC) based on different features of transport-related behavior, General Ecological Behavior, New Environmental Paradigm, and socio-demographic characteristics. Together these sources result in over 100 possible features that indicate someone’s level of environmental concern. Such a large number of features may create several analytical problems, such as overfitting, accuracy reduction, and high computational costs. To this end, a new feature selection technique, named the Coyote Optimization Algorithm-Quadratic Discriminant Analysis (COA-QDA), is first proposed to find the optimal features to predict CC-SoC with the highest accuracy. Different conventional feature selection methods (Lasso, Elastic Net, Random Forest Feature Selection, Extra Trees, and Principal Component Analysis Feature Selection) are employed to compare with the COA-QDA. Afterward, eight classification techniques are applied to solve the prediction problem. Finally, a sensitivity analysis is performed to determine the most important features affecting the prediction of CC-SoC. The results indicate that COA-QDA outperforms conventional feature selection methods by increasing average testing data accuracy from 0.7% to 5.6%. Logistic Regression surpasses other classifiers with the highest prediction accuracy.

The diagnosis of amnestic mild cognitive impairment by combining the characteristics of brain functional network and support vector machine classifier

BACKGROUND

Amnestic mild cognitive impairment (aMCI) is an essential stage of early detection and potential intervention for Alzheimer's disease (AD). Patients with aMCI exhibit partially abnormal functional brain connectivity and it is suggested that these features may represent a new diagnostic marker of early AD.

NEW METHOD

In this paper, we constructed two brain network models, a phase synchronization index (PSI) undirected network and a directed transfer function (DTF) directed network, to evaluate the cognitive function in patients with aMCI. We then built SVM classification models using the network clustering coefficient, global efficiency and average node degree as features to distinguish between aMCI patients and controls.

RESULTS

Our results reveal a classification accuracy and AUC of 66.6 ± 1.7% and 0.7475 and 80.0 ± 2.2% and 0.7825, respectively, for the two network models (PSI and DTF). As the directed network model performed better than the undirected model, we introduced an improved graph theory feature, efficiency density, which resulted in an increased classification accuracy and AUC value 86.6 ± 2.6% and 0.8295, respectively.

COMPARISON WITH EXISTING METHODS

The analysis of network models and the directionality of information flow is suitable for analysis of nonlinear EEG signals for assessment of the functional state of the brain. Compared with traditional network features, our proposed improved features more comprehensively evaluate transmission efficiency and density of the brain.

CONCLUSION

In this study, we demonstrate that an improved efficiency density feature is helpful for enhancing classification the accuracy of aMCI. Moreover, directed brain network models exhibit better classification for aMCI diagnosis than undirected networks.

Decoding the difference between explicit and implicit body expression representation in high level visual, prefrontal and inferior parietal cortex

Recent studies provide an increasing understanding of how visual objects categories like faces or bodies are represented in the brain and also raised the question whether a category based or more dynamic network inspired models are more powerful. Two important and so far sidestepped issues in this debate are, first, how major category attributes like the emotional expression directly influence category representation and second, whether category and attribute representation are sensitive to task demands. This study investigated the impact of a crucial category attribute like emotional expression on category area activity and whether this varies with the participants' task. Using (fMRI) we measured BOLD responses while participants viewed whole body expressions and performed either an explicit (emotion) or an implicit (shape) recognition task. Our results based on multivariate methods show that the type of task is the strongest determinant of brain activity and can be decoded in EBA, VLPFC and IPL. Brain activity was higher for the explicit task condition in VLPFC and was not emotion specific. This pattern suggests that during explicit recognition of the body expression, body category representation may be strengthened, and emotion and action related activity suppressed. Taken together these results stress the importance of the task and of the role of category attributes for understanding the functional organization of high level visual cortex.

Cross-validation and permutations in MVPA: Validity of permutation strategies and power of cross-validation schemes

Multi-Voxel Pattern Analysis (MVPA) is a well established tool to disclose weak, distributed effects in brain activity patterns. The generalization ability is assessed by testing the learning model on new, unseen data. However, when limited data is available, the decoding success is estimated using cross-validation. There is general consensus on assessing statistical significance of cross-validated accuracy with non-parametric permutation tests. In this work we focus on the false positive control of different permutation strategies and on the statistical power of different cross-validation schemes. With simulations, we show that estimating the entire cross-validation error on each permuted dataset is the only statistically valid permutation strategy. Furthermore, using both simulations and real data from the HCP WU-Minn 3T fMRI dataset, we show that, among the different cross-validation schemes, a repeated split-half cross-validation is the most powerful, despite achieving slightly lower classification accuracy, when compared to other schemes. Our findings provide additional insights into the optimization of the experimental design for MVPA, highlighting the benefits of having many short runs.

QSAR Models for Active Substances against Pseudomonas aeruginosa Using Disk-Diffusion Test Data

Pseudomonas aeruginosa is a Gram-negative bacillus included among the six “ESKAPE” microbial species with an outstanding ability to “escape” currently used antibiotics and developing new antibiotics against it is of the highest priority. Whereas minimum inhibitory concentration (MIC) values against Pseudomonas aeruginosa have been used previously for QSAR model development, disk diffusion results (inhibition zones) have not been apparently used for this purpose in the literature and we decided to explore their use in this sense. We developed multiple QSAR methods using several machine learning algorithms (support vector classifier, K nearest neighbors, random forest classifier, decision tree classifier, AdaBoost classifier, logistic regression and naïve Bayes classifier). We used four sets of molecular descriptors and fingerprints and three different methods of data balancing, together with the “native” data set. In total, 32 models were built for each set of descriptors or fingerprint and balancing method, of which 28 were selected and stacked to create meta-models. In terms of balanced accuracy, the best performance was provided by KNN, logistic regression and decision tree classifier, but the ensemble method had slightly superior results in nested cross-validation.

Predicting Postoperative Mortality After Metastatic Intraspinal Neoplasm Excision: Development of a Machine-Learning Approach

OBJECTIVE

Mortality following surgical resection of spinal tumors is a devastating outcome. Naïve Bayes machine learning algorithms may be leveraged in surgical planning to predict mortality. In this investigation, we use a Naïve Bayes classification algorithm to predict mortality following spinal tumor excision within 30 days of surgery.

METHODS

Patients who underwent laminectomies between 2006 and 2018 for excisions of intraspinal neoplasms were selected from the National Surgical Quality Initiative Program. Naïve Bayes classifier analysis was conducted in Python. The area under the receiver operating curve (AUC) was calculated to evaluate the classifier's ability to predict mortality within 30 days of surgery. Multivariable logistic regression analysis was performed in R to identify risk factors for 30-day postoperative mortality.

RESULTS

In total, 2094 spine tumor surgery patients were included in the study. The 30-day mortality rate was 5.16%. The classifier yielded an AUC of 0.898, which exceeds the predictive capacity of the National Surgical Quality Initiative Program mortality probability calculator's AUC of 0.722 (P < 0.0001). The multivariable regression indicated that smoking history, chronic obstructive pulmonary disease, disseminated cancer, bleeding disorder history, dyspnea, and low albumin levels were strongly associated with 30-day mortality.

CONCLUSIONS

The Naïve Bayes classifier may be used to predict 30-day mortality for patients undergoing spine tumor excisions, with an increasing degree of accuracy as the model better performs by learning continuously from the input patient data. Patient outcomes can be improved by identifying high-risk populations early using the algorithm and applying that data to inform preoperative decision making, as well as patient selection and education.

Accurate prediction of RNA 5-hydroxymethylcytosine modification by utilizing novel position-specific gapped k-mer descriptors

RNA modification is an essential step towards generation of new RNA structures. Such modification is potentially able to modify RNA function or its stability. Among different modifications, 5-Hydroxymethylcytosine (5hmC) modification of RNA exhibit significant potential for a series of biological processes. Understanding the distribution of 5hmC in RNA is essential to determine its biological functionality. Although conventional sequencing techniques allow broad identification of 5hmC, they are both time-consuming and resource-intensive. In this study, we propose a new computational tool called iRNA5hmC-PS to tackle this problem. To build iRNA5hmC-PS we extract a set of novel sequence-based features called Position-Specific Gapped k-mer (PSG k-mer) to obtain maximum sequential information. Our feature analysis shows that our proposed PSG k-mer features contain vital information for the identification of 5hmC sites. We also use a group-wise feature importance calculation strategy to select a small subset of features containing maximum discriminative information. Our experimental results demonstrate that iRNA5hmC-PS is able to enhance the prediction performance, dramatically. iRNA5hmC-PS achieves 78.3% prediction performance, which is 12.8% better than those reported in the previous studies. iRNA5hmC-PS is publicly available as an online tool at http://103.109.52.8:81/iRNA5hmC-PS. Its benchmark dataset, source codes, and documentation are available at https://github.com/zahid6454/iRNA5hmC-PS.

Olfactory EEG Signal Classification Using a Trapezoid Difference-Based Electrode Sequence Hashing Approach

Olfactory-induced electroencephalogram (EEG) signal classification is of great significance in a variety of fields, such as disorder treatment, neuroscience research, multimedia applications and brain-computer interface. In this paper, a trapezoid difference-based electrode sequence hashing method is proposed for olfactory EEG signal classification. First, an N-layer trapezoid feature set whose size ratio of the top, bottom and height is 1:2:1 is constructed for each frequency band of each EEG sample. This construction is based on N optimized power-spectral-density features extracted from N real electrodes and N nonreal electrode's features. Subsequently, the N real electrodes' sequence (ES) codes of each layer of the constructed trapezoid feature set are obtained by arranging the feature values in ascending order. Finally, the nearest neighbor classification is used to find a class whose ES codes are the most similar to those of the testing sample. Thirteen-class olfactory EEG signals collected from 11 subjects are used to compare the classification performance of the proposed method with six traditional classification methods. The comparison shows that the proposed method gives average accuracy of 94.3%, Cohen's kappa value of 0.94, precision of 95.0%, and F1-measure of 94.6%, which are higher than those of the existing methods.

Phonatory and articulatory representations of speech production in cortical and subcortical fMRI responses

Speaking involves coordination of multiple neuromotor systems, including respiration, phonation and articulation. Developing non-invasive imaging methods to study how the brain controls these systems is critical for understanding the neurobiology of speech production. Recent models and animal research suggest that regions beyond the primary motor cortex (M1) help orchestrate the neuromotor control needed for speaking, including cortical and sub-cortical regions. Using contrasts between speech conditions with controlled respiratory behavior, this fMRI study investigates articulatory gestures involving the tongue, lips and velum (i.e., alveolars versus bilabials, and nasals versus orals), and phonatory gestures (i.e., voiced versus whispered speech). Multivariate pattern analysis (MVPA) was used to decode articulatory gestures in M1, cerebellum and basal ganglia. Furthermore, apart from confirming the role of a mid-M1 region for phonation, we found that a dorsal M1 region, linked to respiratory control, showed significant differences for voiced compared to whispered speech despite matched lung volume observations. This region was also functionally connected to tongue and lip M1 seed regions, underlying its importance in the coordination of speech. Our study confirms and extends current knowledge regarding the neural mechanisms underlying neuromotor speech control, which hold promise to study neural dysfunctions involved in motor-speech disorders non-invasively.

Odor-induced emotion recognition based on average frequency band division of EEG signals

BACKGROUND

Emotion recognition plays a key role in multimedia. To enhance the sensation of reality, smell has been incorporated into multimedia systems because it can directly stimulate memories and trigger strong emotions.

NEW METHOD

For the recognition of olfactory-induced emotions, this study explored a combination method using a support vector machine (SVM) with an average frequency band division (AFBD) method, where the AFBD method was proposed to extract the power-spectral-density (PSD) features from electroencephalogram (EEG) signals induced by smelling different odors. The so-called AFBD method means that each PSD feature was calculated based on equal frequency bandwidths, rather than the traditional EEG rhythm-based bandwidth. Thirteen odors were used to induce olfactory EEGs and their corresponding emotions. These emotions were then divided into two types of emotions, pleasure and disgust, or five types of emotions that were very unpleasant, slightly unpleasant, neutral, slightly pleasant, and very pleasant.

RESULTS

Comparison between the proposed SVM plus AFBD method and other methods found average accuracies of 98.9 % and 88.5 % for two- and five-emotion recognition, respectively. These values were considerably higher than those of other combination methods, such as the combinations of AFBD or EEG rhythm-based features with naive Bayesian, k-nearest neighbor classification, voting-extreme learning machine, and backpropagation neural network methods.

CONCLUSIONS

The SVM plus AFBD method represents a useful contribution to olfactory-induced emotion recognition. Classification of the five-emotion categories was generally inferior to the classification of the two-emotion categories, suggesting that the recognition performance decreased as the number of emotions in the category increased.

Computational Feature Analysis of Body Movements Reveals Hierarchical Brain Organization

Social species spend considerable time observing the body movements of others to understand their actions, predict their emotions, watch their games, or enjoy their dance movements. Given the important information obtained from body movements, we still know surprisingly little about the details of brain mechanisms underlying movement perception. In this fMRI study, we investigated the relations between movement features obtained from automated computational analyses of video clips and the corresponding brain activity. Our results show that low-level computational features map to specific brain areas related to early visual- and motion-sensitive regions, while mid-level computational features are related to dynamic aspects of posture encoded in occipital–temporal cortex, posterior superior temporal sulcus and superior parietal lobe. Furthermore, behavioral features obtained from subjective ratings correlated with activity in higher action observation regions. Our computational feature-based analysis suggests that the neural mechanism of movement encoding is organized in the brain not so much by semantic categories than by feature statistics of the body movements.

Sensorimotor Representation of Speech Perception. Cross-Decoding of Place of Articulation Features during Selective Attention to Syllables in 7T fMRI

Sensorimotor integration, the translation between acoustic signals and motoric programs, may constitute a crucial mechanism for speech. During speech perception, the acoustic-motoric translations include the recruitment of cortical areas for the representation of speech articulatory features, such as place of articulation. Selective attention can shape the processing and performance of speech perception tasks. Whether and where sensorimotor integration takes place during attentive speech perception remains to be explored. Here, we investigate articulatory feature representations of spoken consonant-vowel (CV) syllables during two distinct tasks. Fourteen healthy humans attended to either the vowel or the consonant within a syllable in separate delayed-match-to-sample tasks. Single-trial fMRI blood oxygenation level-dependent (BOLD) responses from perception periods were analyzed using multivariate pattern classification and a searchlight approach to reveal neural activation patterns sensitive to the processing of place of articulation (i.e., bilabial/labiodental vs. alveolar). To isolate place of articulation representation from acoustic covariation, we applied a cross-decoding (generalization) procedure across distinct features of manner of articulation (i.e., stop, fricative, and nasal). We found evidence for the representation of place of articulation across tasks and in both tasks separately: for attention to vowels, generalization maps included bilateral clusters of superior and posterior temporal, insular, and frontal regions; for attention to consonants, generalization maps encompassed clusters in temporoparietal, insular, and frontal regions within the right hemisphere only. Our results specify the cortical representation of place of articulation features generalized across manner of articulation during attentive syllable perception, thus supporting sensorimotor integration during attentive speech perception and demonstrating the value of generalization.