Hybrid optimization assisted channel selection of EEG for deep learning model-based classification of motor imagery task

OBJECTIVES

To design and develop an approach named HC + SMA-SSA scheme for classifying motor imagery task.

METHODS

The offered model employs a new method for classifying motor imagery task. Initially, down sampling is deployed to pre-process the incoming signal. Subsequently, "Modified Stockwell Transform (ST) and common spatial pattern (CSP) based features are extracted". Then, optimal channel selection is made by a novel hybrid optimization model named as Spider Monkey Assisted SSA (SMA-SSA). Here, "Long Short Term Memory (LSTM) and Bidirectional Gated Recurrent Unit (BI-GRU)" models are used for final classification, whose outcomes are averaged at the end. At last, the improvement of SMA-SSA based model is proven over different metrics.

RESULTS

A superior sensitivity of 0.939 is noted for HC + SMA-SSA that was higher over HC with no optimization and proposed with traditional ST.

CONCLUSIONS

The proposed method achieved effective classification performance in terms of performance measures.

Clusters of Trauma Types as Measured by the Life Events Checklist for DSM-5

Experiences of potentially traumatic events (PTE), commonly assessed with the Life Events Checklist for DSM-5 (LEC-5), can be both varied in pattern and type. An understanding of LEC-assessed PTE type clusters and their relation to psychopathology can enhance research feasibility (e.g., address low base rates for certain PTE types), research communication/comparisons via the use of common terminology, and nuanced trauma assessments/treatments. To this point, the current study examined (1) clusters of PTE types assessed by the LEC-5; and (2) differential relations of these PTE type clusters to mental health correlates (i.e., posttraumatic stress disorder [PTSD] severity, depression severity, emotion dysregulation, reckless and self-destructive behaviors [RSDBs]). A trauma-exposed community sample of 408 participants was recruited via Amazon's Mechanical Turk (M age = 35.90 years; 56.50% female). Network analyses indicated three PTE type clusters: Accidental/Injury Traumas (LEC-5 items 1, 2, 3, 4, 12), Victimization Traumas (LEC-5 items 6, 8, 9), and Predominant Death Threat Traumas (LEC-5 items 5, 7, 10, 11, 13-16). Multiple regression analyses indicated that the Victimization Trauma Cluster significantly predicted PTSD severity (β = .23, p <.001), depression severity (β = .20, p =.001), and negative emotion dysregulation (β = .22, p <.001); and the Predominant Death Threat Trauma Cluster significantly predicted engagement in RSDBs (β = 31, p <.001) and positive emotion dysregulation (β = .26, p <.001), accounting for the influence of other PTE Clusters. Results support three PTE type classifications assessed by the LEC-5, with important clinical and research implications.

Comparison of latent typologies of posttraumatic stress disorder and depression symptoms across military personnel from India and the US

Research has identified heterogeneous subgroups of individuals based on posttraumatic stress disorder (PTSD) and depression symptoms. Using data collected from military personnel in India (N = 146) and U.S. (N = 194), we examined (1) the best-fitting latent class solution; (2) multi-group invariance of the class solution; and (3) construct validity of optimal class solution. Results indicated that the optimal 4-class solution differed in severity and severity/type in the India and U.S. samples respectively. With similarity in the optimal number of classes across cultural samples, the meaning/nature of classes differed. In the India sample, anxiety severity predicted the Low Severity Class vs. all other classes, and the Moderately High Severity/High Severity Classes vs. the Moderately Low Severity Class; number of traumas predicted the High Severity Class vs. other classes; and resilience predicted the Moderately Low Severity Class vs. the Moderately High Severity Class. In the U.S. sample, alcohol use predicted the High Severity Class vs. all other classes, and the High Depression-Low PTSD Class vs. the Low Severity Class; rumination significantly predicted the High Severity and High Depression-Low PTSD Classes vs. each of the High PTSD-Low Depression and Low Severity Classes. Thus, meaning and nature of PTSD-depression subgroups may vary culturally; hence, culturally-sensitive interventions need to account for this heterogeneity.

Fitting Bayesian Models for Single-Case Experimental Designs

Abstract. Single-case experimental designs (SCEDs) are interrupted time-series designs that have recently gained recognition as being able to provide a strong basis for establishing intervention effect. Typically, SCED data are short time series and autocorrelated, which renders maximum likelihood and parametric analyses inadequate for data analysis, respectively. Although Bayesian methods overcome these challenges, most practitioners do not use Bayesian estimation because of: (a) its steep learning curve, (b) lack of Bayesian training, and (c) lack of knowledge of Bayesian software solutions. This study demonstrates two Bayesian interrupted time-series models using freeware programs R and JAGS. Practitioners could modify these codes and run them for their own data by changing the values in the codes where indicated. Providing practitioners with such tools to facilitate their analysis is one way to improve methodological rigor in applied research.

Comparing patients and families perceptions of satisfaction and predictors of overall satisfaction in the emergency department

Study objective

The aim of the study was to investigate factors that best predict patient’s satisfaction with their ED visit, as well as examine whether patients and their families perceived the factors related to satisfaction similarly.

Methods

This is a retrospective study where secondary data analysis was done on patient satisfaction data collected over three quarters for quality improvement purposes. Exploratory factor analysis (EFA) was conducted on the data from the first quarter to identify the factor structure, followed by confirmatory factor analysis (CFA) on the rest of the data to confirm the EFA factor structure. A structural equation model (SEM) was fitted where the factors predicted overall satisfaction with the ED visit. Finally, measurement invariance was conducted to examine if patients and families perceived the factors related to ED services alike.

Results

Two factors were found to be predictive of satisfaction: clinical team and system processes. The SEM showed that system process was a statistically significant predictor of overall satisfaction, while clinical team predicted overall satisfaction to a smaller extent. Multi-group CFA showed that the factor structure fitted neither family nor patient groups adequately. The instrument did not exhibit partial invariance.

Conclusion

This study found that system process was the best predictor of overall satisfaction. Furthermore, this study showed that the same instrument might not reliably compare the perceptions of patients and families.

Assessing Statistical Anxiety Among Online and Traditional Students

The purpose of this study was to determine whether scores on the Statistical Anxiety Scale (SAS) manifest in the same way for students in online and traditional statistics courses. Tests of measurement invariance indicated that invariance of the two-factor model of the SAS held at every level. Therefore, we compared the statistical anxiety of online and traditional students. Results indicated that online and traditional statistics students reported comparable levels of anxiety with slightly less anxiety in terms of seeking help for traditional students. We concluded that online instruction is a viable form of statistics education at least for undergraduate students enrolled in the social sciences.

Invariance of the Construct of Posttraumatic Stress Disorder: A Systematic Review

We conducted a systematic review of studies that have evaluated invariance of the construct of posttraumatic stress disorder (PTSD) to summarize their conclusions related to invariance/noninvariance and sources of noninvariance. In November 2017, we searched Pubmed, PSYCINFO, PILOTS Web of Science, CINAHL, Medline, and Psychological and Behavioral Science Collection for abstracts and articles with these inclusionary criteria: peer-reviewed, including DSM-IV or DSM-5 PTSD invariance as a main study aim, use of multigroup confirmatory factor analyses, and use of an independent PTSD instrument or module. In total, 45 articles out of 1,169 initially identified abstracts met inclusion criteria. Research assistants then followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to complete a secondary search and independently extract data. Results indicated that DSM-IV dysphoric arousal and DSM-5 hybrid model factors demonstrated the most stability; sources of instability were some intrusion (distress to trauma cues), dysphoria/numbing (traumatic amnesia, foreshortened future, emotional numbness, detachment), and arousal (hypervigilance) items. The PTSD Checklist and PTSD Reaction Index were most often used to assess PTSD in studies investigating its invariance; however, these measures demonstrated partial conceptual equivalence of PTSD across subgroups. Instead, clinician-administered measures demonstrated more conceptual equivalence across subgroups. Age, gender, cultural/linguistic factors, and sample diversity had the least moderating effect on PTSD's symptom structure. Our review demonstrates the need to examine invariance of the PTSD construct following recommended guidelines for each empirical and clinical trial study to draw meaningful multigroup comparative conclusions.

Challenges in measuring ACGME competencies: considerations for milestones

Background

Measuring milestones, competencies, and sub-competencies as residents progress through a training program is an essential strategy in Accreditation Council for Graduate Medical Education (ACGME)’s attempts to ensure graduates meet expected professional standards. Previous studies have found, however, that physicians make global ratings often by using a single criterion.

Methods

We use advanced statistical analysis to extend these studies by examining the validity of ACGME International competency measures for an international setting, across emergency medicine (EM) and neurology, and across evaluators. Confirmatory factor analysis (CFA) models were fitted to both EM and neurology data. A single-factor CFA was hypothesized to fit each dataset. This model was modified based on model fit indices. Differences in how different EM physicians perceived the core competencies were tested using a series of measurement invariance tests.

Results

Extremely high alpha reliability coefficients, factor coefficients (> .93), and item correlations indicated multicollinearity, that is, most items being evaluated could essentially replace the underlying construct itself. This was true for both EM and neurology data, as well as all six EM faculty.

Conclusions

Evaluation forms measuring the six core ACGME competencies did not possess adequate validity. Severe multicollinearity exists for the six competencies in this study. ACGME is introducing milestones with 24 sub-competencies. Attempting to measure these as discrete elements, without recognizing the inherent weaknesses in the tools used will likely serve to exacerbate an already flawed strategy. Physicians likely use their “gut feelings” to judge a resident’s overall performance. A better process could be conceived in which this subjectivity is acknowledged, contributing to more meaningful evaluation and feedback.

Development and initial psychometric validation of the Brief-Caffeine Expectancy Questionnaire (B-CaffEQ)

Caffeine is the most widely available and consumed psychoactive substance in the United States. Extant work indicates that across substances, use expectancies play a marked role in the development and maintenance of consumption patterns. Despite a burgeoning line of etiological and intervention-oriented research focused on expectancies (e.g., alcohol), there is a limited literature regarding caffeine use effect expectancies, specifically. To facilitate this work, the Caffeine Expectancy Questionnaire (CaffEQ) was developed and psychometrically validated; however, the length of the CaffEQ (i.e., 47 items) may hinder widespread adoption of this tool. As such, the current study provides an initial psychometric validation of a brief, 20-item version-the Brief-Caffeine Expectancy Questionnaire (B-CaffEQ)-in a multiethnic sample of undergraduate students (N = 975). Results showed that the B-CaffEQ replicated the 7-factor structure of the original CaffEQ using both constrained (confirmatory factor analysis) and less constrained (exploratory structural equation modeling) structural models and evidenced good internal consistency across subscales. The B-CaffEQ also demonstrated concurrent validity with caffeine use frequency indices, replicated and extended convergent validity between caffeine expectancy subscales and related behavioral and psychological constructs, and demonstrated discriminant validity with other related, but notably distinct, stimulant use metrics (e.g., cocaine, Ritalin). Lastly, the B-CaffEQ appears to provide an invariant measure of expectancies across types of caffeine users. These findings indicate that the B-CaffEQ is a reliable assessment of caffeine use effect expectancies, with acceptable-to-good psychometric properties-comparable in length to other substance use expectancies measures-that may be more readily incorporated into research and clinical settings. (PsycINFO Database Record (c) 2018 APA, all rights reserved).

Evaluation of threshold selection methods for adaptive kernel density estimation in disease mapping

Background

Maps of disease rates produced without careful consideration of the underlying population distribution may be unreliable due to the well-known small numbers problem. Smoothing methods such as Kernel Density Estimation (KDE) are employed to control the population basis of spatial support used to calculate each disease rate. The degree of smoothing is controlled by a user-defined parameter (bandwidth or threshold) which influences the resolution of the disease map and the reliability of the computed rates. Methods for automatically selecting a smoothing parameter such as normal scale, plug-in, and smoothed cross validation bandwidth selectors have been proposed for use with non-spatial data, but their relative utilities remain unknown. This study assesses the relative performance of these methods in terms of resolution and reliability for disease mapping.

Results

Using a simulated dataset of heart disease mortality among males aged 35 years and older in Texas, we assess methods for automatically selecting a smoothing parameter. Our results show that while all parameter choices accurately estimate the overall state rates, they vary in terms of the degree of spatial resolution. Further, parameter choices resulting in desirable characteristics for one sub group of the population (e.g., a specific age-group) may not necessarily be appropriate for other groups.

Conclusion

We show that the appropriate threshold value depends on the characteristics of the data, and that bandwidth selector algorithms can be used to guide such decisions about mapping parameters. An unguided choice may produce maps that distort the balance of resolution and statistical reliability.

Bayesian Prior Choice in IRT Estimation Using MCMC and Variational Bayes

This study investigated the impact of three prior distributions: matched, standard vague, and hierarchical in Bayesian estimation parameter recovery in two and one parameter models. Two Bayesian estimation methods were utilized: Markov chain Monte Carlo (MCMC) and the relatively new, Variational Bayesian (VB). Conditional (CML) and Marginal Maximum Likelihood (MML) estimates were used as baseline methods for comparison. Vague priors produced large errors or convergence issues and are not recommended. For both MCMC and VB, the hierarchical and matched priors showed the lowest root mean squared errors (RMSEs) for ability estimates; RMSEs of difficulty estimates were similar across estimation methods. For the standard errors (SEs), MCMC-hierarchical displayed the largest values across most conditions. SEs from the VB estimation were among the lowest in all but one case. Overall, VB-hierarchical, VB-matched, and MCMC-matched performed best. VB with hierarchical priors are recommended in terms of their accuracy, and cost and (subsequently) time effectiveness.

Comparing interval estimates for small sample ordinal CFA models

Robust maximum likelihood (RML) and asymptotically generalized least squares (AGLS) methods have been recommended for fitting ordinal structural equation models. Studies show that some of these methods underestimate standard errors. However, these studies have not investigated the coverage and bias of interval estimates. An estimate with a reasonable standard error could still be severely biased. This can only be known by systematically investigating the interval estimates. The present study compares Bayesian, RML, and AGLS interval estimates of factor correlations in ordinal confirmatory factor analysis models (CFA) for small sample data. Six sample sizes, 3 factor correlations, and 2 factor score distributions (multivariate normal and multivariate mildly skewed) were studied. Two Bayesian prior specifications, informative and relatively less informative were studied. Undercoverage of confidence intervals and underestimation of standard errors was common in non-Bayesian methods. Underestimated standard errors may lead to inflated Type-I error rates. Non-Bayesian intervals were more positive biased than negatively biased, that is, most intervals that did not contain the true value were greater than the true value. Some non-Bayesian methods had non-converging and inadmissible solutions for small samples and non-normal data. Bayesian empirical standard error estimates for informative and relatively less informative priors were closer to the average standard errors of the estimates. The coverage of Bayesian credibility intervals was closer to what was expected with overcoverage in a few cases. Although some Bayesian credibility intervals were wider, they reflected the nature of statistical uncertainty that comes with the data (e.g., small sample). Bayesian point estimates were also more accurate than non-Bayesian estimates. The results illustrate the importance of analyzing coverage and bias of interval estimates, and how ignoring interval estimates can be misleading. Therefore, editors and policymakers should continue to emphasize the inclusion of interval estimates in research.

Measuring Urban Teachers’ Beliefs About African American Students

Understanding urban teachers’ beliefs about African American students has become important because (a) many teachers are reluctant to teach students from other cultures, and (b) most teachers are European American. To construct a psychometrically sound measure of teacher beliefs, the authors investigate the measurement properties of a teacher beliefs factor. This factor was selected from an inventory of items that purported to measure urban teachers’ cultural awareness and beliefs. Measurement invariance of the teacher beliefs factor across European American, African American, and Hispanic American teachers addressed its construct validity. The authors examine the psychometric properties of these items using graded response multilevel analysis. The final 5-item factor showed highest level of invariance for African American and European American teachers but did not fit Hispanic American teachers well. All the five items had good psychometric properties. Analyses of latent means showed that African American teachers had more positive beliefs about African American students than European American teachers did. However, the latent scores were bimodally distributed for African American teachers showing that one subgroup of African American teachers had similar beliefs as European American teachers while another subgroup had more positive beliefs.

An Item Response Theory Analysis of the Mathematics Teaching Efficacy Beliefs Instrument

The mathematics teaching efficacy beliefs of preservice elementary teachers have been the subject of several studies. A widely used measure in these studies is the Mathematics Teaching Efficacy Beliefs Instrument (MTEBI). The present study provides a detailed analysis of the psychometric properties of the MTEBI using Bayesian item response theory. We discuss local dependence between item pairs, psychometric quality of the items, validity of the scoring procedure, and measurement accuracy for teachers with different efficacy levels. Our findings suggest that in its present form, the test reliability of the MTEBI may not be as high as assumed to date. The scale, wording, and placement of the items need revision. Moreover, additional items need to be constructed to measure below average levels of efficacy more accurately. Ordering the items according to difficulty, we describe the structure of mathematics teaching efficacy beliefs and draw some implications for mathematics teacher educators.