A survey of methodologies for the treatment of missing values within datasets: limitations and benefits

Lightweight federated learning for STIs/HIV prediction

This paper presents a solution that prioritises high privacy protection and improves communication throughput for predicting the risk of sexually transmissible infections/human immunodeficiency virus (STIs/HIV). The approach utilised Federated Learning (FL) to construct a model from multiple clinics and key stakeholders. FL ensured that only models were shared between clinics, minimising the risk of personal information leakage. Additionally, an algorithm was explored on the FL manager side to construct a global model that aligns with the communication status of the system. Our proposed method introduced Random Forest Federated Learning for assessing the risk of STIs/HIV, incorporating a flexible aggregation process that can be adjusted to accommodate the capacious communication system. Experimental results demonstrated the significant potential of a solution for estimating STIs/HIV risk. In comparison with recent studies, our approach yielded superior results in terms of AUC (0.97) and accuracy ( 93 % ). Despite these promising findings, a limitation of the study lies in the experiment for man's data, due to the self-reported nature of the data and sensitive content. which may be subject to participant bias. Future research could check the performance of the proposed framework in partnership with high-risk populations (e.g., men who have sex with men) to provide a more comprehensive understanding of the proposed framework's impact and ultimately aim to improve health outcomes/health service optimisation.

Working under the Shadow of Drones: Investigating Occupational Safety Hazards among Commercial Drone Pilots

TECHNICAL ABSTRACTBackground: Commercial drones are rapidly transforming business operations, however there is a paucity of research evaluating occupational hazards and risks associated with drone deployment in the workplace.Purpose: We aimed to identify challenges of human-drone collaborations and assess drone pilot perceptions of workplace safety.Methods: An online questionnaire was generated and sent to 308 drone pilots working in different industries. A total of 75 of responses were included for data analysis. Descriptive statistics, principal component analysis, and association rule mining were employed to extract knowledge from the obtained data.Results: Our results indicate that human factors are the main contributors to workplace drone mishaps. Poor communication, information display, and control modes were found to be chief obstacles to effective human-drone collaboration. Drone pilots indicated a propensity for complying with and participating in safety practices. Following safety procedures, receiving technical training, and flying outdoors may all be associated with a lower risk of drone mishaps.Conclusions: Offering professional training to pilots and following safety procedures could decrease the risks associated with occupational drones.

Perceived household financial decline and physical/mental health among adolescents during the COVID-19 crisis: Focusing on gender differences

COVID-19 has worsened adolescents' mental and physical health. Several studies have reported that the effect may be greater in girls; however, relevant socio-ecological factors have not been examined. This study aimed to examine the factors associated with physical and mental health status among adolescents and the moderating role of gender on the relationship between physical and mental health status and perceived household financial decline. We analyzed the cross-sectional 2020 Korea Youth Risk Behavior Web-based Survey (KYRBS) collected between August and November 2020 in South Korea. It included 54,809 adolescents (28,269 males and 26,540 females), on average aged 15.1. We conducted ordinary least squares (OLS) regressions to examine the factors associated with physical and mental health outcomes. Gender differences were observed in associated factors. Then, we tested the moderating effect of gender by including an interaction term between gender and perceived household financial decline due to the COVID-19 pandemic. Perceived household financial decline due to COVID-19 negatively affected both groups. Perceiving moderate and severe financial decline due to COVID-19 is negatively associated with self-rated health among female adolescents than male counterparts. Female adolescents were also more vulnerable to mental health outcomes (i.e., distress, anxiety, and loneliness) when they perceived severe or moderate household financial decline due to COVID-19 compared to their male peers. Our findings suggest that female adolescents are more vulnerable to household financial shocks due to COVID-19, especially in households that have experienced a severe decline. We suggest the need for gender-sensitive policy interventions for adolescent mental health.

Longitudinal associations between specific types/amounts social contact and cognitive function among middle-aged and elderly Chinese: A causal inference and longitudinal targeted maximum likelihood estimation analysis

BACKGROUND

Social contact has demonstrated associations with cognitive function, while the literature on the effect of specific social relationship subdomains on cognitive function is limited. This study aimed to examine the causal effects of specific types/amounts of social contact on cognitive function among middle-aged and elderly Chinese.

METHODS

A total of 38,883 middle-aged and elderly adults from the China Health and Retirement Longitudinal Study were involved. Social contact in this study included interaction with families, taking care of grandchildren, interaction with friends, and participation in three types of social activities. We performed the linear mixed-effects model analysis with propensity score approach and the longitudinal targeted maximum likelihood-based estimation analysis after adjusting for potential covariates and confounders.

RESULTS

Interaction with families, caring for grandchildren, interaction with friends and participation in social activities were all associated with cognitive z-scores. Participants who interacted with families "2-3 times a week" and "once a week" versus "almost every day" had higher cognitive z-scores. Those who interacted with friends and participated in social activities "almost every week" versus "almost daily" had higher cognitive z-scores.

LIMITATIONS

The assessment of cognition was biased against people with poor education due to elements of language and mathematical testing, and against those with visual impairment.

CONCLUSIONS

Social contact was associated with better cognitive function among middle-aged and elderly Chinese. Social contact "1-3 times a week" was optimal for cognitive function. More social contact in middle-aged and elderly Chinese led to less cognitive decline in later life than in their inactive peers.

BiT-MAC: Mortality prediction by bidirectional time and multi-feature attention coupled network on multivariate irregular time series

Mortality prediction is crucial to evaluate the severity of illness and assist in improving the prognosis of patients. In clinical settings, one way is to analyze the multivariate time series (MTSs) of patients based on their medical data, such as heart rates and invasive mean arterial blood pressure. However, this suffers from sparse, irregularly sampled, and incomplete data issues. These issues can compromise the performance of follow-up MTS-based analytic applications. Plenty of existing methods try to deal with such irregular MTSs with missing values by capturing the temporal dependencies within a time series, yet in-depth research on modeling inter-MTS couplings remains rare and lacks model interpretability. To this end, we propose a bidirectional time and multi-feature attention coupled network (BiT-MAC) to capture the temporal dependencies (i.e., intra-time series coupling) and the hidden relationships among variables (i.e., inter-time series coupling) with a bidirectional recurrent neural network and multi-head attention, respectively. The resulting intra- and inter-time series coupling representations are then fused to estimate the missing values for a more robust MTS-based prediction. We evaluate BiT-MAC by applying it to the missing-data corrupted mortality prediction on two real-world clinical datasets, i.e., PhysioNet'2012 and COVID-19. Extensive experiments demonstrate the superiority of BiT-MAC over cutting-edge models, verifying the great value of the deep and hidden relations captured by MTSs. The interpretability of features is further demonstrated through a case study.

Machine learning algorithms identify demographics, dietary features, and blood biomarkers associated with stroke records

OBJECTIVE

We conducted a comprehensive evaluation of features associated with stroke records.

METHODS

We screened the dietary nutrients, blood biomarkers, and clinical information from the National Health and Nutrition Examination Survey (NHANES) 2015-16 database to assess a self-reported history of all strokes (136 strokes, n = 4381). We computed feature importance, built machine learning (ML) models, developed a nomogram, and validated the nomogram on NHANES 2007-08, 2017-18, and the baseline UK Biobank. We calculated the odds ratios with/without adjusting sampling weights (OR/OR_w).

RESULTS

The clinical features have the best predictive power compared to dietary nutrients and blood biomarkers, with 22.8% increased average area under the receiver operating characteristic curves (AUROC) in ML models. We further modeled with ten most important clinical features without compromising the predictive performance. The key features positively associated with stroke include age, cigarette smoking, tobacco smoking, Caucasian or African American race, hypertension, diabetes mellitus, asthma history; the negatively associated feature is the family income. The nomogram based on these key features achieved good performances (AUROC between 0.753 and 0.822) on the test set, the NHANES 2007-08, 2017-18, and the UK Biobank. Key features from the nomogram model include age (OR = 1.05, OR_w = 1.06), Caucasian/African American (OR = 2.68, OR_w = 2.67), diabetes mellitus (OR = 2.30, OR_w = 1.99), asthma (OR = 2.10, OR_w = 2.41), hypertension (OR = 1.86, OR_w = 2.10), and income (OR = 0.83, OR_w = 0.81).

CONCLUSIONS

We identified clinical key features and built predictive models for assessing stroke records with high performance. A nomogram consisting of questionnaire-based variables would help identify stroke survivors and evaluate the potential risk of stroke.

The Influence of Stigma and Trust in Young People Seeking Support for Their Own or a Friend’s Symptoms: The Role of Threat Appraisals

Background

Stigma and trust influence how adolescents seek support for mental illness, though it is unclear how these influence their decisions to approach a range of potential sources of support. Moreover, even less is known about the ways in which these issues are related when a friend discloses symptoms of mental illness.

Objective

The study’s aims were to understand the role of stigma, trust, and threat appraisals in adolescents’ support seeking when exposed to their own, or to a friend’s, symptoms of mental illness.

Method

A vignette-based study comparing reports of support (friends, parents, teachers, professionals, and online) was completed with reference to either (i) experiencing symptoms of mental illness or (ii) having a friend disclose these types of symptoms. Two hundred and fifty adolescents (M = 12.75 years) answered questions pertaining to stigma (public and self), trust levels, threat appraisals, and support seeking.

Results

When dealing with their own symptoms, threat accounted for 4.8 and 2.5% of the variance when seeking support from parents and professionals, respectively. Self-stigma accounted for 2.4% of variance when seeking support from parents and 0.8% of variance when seeking support from professionals. Trust moderated the association between threat and the use of online support. When responding to a friend’s disclosure, higher levels of public-stigma were associated with lower support seeking from friends, parents, and professionals.

Conclusions

This study showed a distinction in how adolescents deal with their own or a friend’s symptoms of mental illness, and what resources they choose to ask for support from. Self-stigma, threat, and trust levels were particularly relevant when experiencing their own symptoms, while dealing with a friend’s disclosure was related to levels of public-stigma.

Software Defect Prediction Using Stacking Generalization of Optimized Tree-Based Ensembles

Software defect prediction refers to the automatic identification of defective parts of software through machine learning techniques. Ensemble learning has exhibited excellent prediction outcomes in comparison with individual classifiers. However, most of the previous work utilized ensemble models in the context of software defect prediction with the default hyperparameter values, which are considered suboptimal. In this paper, we investigate the applicability of a stacking ensemble built with fine-tuned tree-based ensembles for defect prediction. We used grid search to optimize the hyperparameters of seven tree-based ensembles: random forest, extra trees, AdaBoost, gradient boosting, histogram-based gradient boosting, XGBoost and CatBoost. Then, a stacking ensemble was built utilizing the fine-tuned tree-based ensembles. The ensembles were evaluated using 21 publicly available defect datasets. Empirical results showed large impacts of hyperparameter optimization on extra trees and random forest ensembles. Moreover, our results demonstrated the superiority of the stacking ensemble over all fine-tuned tree-based ensembles.

Cardio-ML: Detection of malicious clinical programmings aimed at cardiac implantable electronic devices based on machine learning and a missing values resemblance framework

Patients with life-threatening arrhythmias are often treated with cardiac implantable electronic devices (CIEDs), such as pacemakers and implantable cardioverter defibrillators (ICDs). Recent advancements in CIEDs have enabled advanced functionality and connectivity that make such devices (particularly ICDs) vulnerable to cyber-attacks. One of the most dangerous attacks on CIED ecosystems is a data manipulation attack from a compromised programmer device that sends malicious clinical programmings to the CIED. Such attacks can affect the CIED functioning and impact patient's survival and quality of life. In this paper, we propose Cardio-ML - an automated system for the detection of malicious clinical programmings that is based on machine learning algorithms and a novel missing values resemblance framework. Our system is designed to detect new variants of existing attacks and, more importantly, new unknown (zero-day) attacks, aimed at ICDs. We collected 1651 legitimate clinical programmings from 514 patients, over a four-year period, from programmer devices at two medical centers. Our collection also includes 28 core malicious functionalities created by cardiac electrophysiology experts that were later used to create different variants of malicious programmings. Cardio-ML was evaluated extensively in three comprehensive experiments and showed high detection capabilities in most attack scenarios. We achieved perfect classification results for detecting newly created variants of existing core malicious functionalities, with an AUC of 100%; for completely new unknown (zero-day) malicious clinical programmings, an AUC of 80% was obtained, which is 14% better than the state-of-the-art method. We were able to further improve our detection results by identifying the best combination of legitimate and zero-day malicious programmings in the dataset, achieving an AUC of 87%. CIED clinical programmings have many parameters without values for a large number of samples (programmings). To cope with the extreme amount of missing values in our dataset, we developed a novel missing values-based resemblance framework and evaluated it using three dataset-creation approaches: a standard expert-driven approach, our novel data-driven approach, and a combined approach incorporating both approaches. The results showed that our novel framework handles missing values in the data better than the expert-driven approach which yields an empty dataset. In particular, the combined approach showed a 40% improvement in data utilization compared to the data-driven approach.

Advanced methods for missing values imputation based on similarity learning

The real-world data analysis and processing using data mining techniques often are facing observations that contain missing values. The main challenge of mining datasets is the existence of missing values. The missing values in a dataset should be imputed using the imputation method to improve the data mining methods' accuracy and performance. There are existing techniques that use k-nearest neighbors algorithm for imputing the missing values but determining the appropriate k value can be a challenging task. There are other existing imputation techniques that are based on hard clustering algorithms. When records are not well-separated, as in the case of missing data, hard clustering provides a poor description tool in many cases. In general, the imputation depending on similar records is more accurate than the imputation depending on the entire dataset's records. Improving the similarity among records can result in improving the imputation performance. This paper proposes two numerical missing data imputation methods. A hybrid missing data imputation method is initially proposed, called KI, that incorporates k-nearest neighbors and iterative imputation algorithms. The best set of nearest neighbors for each missing record is discovered through the records similarity by using the k-nearest neighbors algorithm (kNN). To improve the similarity, a suitable k value is estimated automatically for the kNN. The iterative imputation method is then used to impute the missing values of the incomplete records by using the global correlation structure among the selected records. An enhanced hybrid missing data imputation method is then proposed, called FCKI, which is an extension of KI. It integrates fuzzy c-means, k-nearest neighbors, and iterative imputation algorithms to impute the missing data in a dataset. The fuzzy c-means algorithm is selected because the records can belong to multiple clusters at the same time. This can lead to further improvement for similarity. FCKI searches a cluster, instead of the whole dataset, to find the best k-nearest neighbors. It applies two levels of similarity to achieve a higher imputation accuracy. The performance of the proposed imputation techniques is assessed by using fifteen datasets with variant missing ratios for three types of missing data; MCAR, MAR, MNAR. These different missing data types are generated in this work. The datasets with different sizes are used in this paper to validate the model. Therefore, proposed imputation techniques are compared with other missing data imputation methods by means of three measures; the root mean square error (RMSE), the normalized root mean square error (NRMSE), and the mean absolute error (MAE). The results show that the proposed methods achieve better imputation accuracy and require significantly less time than other missing data imputation methods.

Work-related extended availability, psychological detachment, and interindividual differences: A cross-lagged panel study

Work-related extended availability (WREA) refers to employees being available for work-related matters during leisure time. Although studies have suggested negative effects of WREA on employee health, there is a scarcity of longitudinal research especially studies trying to disentangle how WREA may impact health. Moreover, there are only few studies dealing with interindividual differences in the effects of WREA on health. These aspects are crucial as they can help laying a foundation for interventions that help coming to terms with negative effects of WREA. The current study implemented a cross-lagged panel design with three waves to clarify how effects of WREA unfold and whether there are interindividual differences. Based on the stressor-detachment-model and person-environmental-fit theory, we proposed that (1) the relationship between WREA and sleep as well as between WREA and exhaustion is mediated by psychological detachment, and (2) that the relationship between WREA and the outcomes is moderated by segmentation preferences. In total, 528 employees (320 women, mean age = 48 years) participated in the study. Although there was a cross-lagged negative association between WREA and detachment, we did not find an indirect relationship between WREA and either sleep or exhaustion via detachment. Moreover, we did not find evidence for interindividual differences in the effects of WREA on any of the outcomes. On the basis of the negative cross-lagged relationship between WREA and detachment from work, we recommend organizations to discourage employees from WREA, because failure to regularly recover from work may lead to health issues in the long run.

Do Gender-role Values Matter? Explaining New Refugee Women’s Social Contact in Germany

This article investigates whether gender-role values are linked to refugee women’s social contact in Germany. By building on the “preferences–third parties–opportunities” framework, we explicate a direct and an indirect path through which gender-role values may be related to refugee women’s minority-majority, intra-minority, and inter-minority contact. By applying median regressions, marginal structural models, and inverse probability of treatment weighting to data from the 2016 IAB-BAMF-SOEP refugee survey, we show that refugee women’s own gender-traditional values and those of their partners are associated both directly and indirectly with less social contact for these women. Effects of gender-role values on refugee women’s social contact are more pronounced for minority-majority contact than for the other two types of social contact assessed. With the effects of refugee women’s and their partners’ gender-role values being rather small against alternative explanatory factors, we conclude that in contrast to the view traditionally held by the populist right, traditional gender-role values hold refugee women back from establishing social contact in the host society only to a very limited extent.

Fusing imperfect experimental data for risk assessment of musculoskeletal disorders in construction using canonical polyadic decomposition

Field or laboratory data collected for work-related musculoskeletal disorder (WMSD) risk assessment in construction often becomes unreliable as a large amount of data go missing due to technology-induced errors, instrument failures or sometimes at random. Missing data can adversely affect the assessment conclusions. This study proposes a method that applies Canonical Polyadic Decomposition (CPD) tensor decomposition to fuse multiple sparse risk-related datasets and fill in missing data by leveraging the correlation among multiple risk indicators within those datasets. Two knee WMSD risk-related datasets-3D knee rotation (kinematics) and electromyography (EMG) of five knee postural muscles-collected from previous studies were used for the validation and demonstration of the proposed method. The analysis results revealed that for a large portion of missing values (40%), the proposed method can generate a fused dataset that provides reliable risk assessment results highly consistent (70%-87%) with those obtained from the original experimental datasets. This signified the usefulness of the proposed method for use in WMSD risk assessment studies when data collection is affected by a significant amount of missing data, which will facilitate reliable assessment of WMSD risks among construction workers. In the future, findings of this study will be implemented to explore whether, and to what extent, the fused dataset outperforms the datasets with missing values by comparing consistencies of the risk assessment results obtained from these datasets for further investigation of the fusion performance.

Associations between older adults' parental bereavement and their health and well-being: Evidence from the China health and retirement longitudinal study

OBJECTIVES

Few studies have investigated the association between losing a child and parental health and wellbeing, especially among older Chinese bereaved parents. This study examined depressive symptoms, life satisfaction, and self-rated health of older Chinese bereaved parents to estimate the health and well-being of this group.

METHODS

This research used data from the 2015 China Health and Retirement Longitudinal Study (CHARLS). A total number of 11,507 participants age 45 and older were enrolled in the analysis, including 1,758 bereaved adults who had experienced a child's death and 9,749 non-bereaved counterparts. Multivariate linear and logistic regression models were used to examine the effect of bereavement and its interaction effect by sex and age.

RESULTS

Multivariate analyses revealed that the death of a child is associated with an increasing likelihood of experiencing depressive symptoms (adjusted OR = 1.425, p < 0.001), and a reduced probability of a high level of life satisfaction (adjusted OR = 0.725, p < 0.05), whereas experiencing a child's death is not significantly associated with self-reported health status. The effects of bereavement on health and well-being were found to have a much greater impact among participants who were males (compared to females) and who aged<60 years (compared to those ≥60 years).

DISCUSSION

Future longitudinal prospective research is expected to examine the causal relationship and explore the attributes of child death and its effects on parental health. Interventions to improve the health and well-being of the older bereaved population are warranted, particularly for those who are male and under 60 years of age.

A deep learning technique for imputing missing healthcare data

Missing data is a frequent occurrence in medical and health datasets. The analysis of datasets with missing data can lead to loss in statistical power or biased results. We address this issue with a novel deep learning technique to impute missing values in health data. Our method extends upon an autoencoder to derive a deep learning architecture that can learn the hidden representations of data even when data is perturbed by missing values (noise). Our model is constructed with overcomplete representation and trained with denoising regularization. This allows the latent/hidden layers of our model to effectively extract the relationships between different variables; these relationships are then used to reconstruct missing values. Our contributions include a new loss function designed to avoid local optima, and this helps the model to learn the real distribution of variables in the dataset. We evaluate our method in comparison with other well-established imputation strategies (mean, median imputation, SVD, KNN, matrix factorization and soft impute) on 48,350 Linked Birth/Infant Death Cohort Data records. Our experiments demonstrate that our method achieved lower imputation mean squared error (MSE=0.00988) compared with other imputation methods (with MSE ranging from 0.02 to 0.08). When assessing the imputation quality using the imputed data for prediction tasks, our experiments show that the data imputed by our method yielded better results (F1=70.37%) compared with other imputation methods (ranging from 66 to 69%).

Health Care Transition Services and Adaptive and Social-Emotional Functioning of Youth with Autism Spectrum Disorder

To promote health care transition services (HCTS) among youth with autism spectrum disorder (ASD), it is important to understand their access to HCTS and the association with functioning. We conducted weighted descriptive statistics and regressions. Findings suggested that HCTS were inconsistently provided to youth with ASD. Access to two or more HCTS was associated with positive social-emotional functioning. Helping youth with ASD understand health care changes and working with them to gain skills in managing health needs were found to be significant determinants of positive social-emotional functioning. The present study sheds light on HCTS that are essential for youth with ASD and highlights the necessity of health care system changes to promote service access and optimal functioning for youth with ASD.

Participation in Part C Early Intervention: One Key to an Earlier Diagnosis of Autism Spectrum Disorder?

OBJECTIVE

To determine whether participation in a state early intervention program is associated with reduction in the age of diagnosis of autism spectrum disorder (ASD).

STUDY DESIGN

State agency, Medicaid, and Census data were integrated for children with ASD enrolled in a Medicaid waiver between February 2007 and March 2015 (N = 1613). Ordinary least squares regression was used to estimate the relationship between participation in a state early intervention program and their age of diagnosis of ASD.

RESULTS

The model explained 34% of variation in age of diagnosis (F[17,1595] = 49.20, P < .0001, adj R² = 0.34). After adjustment for key variables, compared with children who did not participate in early intervention, children who did participate were diagnosed 2 years earlier (β = -23.97, P < .0001).

CONCLUSIONS

Although conducted in only 1 state, this study suggests that participation in early intervention programs may be instrumental in earlier diagnosis of ASD. These findings underscore the importance of identifying children who qualify for early intervention programs, the value of encouraging childhood professionals (eg, early care providers and educators) to refer given documented barriers to pediatrician referral, and the need for research that identifies the mechanisms by which programs may promote earlier diagnosis (eg, service coordination, parent support).

Adverse childhood experiences and preschool suspension expulsion: A population study

BACKGROUND

Preschool suspension and expulsion rates are typically based on teacher reports, and don't simultaneously account for adverse childhood experiences (ACEs).

OBJECTIVE

To examine estimates in the United States of parent-reported preschool suspension and expulsion rates, in the context of ACEs.

PARTICIPANTS AND SETTING

Parents of children aged 3-5 years old (N = 6,100) in the 2016 National Survey of Children's Health dataset.

METHOD

We reported the prevalence estimates of preschool suspension and expulsion, and estimated the unique variance of ACEs as risk factors using weighted sequential logistic regression.

RESULTS

An estimated 174,309 preschoolers (2.0%) were suspended, and 17,248 (0.2%) children were expelled annually. If divided by 36 school weeks, the instances of weekly suspension and expulsion were at least 4,842 and 479 respectively. Controlling for previous risk factors (i.e., age, gender, race, ethnicity), the odds ratio increased by 80% for every unit of ACEs increment. Children were more likely to be suspended or expelled if they had domestic violence (OR = 10.6, p <  .001), living with mental illness (OR = 9.8, p <  .001), adult substance abuse (OR = 4.8, p <  .001), and victim of violence (OR = 4.5, p =  .004), living in high poverty (OR = 3.9, p =  .001), divorced parents (OR = 3.3, p =  .001), and parent incarceration (OR = 3.0, p =  .009).

CONCLUSION

The alarming suspension and expulsion rates call for more comprehensive outreach prevention and response efforts in preschool settings. Cross system collaboration and family support are essential to this work.

Treatment Utilization Trajectories among Children with Autism Spectrum Disorder: Differences by Race-Ethnicity and Neighborhood

Health coverage of early intensive behavioral intervention (EIBI) for children with autism spectrum disorder is expanding. Yet there is no longitudinal research on patterns of or inequities in utilization of EIBI. We integrated state administrative records with Medicaid and Census data for children enrolled in an EIBI Medicaid waiver (N = 730) to identify and describe the type and prevalence of treatment utilization trajectories, and to examine the association between trajectory types and (a) child race-ethnicity and (b) neighborhood racial composition, poverty, affluence, and urbanicity. We identified four utilization trajectories (Low, Low-Moderate, Moderate, and High users). Race-ethnicity and neighborhood affluence were associated with trajectory membership. As coverage expands, policy makers should consider strategies to improve overall treatment utilization and enhance equity.

Application of Data Mining to "Big Data" Acquired in Audiology: Principles and Potential

The ubiquity and cheapness of miniature low-power sensors, digital processing, and large amounts of storage contained in small packages has heralded the ability to acquire large amounts of data about systems during their course of operation. The size and complexity of the data sets so generated have colloquially been labeled “big data.” The computer science field of “data mining” has arisen with the purpose of extracting meaning from such data, expressly looking for patterns that not only link historic observations but also predict future behavior. This overview article considers the process, techniques, and interpretation of data mining, with specific focus on its application in audiology. Modern hearing instruments contain data-logging technology to record data separate from the audio stream, such as the acoustic environments in which the device was being used and how the signal processing was consequently operating. Combined with details about the patient, such as the audiogram, the variety of data generated lends itself to a data mining approach. To date, reports of the use and interpretation of these data have been mostly constrained to questions such as looking for changes in patterns of daily use, or the degree and direction of volume control manipulation as the patient’s experience with a hearing aid changes. In this, and an accompanying results paper, the practical applications of some data mining techniques are described as applied to a large data set of examples of real-world device usage, as supplied by a hearing aid manufacturer.

Vector Autoregressive Models and Granger Causality in Time Series Analysis in Nursing Research: Dynamic Changes Among Vital Signs Prior to Cardiorespiratory Instability Events as an Example

BACKGROUND

Patients undergoing continuous vital sign monitoring (heart rate [HR], respiratory rate [RR], pulse oximetry [SpO2]) in real time display interrelated vital sign changes during situations of physiological stress. Patterns in this physiological cross-talk could portend impending cardiorespiratory instability (CRI). Vector autoregressive (VAR) modeling with Granger causality tests is one of the most flexible ways to elucidate underlying causal mechanisms in time series data.

PURPOSE

The purpose of this article is to illustrate the development of patient-specific VAR models using vital sign time series data in a sample of acutely ill, monitored, step-down unit patients and determine their Granger causal dynamics prior to onset of an incident CRI.

APPROACH

CRI was defined as vital signs beyond stipulated normality thresholds (HR = 40-140/minute, RR = 8-36/minute, SpO2 < 85%) and persisting for 3 minutes within a 5-minute moving window (60% of the duration of the window). A 6-hour time segment prior to onset of first CRI was chosen for time series modeling in 20 patients using a six-step procedure: (a) the uniform time series for each vital sign was assessed for stationarity, (b) appropriate lag was determined using a lag-length selection criteria, (c) the VAR model was constructed, (d) residual autocorrelation was assessed with the Lagrange Multiplier test, (e) stability of the VAR system was checked, and (f) Granger causality was evaluated in the final stable model.

RESULTS

The primary cause of incident CRI was low SpO2 (60% of cases), followed by out-of-range RR (30%) and HR (10%). Granger causality testing revealed that change in RR caused change in HR (21%; i.e., RR changed before HR changed) more often than change in HR causing change in RR (15%). Similarly, changes in RR caused changes in SpO2 (15%) more often than changes in SpO2 caused changes in RR (9%). For HR and SpO2, changes in HR causing changes in SpO2 and changes in SpO2 causing changes in HR occurred with equal frequency (18%).

DISCUSSION

Within this sample of acutely ill patients who experienced a CRI event, VAR modeling indicated that RR changes tend to occur before changes in HR and SpO2. These findings suggest that contextual assessment of RR changes as the earliest sign of CRI is warranted. Use of VAR modeling may be helpful in other nursing research applications based on time series data.

Parent Management Training and Collaborative & Proactive Solutions: A Randomized Control Trial for Oppositional Youth

This study examined the efficacy of Collaborative & Proactive Solutions (CPS) in treating oppositional defiant disorder (ODD) in youth by comparing this novel treatment to Parent Management Training (PMT), a well-established treatment, and a waitlist control (WLC) group. One hundred thirty-four youth (ages 7-14, 61.9% male, 83.6% White) who fulfilled Diagnostic and Statistical Manual of Mental Disorders (4th ed.) criteria for ODD were randomized to CPS, PMT, or WLC groups. ODD was assessed with semistructured diagnostic interviews, clinical global severity and improvement ratings, and parent report measures. Assessments were completed pretreatment, posttreatment, and at 6 months following treatment. Responder and remitter analyses were undertaken using intent-to-treat mixed-models analyses. Chronological age, gender, and socioeconomic status as well as the presence of comorbid attention deficit/hyperactivity and anxiety disorders were examined as predictors of treatment outcome. Both treatment conditions were superior to the WLC condition but did not differ from one another in either our responder or remitter analyses. Approximately 50% of youth in both active treatments were diagnosis free and were judged to be much or very much improved at posttreatment, compared to 0% in the waitlist condition. Younger age and presence of an anxiety disorder predicted better treatment outcomes for both PMT and CPS. Treatment gains were maintained at 6-month follow-up. CPS proved to be equivalent to PMT and can be considered an evidence-based, alternative treatment for youth with ODD and their families.

Open challenges for data stream mining research

Every day, huge volumes of sensory, transactional, and web data are continuously generated as streams, which need to be analyzed online as they arrive. Streaming data can be considered as one of the main sources of what is called big data. While predictive modeling for data streams and big data have received a lot of attention over the last decade, many research approaches are typically designed for well-behaved controlled problem settings, overlooking important challenges imposed by real-world applications. This article presents a discussion on eight open challenges for data stream mining. Our goal is to identify gaps between current research and meaningful applications, highlight open problems, and define new application-relevant research directions for data stream mining. The identified challenges cover the full cycle of knowledge discovery and involve such problems as: protecting data privacy, dealing with legacy systems, handling incomplete and delayed information, analysis of complex data, and evaluation of stream mining algorithms. The resulting analysis is illustrated by practical applications and provides general suggestions concerning lines of future research in data stream mining.