Predicting in-hospital mortality after transcatheter aortic valve replacement using administrative data and machine learning.

Transcatheter aortic valve replacement (TAVR) is the gold standard treatment for patients with symptomatic aortic stenosis. The utility of existing risk prediction tools for in-hospital mortality post-TAVR is limited due to two major factors: (a) the predictive accuracy of these tools is insufficient when only preoperative variables are incorporated, and (b) their efficacy is also compromised when solely postoperative variables are employed, subsequently constraining their application in preoperative decision support. This study examined whether statistical/machine learning models trained with solely preoperative information encoded in the administrative National Inpatient Sample database could accurately predict in-hospital outcomes (death/survival) post-TAVR. Fifteen popular binary classification methods were used to model in-hospital survival/death. These methods were evaluated using multiple classification metrics, including the area under the receiver operating characteristic curve (AUC). By analyzing 54,739 TAVRs, the top five classification models had an AUC ≥ 0.80 for two sampling scenarios: random, consistent with previous studies, and time-based, which assessed whether the models could be deployed without frequent retraining. Given the minimal practical differences in the predictive accuracies of the top five models, the L2 regularized logistic regression model is recommended as the best overall model since it is computationally efficient and easy to interpret.

The relationship between ratings of perceived exertion (RPE) and relative strength for a fatiguing dynamic upper extremity task: A consideration of multiple cycles and conditions.

The goal of this study was to evaluate the relationship between ratings of perceived exertion (RPE) and relative strength with respect to baseline for a fatiguing free dynamic task targeting the upper extremity, namely simulated order picking, and determine whether the relationship remains the same for different conditions (i.e., pace and weight) and with fatigue. Fourteen participants (seven males, seven females) performed four sessions that included two 45-min work periods separated by 15 min of rest. The work periods involved picking weighted bottles from shoulder height and packaging them at waist height for four combinations of bottle mass and picking rate: 2.5 kg-15 bottles per minute (bpm), 2.5 kg-10 bpm, 2.5 kg-5 bpm, and 1.5 kg-15 bpm. Participants reported their RPEs every 5 min and performed a maximum isometric shoulder flexion exertion every 9 min. Pearson product-moment correlation was used to evaluate the linear relationship between RPE and relative strength for each subject and work period. Then, the effects of condition and work period on the average relationship were assessed using a repeated-measures analysis of variance (ANOVA). For the first 45-min period, there were no significantly different correlations between RPE and relative strength across conditions (average r = -0.62 (standard deviation = 0.38); p = 0.57). There was a significant decrease in average correlation for the second work period (r = -0.39 (0.53)). These results suggest that individual subjective responses consistently increase while relative strength declines when starting from a non-fatigued state. However, correlations are weaker when re-engaging in work following incomplete recovery. Thus, starting fatigue levels should be accounted for when considering the expected relationship between RPE and relative strength.

Explaining Predictive Model Performance: An Experimental Study of Data Preparation and Model Choice.

Although confirmatory modeling has dominated much of applied research in medical, business, and behavioral sciences, modeling large data sets with the goal of accurate prediction has become more widely accepted. The current practice for fitting predictive models is guided by heuristic-based modeling frameworks that lead researchers to make a series of often isolated decisions regarding data preparation and cleaning that may result in substandard predictive performance. In this article, we use an experimental design to evaluate the impact of six factors related to data preparation and model selection (techniques for numerical imputation, categorical imputation, encoding, subsampling for unbalanced data, feature selection, and machine learning algorithm) and their interactions on the predictive accuracy of models applied to a large, publicly available heart transplantation database. Our factorial experiment includes 10,800 models evaluated on 5 independent test partitions of the data. Results confirm that some decisions made early in the modeling process interact with later decisions to affect predictive performance; therefore, the current practice of making these decisions independently can negatively affect predictive outcomes. A key result of this case study is to highlight the need for improved rigor in applied predictive research. By using the scientific method to inform predictive modeling, we can work toward a framework for applied predictive modeling and a standard for reproducibility in predictive research.

Functional ANOVA for Upper Extremity Fatigue Analysis during Dynamic Order Picking.

OCCUPATIONAL APPLICATIONSMusculoskeletal disorders are prevalent among warehouse workers who engage in repetitive and dynamic tasks. To prevent such injuries, it is vital to identify the factors that influence fatigue in the upper extremities during these repetitive activities. Our study reveals that task factors, namely the bottle mass and picking rate, significantly influence upper extremity fatigue. In most cases, the fatigue indicator is a functional variable, meaning that the fatigue score or measurement is a curve captured over time, which could be modeled as a function. In this study, we demonstrate that functional data analysis tools, such as functional analysis of variance (FANOVA), prove more effective than traditional methods in specifying how task factors contribute to the development of fatigue in the upper extremities. Furthermore, since there are inherent differences among workers that could affect their fatigue development process, the data heterogeneity could be tackled by employing clustering methods.

Background: Preventing musculoskeletal disorders is a paramount safety concern for industries, with order pickers in warehouses being particularly vulnerable due to their repetitive and dynamic tasks. Understanding the factors contributing to upper-extremity fatigue in such settings is crucial. Purpose: This paper investigates the impact of task-related factors on two upper-extremity fatigue indicators: ratings of perceived fatigue and relative muscle strength. Several statistical approaches were used and compared in terms of their capability in eliciting these effects. Methods: Simulated over-shoulder, order-picking lab experiments were conducted under different combinations of two bottle loads and three picking paces. Fourteen participants, evenly distributed between genders, completed the experiment. A FANOVA was executed as the principal analytical approach, considering the functional nature of the two fatigue indicators measured over the work period. To underscore the benefits of considering the whole functional curve instead of discrete variables, we also conducted repeated-measures and two-way ANOVA as benchmark analyses. Results: FANOVA outcomes affirmed that both task factors (load and pace) significantly influenced both fatigue indicators. The FANOVA method identified larger effect sizes (0.11< ηp2 < 0.19) for both task factors compared to the conventional methods (0< ηp2 < 0.11), supporting the efficacy of FANOVA in identifying the importance of these factors. Conclusions: The FANOVA approach proved effective in detecting the impact of task factors on fatigue indicators, yielding superior results compared to conventional benchmark methods. To address participant heterogeneity, functional clustering and gender-based clustering were introduced into the FANOVA framework, both effectively mitigating this challenge. Notably, FANOVA with functional clusters had superior performance compared to the one with gender clustering, suggesting functional clustering as a more suitable method in overcoming participant heterogeneity.

Explaining the Varying Patterns of COVID-19 Deaths Across the United States: 2-Stage Time Series Clustering Framework.

BACKGROUND: Socially vulnerable communities are at increased risk for adverse health outcomes during a pandemic. Although this association has been established for H1N1, Middle East respiratory syndrome (MERS), and COVID-19 outbreaks, understanding the factors influencing the outbreak pattern for different communities remains limited. OBJECTIVE: Our 3 objectives are to determine how many distinct clusters of time series there are for COVID-19 deaths in 3108 contiguous counties in the United States, how the clusters are geographically distributed, and what factors influence the probability of cluster membership. METHODS: We proposed a 2-stage data analytic framework that can account for different levels of temporal aggregation for the pandemic outcomes and community-level predictors. Specifically, we used time-series clustering to identify clusters with similar outcome patterns for the 3108 contiguous US counties. Multinomial logistic regression was used to explain the relationship between community-level predictors and cluster assignment. We analyzed county-level confirmed COVID-19 deaths from Sunday, March 1, 2020, to Saturday, February 27, 2021. RESULTS: Four distinct patterns of deaths were observed across the contiguous US counties. The multinomial regression model correctly classified 1904 (61.25%) of the counties' outbreak patterns/clusters. CONCLUSIONS: Our results provide evidence that county-level patterns of COVID-19 deaths are different and can be explained in part by social and political predictors.

A data analytic end-to-end framework for the automated quantification of ergonomic risk factors across multiple tasks using a single wearable sensor.

Existing ergonomic risk assessment tools require monitoring of multiple risk factors. To eliminate the direct observation, we investigated the effectiveness of an end-to-end framework that works with the data from a single wearable sensor. The framework is used to identify the performed task as the major contextual risk factor, and then estimate the task duration and number of repetitions as two main indicators of task intensity. For evaluation of the framework, we recruited 37 participants to complete 10 simulated work tasks in a laboratory setting. In testing, we achieved an average accuracy of 92% for task identification, 7.3% error in estimation of task duration, and 7.1% error for counting the number of task repetitions. Moreover, we showed the utility of the framework outputs in two ergonomic tools to estimate the risk of injury. Overall, we indicated the feasibility of using data from wearable sensors to automate the ergonomic risk assessment in workplaces.

Modeling the differences in the time-series profiles of new COVID-19 daily confirmed cases in 3,108 contiguous U.S. counties: A retrospective analysis.

OBJECTIVE: The COVID-19 pandemic in the U.S. has exhibited a distinct multiwave pattern beginning in March 2020. Paradoxically, most counties do not exhibit this same multiwave pattern. We aim to answer three research questions: (1) How many distinct clusters of counties exhibit similar COVID-19 patterns in the time-series of daily confirmed cases? (2) What is the geographic distribution of the counties within each cluster? and (3) Are county-level demographic, socioeconomic and political variables associated with the COVID-19 case patterns? MATERIALS AND METHODS: We analyzed data from counties in the U.S. from March 1, 2020 to January 2, 2021. Time series clustering identified clusters in the daily confirmed cases of COVID-19. An explanatory model was used to identify demographic, socioeconomic and political variables associated with the outbreak patterns. RESULTS: Three patterns were identified from the cluster solution including counties in which cases are still increasing, those that peaked in the late fall, and those with low case counts to date. Several county-level demographic, socioeconomic, and political variables showed significant associations with the identified clusters. DISCUSSION: The pattern of the outbreak is related both to the geographic location within the U.S. and several variables including population density and government response. CONCLUSION: The reported pattern of cases in the U.S. is observed through aggregation of the daily confirmed COVID-19 cases, suggesting that local trends may be more informative. The pattern of the outbreak varies by county, and is associated with important demographic, socioeconomic, political and geographic factors.

Investigation of Heterogeneity Sources for Occupational Task Recognition via Transfer Learning.

Human activity recognition has been extensively used for the classification of occupational tasks. Existing activity recognition approaches perform well when training and testing data follow an identical distribution. However, in the real world, this condition may be violated due to existing heterogeneities among training and testing data, which results in degradation of classification performance. This study aims to investigate the impact of four heterogeneity sources, cross-sensor, cross-subject, joint cross-sensor and cross-subject, and cross-scenario heterogeneities, on classification performance. To that end, two experiments called separate task scenario and mixed task scenario were conducted to simulate tasks of electrical line workers under various heterogeneity sources. Furthermore, a support vector machine classifier equipped with domain adaptation was used to classify the tasks and benchmarked against a standard support vector machine baseline. Our results demonstrated that the support vector machine equipped with domain adaptation outperformed the baseline for cross-sensor, joint cross-subject and cross-sensor, and cross-subject cases, while the performance of support vector machine equipped with domain adaptation was not better than that of the baseline for cross-scenario case. Therefore, it is of great importance to investigate the impact of heterogeneity sources on classification performance and if needed, leverage domain adaptation methods to improve the performance.

Predicting unsafe driving risk among commercial truck drivers using machine learning: Lessons learned from the surveillance of 20 million driving miles.

The emergence of sensor-based Internet of Things (IoT) monitoring technologies have paved the way for conducting large-scale naturalistic driving studies, where continuous kinematic driver-based data are generated, capturing crash/near-crash safety critical events (SCEs) and their precursors. However, it is unknown whether the SCEs risk can be predicted to inform driver decisions in the medium term (e.g., hours ahead) since the literature has focused on SCE predictions either for a given road segment or for automated breaking applications, i.e., immediately before the event. In this paper, we examine the SCE data generated from 20+ million miles-driven by 496 commercial truck drivers to address three main questions. First, whether SCEs can be predicted using disparate driving-related data sources. Second, if so, what the relative importance of the different predictors examined is. Third, whether the prediction models can be generalized to new drivers and future time periods. We show that SCEs can be predicted 30 min in advance, using machine learning techniques and dependent variables capturing the driver's characteristics, weather conditions, and day/time categories, where an area under the curve (AUC) up to 76% can be achieved. Moreover, the predictive performance remains relatively stable when tested on new (i.e., not in the training set) drivers and a future two-month time period. Our results can inform dispatching and routing applications, and lead to the development of technological interventions to improve driver safety.

Seat Assignments With Physical Distancing in Single-Destination Public Transit Settings.

While the importance of physical (social) distancing in reducing the spread of COVID-19 has been well-documented, implementing similar controls in public transit remains an open question. For instance, in the United States, guidance for maximum seating capacity in single-destination public transit settings, such as school buses, is only dependent on the physical distance between passengers. In our estimation, the available models/guidance are suboptimal/inefficient since they do not account for the possibility of passengers being from the same household. This paper discusses and addresses the aforementioned limitation through two types of physical distancing models. First, a mixed-integer programming model is used to assign passengers to seats based on the reported configuration of the vehicle and desired physical distancing requirement. In the second model, we present a heuristic that allows for household grouping. Through several illustrative scenarios, we show that seating assignments can be generated in near real-time, and the household grouping heuristic increases the capacity of the transit vehicles (e.g., airplanes, school buses, and trains) without increasing the risk of infection. A running application and its source code are available to the public to facilitate adoption and to encourage enhancements.

A forecasting framework for predicting perceived fatigue: Using time series methods to forecast ratings of perceived exertion with features from wearable sensors.

Advancements in sensing and network technologies have increased the amount of data being collected to monitor the worker conditions. In this study, we consider the use of time series methods to forecast physical fatigue using subjective ratings of perceived exertion (RPE) and gait data from wearable sensors captured during a simulated in-lab manual material handling task (Lab Study 1) and a fatiguing squatting with intermittent walking cycle (Lab Study 2). To determine whether time series models can accurately forecast individual response and for how many time periods ahead, five models were compared: naïve method, autoregression (AR), autoregressive integrated moving average (ARIMA), vector autoregression (VAR), and the vector error correction model (VECM). For forecasts of three or more time periods ahead, the VECM model that incorporates historical RPE and wearable sensor data outperformed the other models with median mean absolute error (MAE) <1.24 and median MAE <1.22 across all participants for Lab Study 1 and Lab Study 2, respectively. These results suggest that wearable sensor data can support forecasting a worker's condition and the forecasts obtained are as good as current state-of-the-art models using multiple sensors for current time prediction.

Interventions to Mitigate Fatigue Induced by Physical Work: A Systematic Review of Research Quality and Levels of Evidence for Intervention Efficacy.

OBJECTIVE: We present a literature review on workplace physical fatigue interventions, focusing on evaluating the methodological quality and strength of evidence. BACKGROUND: Physical fatigue is a recognized workplace problem, with negative effects on performance and health-related complaints. Although many studies have focused on the mechanisms and consequences of fatigue, few have considered the effectiveness of interventions to mitigate fatigue. METHOD: A systematic review of the workplace safety literature for controlled trials of physical fatigue interventions was conducted. Data on intervention type, subject characteristics, targeted tasks and body locations, outcome measures, and study design were extracted. The methodological quality for each study was evaluated using the PEDro scale, and the level of evidence was based on quality, amount, and consistency. RESULTS: Forty-five controlled trials were reviewed, examining 18 interventions. We categorized those interventions into individual-focused (N = 28 studies, nine interventions), workplace-focused (N = 12 studies, five interventions), and multiple interventions (N = 5 studies, four interventions). We identified moderate evidence for interventions related to assistive devices and task variation. There was moderate evidence supporting no fatigue attenuation for the garment change category of interventions. The interventions in the remaining categories had limited to minimal evidence of efficacy. The heterogeneity of the included trials precludes the determination of effect size. CONCLUSION: This review showed a lack of high levels of evidence for the effectiveness of most physical fatigue interventions. APPLICATION: Due to a lack of high levels of evidence for any category of reviewed physical fatigue interventions, further high-quality studies are needed to establish the efficacy of others.

A Review of Data Analytic Applications in Road Traffic Safety. Part 1: Descriptive and Predictive Modeling.

This part of the review aims to reduce the start-up burden of data collection and descriptive analytics for statistical modeling and route optimization of risk associated with motor vehicles. From a data-driven bibliometric analysis, we show that the literature is divided into two disparate research streams: (a) predictive or explanatory models that attempt to understand and quantify crash risk based on different driving conditions, and (b) optimization techniques that focus on minimizing crash risk through route/path-selection and rest-break scheduling. Translation of research outcomes between these two streams is limited. To overcome this issue, we present publicly available high-quality data sources (different study designs, outcome variables, and predictor variables) and descriptive analytic techniques (data summarization, visualization, and dimension reduction) that can be used to achieve safer-routing and provide code to facilitate data collection/exploration by practitioners/researchers. Then, we review the statistical and machine learning models used for crash risk modeling. We show that (near) real-time crash risk is rarely considered, which might explain why the optimization models (reviewed in Part 2) have not capitalized on the research outcomes from the first stream.

A Review of Data Analytic Applications in Road Traffic Safety. Part 2: Prescriptive Modeling.

In the first part of the review, we observed that there exists a significant gap between the predictive and prescriptive models pertaining to crash risk prediction and minimization, respectively. In this part, we review and categorize the optimization/ prescriptive analytic models that focus on minimizing crash risk. Although the majority of works in this segment of the literature are related to the hazardous materials (hazmat) trucking problems, we show that (with some exceptions) many can also be utilized in non-hazmat scenarios. In an effort to highlight the effect of crash risk prediction model on the accumulated risk obtained from the prescriptive model, we present a simulated example where we utilize four risk indicators (obtained from logistic regression, Poisson regression, XGBoost, and neural network) in the k-shortest path algorithm. From our example, we demonstrate two major designed takeaways: (a) the shortest path may not always result in the lowest crash risk, and (b) a similarity in overall predictive performance may not always translate to similar outcomes from the prescriptive models. Based on the review and example, we highlight several avenues for future research.

A machine learning approach to detect changes in gait parameters following a fatiguing occupational task.

The purpose of this study is to provide a method for classifying non-fatigued vs. fatigued states following manual material handling. A method of template matching pattern recognition for feature extraction ($1 Recognizer) along with the support vector machine model for classification were applied on the kinematics of gait cycles segmented by our stepwise search-based segmentation algorithm. A single inertial measurement unit on the ankle was used, providing a minimally intrusive and inexpensive tool for monitoring. The classifier distinguished between states using distance-based scores from the recogniser and the step duration. The results of fatigue detection showed an accuracy of 90% across data from 20 recruited subjects. This method utilises the minimum amount of data and features from only one low-cost sensor to reliably classify the state of fatigue induced by a realistic manufacturing task using a simple machine learning algorithm that can be extended to real-time fatigue monitoring as a future technology to be employed in the manufacturing facilities. Practitioner Summary: We examined the use of a wearable sensor for the detection of fatigue-related changes in gait based on a simulated manual material handling task. Classification based on foot acceleration and position trajectories resulted in 90% accuracy. This method provides a practical framework for predicting realistic levels of fatigue.

A survey of the prevalence of fatigue, its precursors and individual coping mechanisms among U.S. manufacturing workers.

Advanced manufacturing has resulted in significant changes on the shop-floor, influencing work demands and the working environment. The corresponding safety-related effects, including fatigue, have not been captured on an industry-wide scale. This paper presents results of a survey of U.S. manufacturing workers for the: prevalence of fatigue, its root causes and significant factors, and adopted individual fatigue coping methods. The responses from 451 manufacturing employees were analyzed using descriptive data analysis, bivariate analysis and Market Basket Analysis. 57.9% of respondents indicated that they were somewhat fatigued during the past week. They reported the ankles/feet, lower back and eyes were frequently affected body parts and a lack of sleep, work stress and shift schedule were top selected root causes for fatigue. In order to respond to fatigue when it is present, respondents reported coping by drinking caffeinated drinks, stretching/doing exercises and talking with coworkers. Frequent combinations of fatigue causes and individual coping methods were identified. These results may inform the design of fatigue monitoring and mitigation strategies and future research related to fatigue development.

A data-driven approach to modeling physical fatigue in the workplace using wearable sensors.

Wearable sensors are currently being used to manage fatigue in professional athletics, transportation and mining industries. In manufacturing, physical fatigue is a challenging ergonomic/safety "issue" since it lowers productivity and increases the incidence of accidents. Therefore, physical fatigue must be managed. There are two main goals for this study. First, we examine the use of wearable sensors to detect physical fatigue occurrence in simulated manufacturing tasks. The second goal is to estimate the physical fatigue level over time. In order to achieve these goals, sensory data were recorded for eight healthy participants. Penalized logistic and multiple linear regression models were used for physical fatigue detection and level estimation, respectively. Important features from the five sensors locations were selected using Least Absolute Shrinkage and Selection Operator (LASSO), a popular variable selection methodology. The results show that the LASSO model performed well for both physical fatigue detection and modeling. The modeling approach is not participant and/or workload regime specific and thus can be adopted for other applications.