Workload Estimation Research Articles

WORKLOAD DYNAMICS OF MEDICAL PROFESSIONALS IN BULGARIAN HEALTHCARE SYSTEM: EVIDENCES FROM AGGREGATE DATA

Purpose. This study aims to analyze the tendencies of the workload of medical professionals in Bulgaria after the liberal reforms in the national healthcare system. Material/Methods. Basic econometric methods, such as tests for structural breaks and dynamic rows modeling based on "macro-" or aggregate data, are implemented. The period covered is 1980-2023, twenty years before and twenty-four years after introducing market elements in national healthcare. Variables such as the number of patients discharged, consultations, doctors, and nurses are used to assess the workload level. Workload estimates are derived from a linear regression model of hospital output as the dependent variable with the number of doctors and nurses in the health system as factors. The model is tested for the stability of the regression coefficients (structural breaks) dated 2000. Results. A structural break was found in 2000 in all three dynamic rows of doctors, nurses, and discharges. There was a well-defined increasing trend in the number of patients passing through the hospitals after the liberalization of the national healthcare system. In this condition, the number of medical professionals follows its own dynamics – a slight reduction for the group of doctors and a sharp decline for the nurses in the after-reforms sub-period. These facts indicate the presence of an increasing trend in the workload and intensity of medical labour. The dependence is well established, especially in the case of nurses. Conclusions. This study confirms the hypothesis that the liberalization of the Bulgarian healthcare system, together with open-ended payment systems, caused an intensive growth of inpatient system output and workload. The expansion, especially for hospital facilities, which is not supported by a relevant number of medical professionals, threatens workplace performance and the quality of medical services.

Estimating cognitive workload using a commercial in-ear EEG headset

Objective.This study investigated the potential of estimating various mental workload levels during two different tasks using a commercial in-ear electroencephalography (EEG) system, the IDUN 'Guardian'.Approach.Participants performed versions of two classical workload tasks: an n-back task and a mental arithmetic task. Both in-ear and conventional EEG data were simultaneously collected during these tasks. In an effort to facilitate a more comprehensive comparison, the complexity of the tasks was intentionally increased beyond typical levels. Special emphasis was also placed on understanding the significance ofγband activity in workload estimations. Therefore, each signal was analyzed across low frequency (1-35 Hz) and high frequency (1-100 Hz) ranges. Additionally, surrogate in-ear EEG measures, derived from the conventional EEG recordings, were extracted and examined.Main results.Workload estimation using in-ear EEG yielded statistically significant performance levels, surpassing chance levels with 44.1% for four classes and 68.4% for two classes in the n-back task and was better than a naive predictor for the mental arithmetic task. Conventional EEG exhibited significantly higher performance compared to in-ear EEG, achieving 80.3% and 92.9% accuracy for the respective tasks, along with lower error rates than the naive predictor. The developed surrogate measures achieved improved results, reaching accuracies of 57.5% and 85.5%, thus providing insights for enhancing future in-ear systems. Notably, most high frequency range signals outperformed their low frequency counterparts in terms of accuracy validating that high frequencyγband features can improve workload estimation.Significance.The application of EEG-based Brain-Computer Interfaces beyond laboratory settings is often hindered by practical limitations. In-ear EEG systems offer a promising solution to this problem, potentially enabling everyday use. This study evaluates the performance of a commercial in-ear headset and provides guidelines for increased effectiveness.

Self-Supervised Learning for Near-Wild Cognitive Workload Estimation.

Feedback on cognitive workload may reduce decision-making mistakes. Machine learning-based models can produce feedback from physiological data such as electroencephalography (EEG) and electrocardiography (ECG). Supervised machine learning requires large training data sets that are (1) relevant and decontaminated and (2) carefully labeled for accurate approximation, a costly and tedious procedure. Commercial over-the-counter devices are low-cost resolutions for the real-time collection of physiological modalities. However, they produce significant artifacts when employed outside of laboratory settings, compromising machine learning accuracies. Additionally, the physiological modalities that most successfully machine-approximate cognitive workload in everyday settings are unknown. To address these challenges, a first-ever hybrid implementation of feature selection and self-supervised machine learning techniques is introduced. This model is employed on data collected outside controlled laboratory settings to (1) identify relevant physiological modalities to machine approximate six levels of cognitive-physical workloads from a seven-modality repository and (2) postulate limited labeling experiments and machine approximate mental-physical workloads using self-supervised learning techniques.

The National Emergency Department Overcrowding Scale and Perceived Staff Workload: Evidence for Construct Validity in a Pediatric Setting.

The aim of the study is to determine if there is a correlation between perceived staff workload, measured by the NASA Task Load Index (TLX), and the National Emergency Department Overcrowding Scale (NEDOCS) in a pediatric ED. We collected staff questionnaires in a large, urban pediatric ED to assess perceived workload on each of six different TLX subscales, which we weighted evenly to create an overall estimate of workload. We evaluated the correlation between individual TLX responses and NEDOCS overall and by staff subgroup. Additionally, we analyzed: (1) the correlation between mean TLX responses and NEDOCS within a given hour and (2) the performance of a logistic regression model, using TLX as a predictor for "severely overcrowded," as measured by NEDOCS. Four hundred one questionnaires between 6/2018 and 1/2019 demonstrated significant variation between concurrently collected TLX responses and an overall poor correlation between perceived workload and NEDOCS (R2 0.096 [95% confidence interval, 0.048-0.16]). TLX responses by subgroups of fellows (n = 4, R2 0.96) and patient financial advisors (n = 15, R2 0.58) demonstrated the highest correlation with NEDOCS. Taking mean TLX responses within a given hour, during periods with NEDOCS >60 (extremely busy or overcrowded), a polynomial trend line matched the data best (R2 0.638). On logistic regression, the TLX predicts "severely overcrowded" with an area under the curve of the receiver operating characteristic of 0.731. NEDOCS does not have a strong correlation with individual responses on questionnaires of perceived workload for staff in a pediatric ED. NEDOCS, as a measure of overcrowding, may be better correlated with perceived workload during periods with elevated crowding or when interpreted categorically as yes/no "severely overcrowded".

Data-Driven Teamwork

Group assignments are a staple in higher education, providing opportunities to enhance collaboration and communication skills. However, we frequently encounter grievances related to group work as coordinators of large STEM units (800 students). Our student survey identifies work ethic and communication discrepancies between group members as barriers, often leading to uneven workloads, dissatisfaction, and diminished collaborative learning (Bacon et al., 1999). Common mitigation strategies to address work ethic discrepancies, like group charters and peer evaluations, often prove ineffective, since they could be ignored if not properly followed up or enforced (Chang & Brickman, 2018). Another barrier to group productivity lies in the team formation process itself. Traditional team formation methods – self-selection, random assignment, or instructor-led grouping – each present inherent limitation. Self-selection often results in homogenous groups susceptible to "groupthink" (Janis, 1982), limiting the diversity of thought essential for creative problem-solving. Random assignment, while seemingly impartial, can lead to imbalanced skillsets and an uneven distribution of workload (Bacon et al., 1999). While instructor-led grouping might seem ideal, it is logistically impractical within large cohorts (Decker, 1995). Computer-aided team selection offers a promising alternative. Its automated process streamlines team formation, generates diverse teams, and allows instructors to focus on teaching, despite concerns about increased workload (Cavanaugh et al., 2004). To address these group work challenges, we have developed a multi-pronged strategy integrating computer-aided team selection with a comprehensive follow-up and feedback framework. Our goal is to transform group work from a source of anxiety to a rewarding learning experience. Our approach combines technology with strategic interventions: Computer-aided Team Formation: We utilise Gruepr, an open-source team selection software (Hertz, 2021), to generate groups based on student preferences including work styles and meeting preferences (Cavanaugh et al., 2004). Gruepr integrates seamlessly with our learning management system, Canvas. Structured Communication, Collaboration, and Proactive Monitoring: Regular temperature-check surveys deployed through Canvas allow students to rate their group's progress and workload distribution. This data enables early identification of potential issues. To encourage group dialogue, the aggregated and anonymised outcome is shared with team members Facilitating Peer Feedback and Communication through SRES: The Student Relationship Engagement System (SRES), a learning analytics tool (Liu et al., 2017), was used to communicate aggregated peer feedback to individual group members, encouraging constructive dialogues regarding group progress. Our evaluation compared student performance, feedback, and complaints from the current year to previous years, alongside an estimation of staff workload. Preliminary findings demonstrate a reduction in staff time spent on group formation using Gruepr's automated grouping, highly contingent upon accurate survey design. Complaints, though still reported, were less severe, suggesting that regular temperature checks promoted proactive communication. Importantly, student feedback provided valuable insight, with many groups identifying the need for increased in-person interaction and improved scheduling strategies to accommodate diverse commitments. Our poster will elaborate on these results, showcasing both successes and challenges, and propose a refined workflow to further enhance collaborative learning experiences.

Data-Driven Teamwork

Group assignments are a staple in higher education, providing opportunities to enhance collaboration and communication skills. However, we frequently encounter grievances related to group work as coordinators of large STEM units (800 students). Our student survey identifies work ethic and communication discrepancies between group members as barriers, often leading to uneven workloads, dissatisfaction, and diminished collaborative learning (Bacon et al., 1999). Common mitigation strategies to address work ethic discrepancies, like group charters and peer evaluations, often prove ineffective, since they could be ignored if not properly followed up or enforced (Chang & Brickman, 2018). Another barrier to group productivity lies in the team formation process itself. Traditional team formation methods – self-selection, random assignment, or instructor-led grouping – each present inherent limitation. Self-selection often results in homogenous groups susceptible to "groupthink" (Janis, 1982), limiting the diversity of thought essential for creative problem-solving. Random assignment, while seemingly impartial, can lead to imbalanced skillsets and an uneven distribution of workload (Bacon et al., 1999). While instructor-led grouping might seem ideal, it is logistically impractical within large cohorts (Decker, 1995). Computer-aided team selection offers a promising alternative. Its automated process streamlines team formation, generates diverse teams, and allows instructors to focus on teaching, despite concerns about increased workload (Cavanaugh et al., 2004). To address these group work challenges, we have developed a multi-pronged strategy integrating computer-aided team selection with a comprehensive follow-up and feedback framework. Our goal is to transform group work from a source of anxiety to a rewarding learning experience. Our approach combines technology with strategic interventions: Computer-aided Team Formation: We utilise Gruepr, an open-source team selection software (Hertz, 2021), to generate groups based on student preferences including work styles and meeting preferences (Cavanaugh et al., 2004). Gruepr integrates seamlessly with our learning management system, Canvas. Structured Communication, Collaboration, and Proactive Monitoring: Regular temperature-check surveys deployed through Canvas allow students to rate their group's progress and workload distribution. This data enables early identification of potential issues. To encourage group dialogue, the aggregated and anonymised outcome is shared with team members Facilitating Peer Feedback and Communication through SRES: The Student Relationship Engagement System (SRES), a learning analytics tool (Liu et al., 2017), was used to communicate aggregated peer feedback to individual group members, encouraging constructive dialogues regarding group progress. Our evaluation compared student performance, feedback, and complaints from the current year to previous years, alongside an estimation of staff workload. Preliminary findings demonstrate a reduction in staff time spent on group formation using Gruepr's automated grouping, highly contingent upon accurate survey design. Complaints, though still reported, were less severe, suggesting that regular temperature checks promoted proactive communication. Importantly, student feedback provided valuable insight, with many groups identifying the need for increased in-person interaction and improved scheduling strategies to accommodate diverse commitments. Our presentation will elaborate on these results, showcasing both successes and challenges, and propose a refined workflow to further enhance collaborative learning experiences.

EEG workload estimation and classification: a systematic review.

Electroencephalography (EEG) has evolved into an indispensable instrument for estimating cognitive workload in various domains. ML and DL techniques have been increasingly employed to develop accurate workload estimation and classification models based on EEG data. The goal of this systematic review is to compile the body of research on EEG workload estimation and classification using ML and DL approaches. The PRISMA procedures were followed in conducting the review, searches were conducted through databases at SpringerLink, ACM Digital Library, IEEE Explore, PUBMED, and Science Direct from the beginning to the end of February 16, 2024. Studies were selected based on predefined inclusion criteria. Data were extracted to capture study design, participant demographics, EEG features, ML/DL algorithms, and reported performance metrics. Out of the 125 items that emerged, 33 scientific papers were fully evaluated. The study designs, participant demographics, and EEG workload measurement and categorization techniques used in the investigations differed. SVM, CNN, and hybrid networks are examples of ML and DL approaches that were often used. Analyzing the accuracy scores achieved by different ML/DL models. Furthermore, a relationship was noted between sample frequency and model accuracy, with higher sample frequencies generally leading to improved performance. The percentage distribution of ML/DL methods revealed that SVMs, CNNs, and RNNs were the most commonly utilized techniques, reflecting their robustness in handling EEG data. The comprehensive review emphasizes how ML may be used to identify mental workload across a variety of disciplines using EEG data. Optimizing practical applications requires multimodal data integration, standardization efforts, and real-world validation studies. These systems will also be further improved by addressing ethical issues and investigating new EEG properties, which will improve human-computer interaction and performance assessment.

Intensive care unit nursing workload estimation in smart hospitals

PurposeThe objectives of the proposed model are: aiding nursing staff in documentation tasks, which can be onerous and stressful; and helping management by offering an estimate of the nursing workload, which can be considered for administrative purposes, such as staff scheduling.Design/methodology/approachAn exploratory-descriptive study was conducted in order to identify, investigate, and describe the problem of documenting nursing activities and workload estimation in an intensive care unit. Technological solutions were explored, and models were proposed to address these issues.FindingsCross-dataset experiments were performed, and the model was able to offer an adequate estimate of the nursing workload. The results suggest that continuous retraining is essential for maintaining high accuracy. While the proposed model was considered in the context of an adult ICU, it can be adapted to other contexts, such as elderly care.Research limitations/implicationsWhile the proposed solution seems promising, further research is required, such as deploying this system in an ICU and facing challenges in the areas of computer security, medical ethics, and patient data privacy. More patients’ variables could also be collected to improve the workload estimates.Originality/valueNursing workload assessment is critical to improve the cost-benefit ratio in health care, offer high-quality patient care, and reduce unnecessary expenses, and this process is usually manual. An automated device can automatically document the amount of time spent in patient care activities in a more transparent, efficient, and accurate manner, freeing staff for more urgent activities and keeping management better informed about day-to-day nursing operations.

Using emergency department data to define a 'mental health presentation' - implications of different definitions on estimates of emergency department mental health workload.

Objective This study aimed to describe and compare the proportion of patients classified as an emergency department (ED) mental health presentation under different definitions, including the Australian Institute of Health and Welfare (AIHW) definition. Methods This retrospective cohort study enrolled all patients that presented to the EDs of a multi-centre Victorian health service between 1 January 2020 and 30 June 2023. Varying definitions of a mental health presentation were applied to each ED attendance, applying the current AIHW definition (using selected diagnosis codes), broader diagnosis-based coding, the presenting complaint recorded at triage and whether the patient was seen by or referred to the emergency psychiatric service (EPS). The proportion of all ED presentations meeting each definition and any overlap between definitions were calculated. The agreement between each definition and the AIHW definition was evaluated using Kappa's coefficient. Results There were 813,078 presentations to ED of which 34,248 (4.2%) met the AIHW definition for a mental health presentation. Throughout the study, 45,376 (5.6%) patients were seen and/or referred to EPS, and 36,160 (4.4%) patients were allocated a mental health presenting complaint by triage staff. There was moderate interrater agreement between these definitions, with a kappa statistic (95% confidence interval) between the AIHW definition and a mental health presenting complaint recorded at triage of 0.58 (0.58-0.59) and between the AIHW definition and review by EPS of 0.58 (0.57-0.58). Conclusions The AIHW definition is a conservative measure of ED mental health presentations and may underestimate emergency psychiatry workload in Australian EDs.

Classification of mental workload with EEG analysis by using effective connectivity and a hybrid model of CNN and LSTM

Estimation of mental workload from electroencephalogram (EEG) signals aims to accurately measure the cognitive demands placed on an individual during multitasking mental activities. By analyzing the brain activity of the subject, we can determine the level of mental effort required to perform a task and optimize the workload to prevent cognitive overload or underload. This information can be used to enhance performance and productivity in various fields such as healthcare, education, and aviation. In this paper, we propose a method that uses EEG and deep neural networks to estimate the mental workload of human subjects during multitasking mental activities. Notably, our proposed method employs subject-independent classification. We use the “STEW” dataset, which consists of two tasks, namely “No task” and “simultaneous capacity (SIMKAP)-based multitasking activity”. We estimate the different workload levels of two tasks using a composite framework consisting of brain connectivity and deep neural networks. After the initial preprocessing of EEG signals, an analysis of the relationships between the 14 EEG channels is conducted to evaluate effective brain connectivity. This assessment illustrates the information flow between various brain regions, utilizing the direct Directed Transfer Function (dDTF) method. Then, we propose a deep hybrid model based on pre-trained Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) for the classification of workload levels. The accuracy of the proposed deep model achieved 83.12% according to the subject-independent leave-subject-out (LSO) approach. The pre-trained CNN + LSTM approaches to EEG data have been found to be an accurate method for assessing the mental workload.

Design and Refinement of a Faculty Workload Estimator

ObjectiveThis paper describes the creation and refinement of a Workload Estimator to be used at Colleges/Schools of pharmacy to calculate faculty workload. MethodsWe describe the initial development of the Workload Estimator for pharmacy practice faculty, including the consensus-building methodology and the implementation process. Additionally, we describe the annual refinement and expansion of use to pharmaceutical and administrative sciences faculty. ResultsThe final product is an Excel file with built-in multiplication factors used to estimate pharmacy faculty workload. ConclusionThis Workload Estimator was identified as the most comprehensive and readily adaptable tool to estimate pharmacy faculty workload in July 2023 by the American Association of Colleges of Pharmacy, Council of Faculties, and Council of Deans Task Force on Faculty Workload. The Workload Estimator is currently being studied in a multi-institutional implementation study across dozens of diverse programs across the country with norming data and qualitative perspectives on its use being collected and shared with the Academy in 2024.

CPU-GPU Cooperative QoS Optimization of Personalized Digital Healthcare Using Machine Learning and Swarm Intelligence.

In recent decades, the rapid advances in information technology have promoted a widespread deployment of medical cyber-physical systems (MCPS), especially in the area of digital healthcare. In digital healthcare, medical edge devices empowered by CPU-GPU (Graphics Processing Unit) cooperative multiprocessor system-on-chips (MPSoCs) have a great potential in processing and managing the massive amounts of health-related data. However, most of the existing works on CPU-GPU cooperative MPSoCs cannot maintain a high-precision workload estimation since they simply leverage the worst-case execution cycles to pessimistically predict the workload of digital healthcare applications. Besides, they neglect the personalized requirements of individual healthcare applications and the lifetime reliability demands of heterogeneous CPU-GPU cores. As a result, the normal functions of medical edge devices and the quality-of-services (QoS) of digital healthcare applications are likely to suffer from underlying failures and degradation. In this paper, we explore CPU-GPU cooperative QoS optimization of personalized digital healthcare applications running on reliability guaranteed edge devices with the help of machine learning and swarm intelligence techniques. We first develop two novel predictors: one is a machine learning based predictor for application workload estimation, and the other is a feature-driven predictor for application QoS estimation. We then incorporate the two predictors into a swarm intelligent application scheduling scheme upon the cooperative dual-population evolutionary algorithm (c-DPEA) to find optimal application mapping and partitioning settings. Experimental results show that our solution not only augments the average QoS of whole digital healthcare applications by 15.7%, but also balances the QoS of individual digital healthcare applications by 64.3%.

Cataract surgery workload estimates in Theni district, India

Background/aimsTo estimate the annual cataract surgery workload in Theni district, Tamil Nadu, India based on current utilisation of cataract services, prevalence of blindness and vision impairment (VI), and cataract burden-reduction...

Cognitive Workload Estimation in Conditionally Automated Vehicles Using Transformer Networks Based on Physiological Signals

Though driving automation promises to improve driving safety, drivers are still required to be ready to retake control in conditionally automated vehicles, which are defined by the Society of Automotive Engineers (SAE) as SAE L3 vehicles. Thus, drivers’ states can still affect driving safety in SAE L3 vehicles. Previous research found that a high cognitive load may impair drivers’ takeover performance. Thus, it is still necessary to estimate drivers’ cognitive load in SAE L3 vehicles. However, existing driver cognitive load estimation algorithms mostly focus on vehicles with a lower level of driving automation (e.g., SAE L0), which may not be relevant when estimating driver states in SAE L3 vehicles, given that drivers’ responsibilities are different, and several commonly used measures (e.g., driving performance) are unavailable when drivers are not continuously controlling the vehicle. Further, previous driver cognitive load estimation algorithms rarely considered the temporal information in the input features. Thus, we proposed a deep-learning algorithm to estimate driver cognitive load in SAE L3 vehicles, which integrated multiple physiological features (i.e., electrocardiogram, electrodermal activity, respiration) and considered the temporal correlation of the data using a transformer-encoder-based network. The performance of our algorithm was compared with baseline models on an open data set. Results showed that our algorithm outperformed baseline models and achieved an accuracy of 94.4% using within-subject data partition (proportionally splitting data from the same subject into the training and testing data sets) and an accuracy of 89% using across-subjects data partition (dividing the training and testing data sets based on individual subjects).

Process mapping homicide investigations and estimating resource requirements for homicide units: Findings from a case study in the United States

A research team worked with a large urban jurisdiction in the southeastern United States, conducted detective interviews, developed process maps, and estimated the work hours associated with various phases of homicide investigations. The investigative process was organized into separate stages and presented as process maps. The average time elapsed for the initial investigative stage was 72.0 hours, while time to complete the second stage was 10.3 days. Plea-bargained cases consumed 12 hours of lead investigator’s time; homicide trials required an additional 13.2 workdays. These workflow maps and workload estimates could be used to help determine staffing allocations in similar agencies.

The Roman siege system of Masada: a 3D computerized analysis of a conflict landscape

AbstractThe 1st-c. CE Roman siege system of Masada exhibits a high degree of preservation due to its remote location and the arid climate. However, unlike the thoroughly excavated Masada fortress, the siege system has not received due attention. This article is part of a research project aimed at advancing our understanding of the conflict landscape around Masada using contemporary archaeological methods. Following a comprehensive surface survey and photogrammetric 3D modelling, we show that the circumvallation wall stood to a height of 2–2.5 m and served several functions – as an obstacle, a means of psychological warfare, and a platform from which to mount counterattacks. Based on our measurements and workload estimations, we argue that the construction of the siege wall and the camps around Masada occurred fairly quickly.

SmallMap: Low-cost Community Road Map Sensing with Uncertain Delivery Behavior

Accurate road networks play a crucial role in modern mobile applications such as navigation and last-mile delivery. Most existing studies primarily focus on generating road networks in open areas like main roads and avenues, but little attention has been given to the generation of community road networks in closed areas such as residential areas, which becomes more and more significant due to the growing demand for door-to-door services such as food delivery. This lack of research is primarily attributed to challenges related to sensing data availability and quality. In this paper, we design a novel framework called SmallMap that leverages ubiquitous multi-modal sensing data from last-mile delivery to automatically generate community road networks with low costs. Our SmallMap consists of two key modules: (1) a Trajectory of Interest Detection module enhanced by exploiting multi-modal sensing data collected from the delivery process; and (2) a Dual Spatio-temporal Generative Adversarial Network module that incorporates Trajectory of Interest by unsupervised road network adaptation to generate road networks automatically. To evaluate the effectiveness of SmallMap, we utilize a two-month dataset from one of the largest logistics companies in China. The extensive evaluation results demonstrate that our framework significantly outperforms state-of-the-art baselines, achieving a precision of 90.5%, a recall of 87.5%, and an F1-score of 88.9%, respectively. Moreover, we conduct three case studies in Beijing City for courier workload estimation, Estimated Time of Arrival (ETA) in last-mile delivery, and fine-grained order assignment.

Cognitive workload classification of law enforcement officers using physiological responses

Motor vehicle crashes (MVCs) are a leading cause of death for law enforcement officers (LEOs) in the U.S. LEOs and more specifically novice LEOs (nLEOs) are susceptible to high cognitive workload while driving which can lead to fatal MVCs. The objective of this study was to develop a machine learning algorithm (MLA) that can estimate cognitive workload of LEOs while performing secondary tasks in a patrol vehicle. A ride-along study was conducted with 24 nLEOs. Participants performed their normal patrol operations while their physiological responses such as heartrate, eye movement, and galvanic skin response were recorded using unobtrusive devices. Findings suggested that the random forest algorithm could predict cognitive workload with relatively high accuracy (>70%) given that it was entirely reliant on physiological signals. The developed MLA can be used to develop adaptive in-vehicle technology based on real-time estimation of cognitive workload, which can reduce the risk of MVCs in police operations.

An Overview of Approaches and Methods for the Cognitive Workload Estimation in Human–Machine Interaction Scenarios through Wearables Sensors

Background: Human-Machine Interaction (HMI) has been an important field of research in recent years, since machines will continue to be embedded in many human actvities in several contexts, such as industry and healthcare. Monitoring in an ecological mannerthe cognitive workload (CW) of users, who interact with machines, is crucial to assess their level of engagement in activities and the required effort, with the goal of preventing stressful circumstances. This study provides a comprehensive analysis of the assessment of CW using wearable sensors in HMI. Methods: this narrative review explores several techniques and procedures for collecting physiological data through wearable sensors with the possibility to integrate these multiple physiological signals, providing a multimodal monitoring of the individuals’CW. Finally, it focuses on the impact of artificial intelligence methods in the physiological signals data analysis to provide models of the CW to be exploited in HMI. Results: the review provided a comprehensive evaluation of the wearables, physiological signals, and methods of data analysis for CW evaluation in HMI. Conclusion: the literature highlighted the feasibility of employing wearable sensors to collect physiological signals for an ecological CW monitoring in HMI scenarios. However, challenges remain in standardizing these measures across different populations and contexts.

Mental workload estimation with electroencephalogram signals by combining multi-space deep models

The human brain is perpetually active, operating during both work and rest. Excessive mental activity, termed overload, can detrimentally impact health. Advances in predicting mental health conditions aim to prevent severe consequences and improve overall well-being. Consequently, mental status estimation has gained considerable research attention due to its potential benefits. Among the various signals used to assess mental state, the electroencephalogram, with its wealth of brain-related information, is widely employed by researchers. In this paper, we categorize mental workload into three states (low, middle, and high) and estimate a continuum of mental workload levels. Our method leverages information from multiple spatial dimensions (time and frequency domain) to achieve optimal results. In our approach, focusing on the time domain, we utilize Temporal Convolutional Networks to capture temporal information. Meanwhile, in the frequency domain, we implement two distinct branches aimed at acquiring varied information regarding Power Spectral Density by utilizing different lengths of the Fast Fourier Transform. Each branch consists of two blocks, known as the Multi-Dimensional Residual Block, representing a novel architecture by combining residual blocks. The integration of these two domains yields significant results compared to individual estimates in each domain. Our approach achieved a 74.98% accuracy in the three-class classification, surpassing the provided data results at 69.00%. Specially, our method demonstrates efficacy in estimating continuous levels, evidenced by a corresponding Concordance Correlation Coefficient result of 0.629. The combination of time and frequency domain analysis in our approach highlights the exciting potential to improve healthcare applications in the future.