Intelligence of Things: a spatial context-aware control system for smart devices

This study establishes how demanding healthcare work is experienced to be and whether nurses and physicians experience different levels of workload. A meta-analytic review was conducted of 87 studies that reported Task Load Index (TLX) scores for healthcare work. Of these studies, 37 were conducted in real-life settings and 50 in lab settings without real patients. In real-life settings, clinicians experienced a workload with a mean TLX of 49 (on a 0-100 scale). Divided onto staff groups, the mean TLX for nurses was 63, which was significantly higher than the mean of 40 for physicians. Among the six TLX subscales, the main contributors to workload were mental demand, temporal demand, and effort. They were higher than physical demand and frustration. The clinicians experienced their performance - the last subscale - as closer to poor than good in 38% of the studies conducted in real-life settings. The difference between nurses and physicians was consistent across all subscales, except mental demand. Finally, it is methodologically important that TLX scores appeared not to transfer directly from lab to real-life settings. To reduce the risk of errors and burnout, new healthcare procedures and technologies should be evaluated for their impact on workload.

While running a selection procedure, 27 male Belgian Special Forces candidates, with a mean age of 27.4 years (SD = 5.1), were randomly assigned to a no-stress control (n = 14) or a high-intensity stress group (n = 13). Participants in the latter group were exposed to an extremely strenuous mock prisoner of war (POW) exercise. Immediately after stress or control treatment, working memory and visuo-spatial declarative memory performances were measured by the digit span (DS) test and the Rey–Osterrieth complex figure (ROCF), respectively. Concurrently, stress levels were assessed by obtaining salivary cortisol measurements and subjectively by the NASA Task Load Index (TLX). As expected, exposure to high-intensity stress led to both robust cortisol increases and significant differences in TLX scores. Stress induction also significantly impaired DS and ROCF performances. Moreover, delta cortisol increases and ROCF performance in the POW stress group showed a significant negative correlation, while DS performances followed the same tendency. Summarizing, the current findings complement and extend previous work on hormonal stress effects, and the subsequent performance deterioration on two memory tests in a unique high-intensity stress environment.

Maintaining an appropriate level of cognitive load during simulations is crucial to optimize learning. We evaluated 3 subjective measures of cognitive load in a simulated operating room (OR) context across multiple scenarios of varying complexity. This observational study of 25 first-year anesthesiology residents took place during a 5-day simulation "Boot Camp." Each day, participants completed 2 different high-fidelity scenarios in a fully equipped simulated OR. After each simulation, participants completed 3 cognitive load measures: the Paas scale, NASA Task Load Index (TLX), and Cognitive Load Component (CLC) questionnaire. Two-way repeated-measures and mixed-design analyses of variance, with the cognitive load measures and scenarios as independent factors, were used to determine the effect of using different measures to report cognitive load. Cognitive load scores reported by all measures correlated significantly with one another (P < 0.01): TLX and Paas (r = 0.65); Paas and CLC (r = 0.63); and TLX and CLC (r = 0.61). The CLC subscale scores (intrinsic, extraneous, germane) also correlated significantly with composite TLX and Paas scores (P < 0.01). Scenarios and measures displayed significant interaction: F(10, 210) = 3.01, P = 0.001. Participants reported highest overall cognitive load using the Paas scale. All cognitive load measures were sensitive to scenario variability and showed similar fluctuation patterns across the 10 scenarios. The findings suggest that cognitive load measures can help create benchmarks based on learner perceptions of cognitive burden for different simulation scenarios.

Introduction: the Generative Pre-trained Transformer (GPT) is a deep learning language model architecture developed by OpenAI. Aim: to describe the knowledge networks (both at the theoretical and country levels) of the Generative Pre-trained Transformer (GPT) as an emerging technology. Results: 222 Documents were identified, of which 69 were articles, 50 were conference papers, 36 were editorials, 29 were notes, 19 were letters, 14 were reviews, 3 were conference reviews, and 2 were short surveys. In terms of the number of documents per year, 2 were found in 2019, 10 in 2020, 22 in 2021, 44 in 2022, and 144 in 2023. The year-on-year growth rate was over 100 % in all years. The subject area with the highest number of documents was Computer Science with 90 documents. The most productive countries in relation to GPT were the United States with 60 documents, followed by China with 19, the United Kingdom with 18, India with 15, and Australia with 12. Co-occurrence illustrated the centrality of Artificial Intelligence, Natural Language Processing, Deep Learning, and the term Human around ChatGPT and GPT. Conclusions: this bibliometric study aimed to describe the knowledge networks of the Generative Pre-trained Transformer (GPT) as an emerging technology. Although only 222 documents were found, this study revealed a high level of international scientific collaboration in the field. The results suggest that GPT is a highly relevant technology with a wide range of potential applications in natural language processing, artificial intelligence, and deep learning. Moreover, the study was able to qualitatively characterize the main thematic areas surrounding GPT, including its applications in chatbots, text generation, machine translation, sentiment analysis, and more.

ObjectivesTo explore the impact of self-reported motor skills on sex-specific training outcome in early interventional radiology (INR).Materials and methodsBased on the population of Reder et al., the study assessed the frequency, duration, and intensity of manual-focused activities, encompassing both non-professional and professional engagements (64 participants; 25 women), using the NASA Task Load Index (NASA-TLX) scores for mental workload following a standardized INR training session for basic techniques. Spearman’s correlation and multiple regression analyses with backward elimination and stepwise variable selection were conducted.ResultsSex-specific significant differences exist for predicting objective performance (OP) based on lifetime-developed hand focus. Leisure activities revealed a positive correlation with OP in men (β = 0.526; p = 0.001), unlike women. Conversely, profession-based activities correlated negatively with OP in men (β = −0.579; p = 0.022). In women, non-profession-based and profession-based activities did not significantly correlate with achieved OP. However, self-assessed performance (NASA-TLX) correlated with OP in women (β > 0.5 and p < 0.01 each). In men, cumulative hand focus (or fine motor skills) is crucial for success, influenced negatively by excessive profession-related work load (p = 0.025). In women, only self-assessment was responsible for objectively determined success or failure.ConclusionIn INR training, women might benefit more from external motivation, mental training, and empowerment, while men might profit more from physical training resembling observational learning. Considering the observed negative impact of excessive profession-related work load on men, a balanced lifestyle might lead to better outcomes related to fine motor skill demanding tasks.

BackgroundIn this study, we aimed to evaluate the impact of the COVID-19 epidemic on the workload and mental health of Iranian medical staff using the General Health Questionnaire (GHQ-12) and NASA -Task Load Index (NASA-TLX) Questionnaire between March and April 2020, respectively.MethodsThe present cross-sectional study was conducted from March 5th to April 5th, 2020. To evaluate the workload and mental health of participants NASA-TLX and GHQ-12 online questionnaires were distributed. Data were entered into software SPSS (Version 23) and T-test, ANOVA, Regression methods were used for data analysis.ResultsHealth workers who encountered COVID- 19 patients, were subjected to more task load compared to those who had no contact with COVID- 19 patients at the workplace (p < 0.001). In terms of the subscale score of NASA-TLX, nurses had more scores in mental pressure, physical pressure, time pressure (temporal), and frustration compared to the other jobs (p < 0.05). Moreover, nurses had significantly more workload compared to the other jobs.ConclusionsType of job, the shift of work, educational level, and facing COVID-19 affected the score of NASA-TLX. NASA-TLX scores were higher in nursing compared to the scores of other health staff groups. The results of this study indicate that the scores of NASA-TLX and GHQ-12 among staff who had contact with COVID-19 patients were significantly higher than those who did not face COVID-19 patients. We suggested that a comprehensive assistance should be provided to support the well-being of healthcare workers especially nurses and healthcare workers who treated COVID-19 patients.

Recent breakthroughs in artificial intelligence (AI) research include advancements in natural language processing (NLP) achieved by large language models (LLMs), and; in particular, generative pre–trained transformer (GPT) architectures. The latest GPT developed by OpenAI, GPT–4, has shown remarkable intelligence across various domains and tasks. It exhibits capabilities in abstraction, comprehension, vision, computer coding, mathematics, and more, suggesting it to be a significant step towards artificial general intelligence (AGI), a level of AI that possesses capabilities similar to human intelligence. This paper explores this AGI, its knowledge diffusive and societal influences, and its governance. In addition to coverage of the major associated topics studied in the literature, and making up for their loopholes, we scrutinize how GPT-4 can facilitate the diffusion of knowledge across different areas of science by promoting their interpretability and explainability (IE) to inexperts. Where applicable, the topics are also accompanied by their specific potential implications on medical imaging.

PT. SPU Palembang is an industrial company that produces Hospital Furniture and Rehabilitation products with the trademark SHIMA. This study aims to reduce the physical and mental workload received by workers in the Machine Shop Department. This research is focused on the production process .Measurement of physical workload uses the calculation of the worker's heart rate by calculating the percentage of Cardiovascular Load (CVL). Measurement of mental workload using the National Aeronautics and Space Administration Task Load Index (NASA-TLX) score calculation method. The results obtained based on the aspect of NASA-TLX before the proposal almost all workers receive a mental workload with a value of &gt; 80 that is included in the category of heavy mental workloads except Milling-Drilling Workers (4) have a score &lt; 80 in the category of moderate mental workload. After the proposal all workers have a score &lt; 80 so that the mental workload received falls into the medium category. Based on the proposals applied to the Lathe Worker (1) and the Milling-Drilling Worker (4) namely the improvement of work methods in the process of turning, punching, and the proposed holding of a reminder as a tool to remember, the results obtained %CVL from both workers &lt; 30%. Three other workers, although experiencing a decline, remained at &gt; 30%. The category of mental workload received by workers from previously included in the category of heavy mental workload dropped to moderate

Looping blood vessels in robot-assisted surgery remains challenging because of limitations in instrument articulation, joint length, and the lack of tactile feedback. To address this critical issue, we developed the Vas Guide, a novel device, for safe blood vessel handling. This study aims to establish the efficacy, safety, and durability of the Vas Guide. The Vas Guide is a flat, curved stainless-steel strip with blunt, rounded ends and a hole near the distal end. After sufficient dissection of the posterior surface of the target blood vessel, the Vas Guide with a vessel loop tied to the distal end was passed around the blood vessel in a single step using a suturing-like motion, thereby enabling secure vessel management. It is reusable and has a unit price of approximately $600. We analyzed 83 patients undergoing robot-assisted partial nephrectomy or nephroureterectomy at Tokushima University Hospital: 39 in the Vas Guide group and 44 in the conventional group. Outcomes including looping time, looping-related complication rates, such as collisions between instruments and bleeding, and additional instrument usage were recorded. The National Aeronautics and Space Administration task load index (NASA-TLX) scores were assessed. The patients' characteristics were comparable between the two groups. The Vas Guide group demonstrated significantly better outcomes, including faster looping times (31 vs. 90s, p < 0.001), fewer looping-related complications (0% vs. 30%, p < 0.001), no additional instrument usage (0% vs. 50%, p < 0.001), and better NASA-TLX scores (26.5 vs. 75.7, p < 0.001). The Vas Guide showed no usability issues, with only minor surface scratches and a slight radius change (8.60mm to 8.69mm). This reusable device combines high efficacy, safety, and durability and can be implemented at a relatively low cost.

Ensuring human safety has been one of the most critical considerations within the field of human–robot interaction. To explore the effects of working with autonomous robots on human coworkers’ perceived workload and job performance, two experiments were conducted in this study. Eight participants were recruited in the first experiment. Results revealed an increase in both subjective and objective workload measurements: compared to the baseline “no robot” to “empty payload” condition, the sum of NASA Task Load Index (NASA-TLX) scores increased from 129.3 (58.8) to 147.6 (53.7) ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p$</tex-math></inline-formula> -value = 0.041) and the pupil diameter increased from 4.07 (0.64) mm to 4.11 (0.67) mm; while working with a full payload robot, the sum of NASA-TLX scores increased to 151.7 (55.9) ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$p$</tex-math></inline-formula> -value = 0.010) and the pupil diameter increased to 4.12 (0.66) mm. Increased task completion time (3.1% for empty payload condition and 6.6% for full payload condition) showed a decrease in human productivity. However, this slight human output reduction was compensated by the substantial gain from the robot. Similar effects of autonomous mobile robots on participants’ NASA-TLX scores and task completion time were observed in the second experiment, where eight participants performed the order picking and sorting tasks in a high-fidelity grocery store setting. Results from the study suggested the feasibility of applying fully autonomous mobile robots in Wholesale and Retail Trade settings to improve human–robot team productivity while prioritizing physical safety and reasonable increases in mental workload.

Natural Language Processing (NLP) is a pivotal area in artificial intelligence, aiming to make computers capable of understanding and generating human language. This study evaluates and compares three prominent NLP models—the Generative Pre-trained Transformer (GPT) model, Bidirectional Encoder Representations from Transformers (BERT) model, and Generalized Autoregressive Pretraining for Language Understanding (XLNet)—to determine their strengths, limitations, and suitability for various tasks. The research involves a comprehensive analysis of these models, utilizing well-established datasets such as the Stanford Question Answering Dataset (SQuAD), General Language Understanding Evaluation (GLUE), Reading Comprehension from Examinations (RACE), and the Situations with Adversarial Generations (SWAG). The study explores each model's architecture, pre-training, and fine-tuning processes: GPT’s unidirectional approach is assessed for its language generation and handling of long-range dependencies; Bidirectional encoding is examined for its effectiveness in context understanding, and XLNet permutation-based training is analyzed for its robust contextual comprehension. The experimental results reveal that GPT excels in generative tasks but is constrained by its unidirectional nature. BERT achieves superior accuracy in comprehension tasks but is computationally demanding and susceptible to pre-training bias. XLNet outperforms both GPT and BERT in accuracy and contextual understanding, though at the cost of increased complexity. The results offer a significant understanding of the effectiveness and applicability of these models, suggesting future research directions such as hybrid models and improvements in efficiency.

In order to evaluate the mental workload of the operators more systematically, a subjective workload rating scale named ICO (short for Input, Cognition and Output) was developed. The structure of the ICO was based on the multiple resource theory. And the Analytic Hierarchy Process was applied in the development of the ICO. 12 subjects participated in a series of experiments, and rated the ICO and the NASA Task Load Index (TLX) on the completion of each experiment. By analyzing the ICO scores and TLX scores statistically, the sensitivity, diagnostic ability, and validity of the ICO rating scale were evaluated. Results show that the ICO demonstrates satisfying sensitivity, diagnostic ability and validity, and can be applied in the design and optimization of the human-in-the-loop system.

The urgent need to scale up testing capacity during the COVID-19 pandemic has prompted the rapid development of point-of-care diagnostic tools such as lateral flow immunoassays (LFIA) for large-scale community-based rapid testing. However, studies of how the general public perform when using LFIA tests in different environmental settings are scarce. This user experience (UX) study of 264 participants in Northern Ireland aimed to gather a better understanding of how self-administered LFIA tests were performed by the general public at home. The UX performance was assessed via analysis of a post-test questionnaire including 30 polar questions and 11 7-point Likert scale questions, which covers the multidimensional aspects of UX in terms of ease of use, effectiveness, efficiency, accuracy and satisfaction. Results show that 96.6% of participants completed the test with an overall average UX score of 95.27% [95% confidence interval (CI) 92.71–97.83%], which suggests a good degree of user experience and effectiveness. Efficiency was assessed based on the use of physical resources and human support received, together with the mental effort of self-administering the test measured via NASA Task Load Index (TLX). The results for six TLX subscales show that the participants scored the test highest for mental demand and lowest for physical demand, but the average TLX score suggests that the general public have a relatively low level of mental workload when using LFIA self-testing at home. Five printed LFIA testing results (i.e. the ‘simulated’ results) were used as the ground truth to assess the participant’s performance in interpreting the test results. The overall agreement (accuracy) was 80.63% [95% CI 75.21–86.05%] with a Kappa score 0.67 [95% CI 0.58–0.75] indicating substantial agreement. The users scored lower in confidence when interpreting test results that were weak positive cases (due to the relatively low signal intensity in the test-line) compared to strong positive cases. The end-users also found that the kit was easier to use than they expected (p < 0.001) and 231 of 264 (87.5%) reported that the test kit would meet their requirements if they needed an antibody testing kit. The overall findings provide an insight into the opportunities for improving the design of self-administered SARS-CoV-2 antibody testing kits for the general public and to inform protocols for future UX studies of LFIA rapid test kits.

Large language models (LLMs) offer substantial promise for improving health care; however, some risks warrant evaluation and discussion. This study assessed the effectiveness of safeguards in foundational LLMs against malicious instruction into health disinformation chatbots. Five foundational LLMs-OpenAI's GPT-4o, Google's Gemini 1.5 Pro, Anthropic's Claude 3.5 Sonnet, Meta's Llama 3.2-90B Vision, and xAI's Grok Beta-were evaluated via their application programming interfaces (APIs). Each API received system-level instructions to produce incorrect responses to health queries, delivered in a formal, authoritative, convincing, and scientific tone. Ten health questions were posed to each customized chatbot in duplicate. Exploratory analyses assessed the feasibility of creating a customized generative pretrained transformer (GPT) within the OpenAI GPT Store and searched to identify if any publicly accessible GPTs in the store seemed to respond with disinformation. Of the 100 health queries posed across the 5 customized LLM API chatbots, 88 (88%) responses were health disinformation. Four of the 5 chatbots (GPT-4o, Gemini 1.5 Pro, Llama 3.2-90B Vision, and Grok Beta) generated disinformation in 100% (20 of 20) of their responses, whereas Claude 3.5 Sonnet responded with disinformation in 40% (8 of 20). The disinformation included claimed vaccine-autism links, HIV being airborne, cancer-curing diets, sunscreen risks, genetically modified organism conspiracies, attention deficit-hyperactivity disorder and depression myths, garlic replacing antibiotics, and 5G causing infertility. Exploratory analyses further showed that the OpenAI GPT Store could currently be instructed to generate similar disinformation. Overall, LLM APIs and the OpenAI GPT Store were shown to be vulnerable to malicious system-level instructions to covertly create health disinformation chatbots. These findings highlight the urgent need for robust output screening safeguards to ensure public health safety in an era of rapidly evolving technologies.

The systematic review of clinical research papers is a labor-intensive and time-consuming process that often involves the screening of thousands of titles and abstracts. The accuracy and efficiency of this process are critical for the quality of the review and subsequent health care decisions. Traditional methods rely heavily on human reviewers, often requiring a significant investment of time and resources. This study aims to assess the performance of the OpenAI generative pretrained transformer (GPT) and GPT-4 application programming interfaces (APIs) in accurately and efficiently identifying relevant titles and abstracts from real-world clinical review data sets and comparing their performance against ground truth labeling by 2 independent human reviewers. We introduce a novel workflow using the Chat GPT and GPT-4 APIs for screening titles and abstracts in clinical reviews. A Python script was created to make calls to the API with the screening criteria in natural language and a corpus of title and abstract data sets filtered by a minimum of 2 human reviewers. We compared the performance of our model against human-reviewed papers across 6 review papers, screening over 24,000 titles and abstracts. Our results show an accuracy of 0.91, a macro F1-score of 0.60, a sensitivity of excluded papers of 0.91, and a sensitivity of included papers of 0.76. The interrater variability between 2 independent human screeners was κ=0.46, and the prevalence and bias-adjusted κ between our proposed methods and the consensus-based human decisions was κ=0.96. On a randomly selected subset of papers, the GPT models demonstrated the ability to provide reasoning for their decisions and corrected their initial decisions upon being asked to explain their reasoning for incorrect classifications. Large language models have the potential to streamline the clinical review process, save valuable time and effort for researchers, and contribute to the overall quality of clinical reviews. By prioritizing the workflow and acting as an aid rather than a replacement for researchers and reviewers, models such as GPT-4 can enhance efficiency and lead to more accurate and reliable conclusions in medical research.