A Deep Learning-Based Platform for Workers' Stress Detection Using Minimally Intrusive Multisensory Devices.

The advent of Industry 4.0 necessitates substantial interaction between humans and machines, presenting new challenges when it comes to evaluating the stress levels of workers who operate in increasingly intricate work environments. Undoubtedly, work-related stress exerts a significant influence on individuals' overall stress levels, leading to enduring health issues and adverse impacts on their quality of life. Although psychological questionnaires have traditionally been employed to assess stress, they lack the capability to monitor stress levels in real-time or on an ongoing basis, thus making it arduous to identify the causes and demanding aspects of work. To surmount this limitation, an effective solution lies in the analysis of physiological signals that can be continuously measured through wearable or ambient sensors. Previous studies in this field have mainly focused on stress assessment through intrusive wearable systems susceptible to noise and artifacts that degrade performance. One of our recently published papers presented a wearable and ambient hardware-software platform that is minimally intrusive, able to detect human stress without hindering normal work activities, and slightly susceptible to artifacts due to movements. A limitation of this system is its not very high performance in terms of the accuracy of detecting multiple stress levels; therefore, in this work, the focus was on improving the software performance of the platform, using a deep learning approach. To this purpose, three neural networks were implemented, and the best performance was achieved by the 1D-convolutional neural network with an accuracy of 95.38% for the identification of two levels of stress, which is a significant improvement over those obtained previously.

Stress responses in high-fidelity simulation and standard simulation training among medical students.

BACKGROUND: Simulation has been recognized as a shift in healthcare education that can improve skills and patient safety and outcomes. High-fidelity simulation of critical medical situations can be a source of stress among participants that can interfere with students' abilities leading to unexpected emotional responses. The aim of this study is to determine if two simulation methods, high-fidelity (HF) and procedural simulation (PS), may be associated with stress responses at a self-perceived and biological level (salivary cortisol variations), and to compare stress levels of the two different simulation method. We also wanted to find independent variables associated with cortisol total hormonal output. METHODS: A quasi-experimental before-after study was used including the administration of questionnaires, and biomarkers evaluation by salivary cortisol samples before and after simulation. A total of 148 students were eligible and agreed to participate in the study. We used paired T-test for mean comparison regarding State-trait anxiety for both HF and PT simulations. For NASA-TLX we performed a T-test mean comparison between groups. We used paired T-test mean comparison for cortisol analysis. Multivariable linear regression has been used to assess variables associated with AUCg values and perceived stress. RESULTS: values of STAI-Y scores were relatively higher at the end of the HF and PS sessions. NASA-TLX was significantly higher at baseline for the HF simulations, with respect to the PS simulation. Cortisol fold increase was significantly different in the two groups. Linear regression showed that cortisol AUCg was related to the STAI-Y score in both groups. CONCLUSION: Participating students developed a stress response both after in the HF and PS training, testified by psychological and biological outputs. According to our results, stress levels were increased for simply being in a simulation scenario than to the intrinsic complexity of the task required. More studies are needed to confirm this trend and to clarify the role of simulated stress response in a long-term learning scenario.

Development of the optimal touchscreen interface for patients with scleroderma.

Impaired hand function is a major contributor to overall disability and reduced health-related quality of life in scleroderma patients. A relevant issue concerns interaction of scleroderma subjects with touchscreen interfaces. This study aims at investigating this problem assessing scleroderma patients' performance with a novel, aptly designed, touchscreen application in order to identify critical items of touchscreen technology which may impair or facilitate the use by scleroderma subjects. Eighty scleroderma patients performed this novel application including three games, each of which tested a different gesture: tapping, dragging/dropping, and pinching-to-zoom. Eighty healthy subjects without hand impairments were recruited as controls. Scleroderma patients performed worse than healthy users in each game, and statistically significant negatively impacting items were identified. In the second phase of the study, the 10 worst touchscreen performers within the scleroderma cohort were recruited for a physio-rehabilitation trial based on the daily use at home of a modified version of the software application downloaded into the personal devices of patients. The results of this study allow introduction of guidelines to design accessible touchscreen interfaces for subjects with scleroderma and suggest that touchscreen technology may be included in self-administered physio-rehabilitation programs for scleroderma hand.

Respiratory muscle training in patients recovering recent open cardiothoracic surgery: a randomized-controlled trial.

OBJECTIVES: To evaluate the clinical efficacy and feasibility of an expiratory muscle training (EMT) device (Respilift) applied to patients recovering from recent open cardiothoracic surgery (CTS). DESIGN: Prospective, double-blind, 14-day randomised-controlled trial. PARTICIPANTS AND SETTING: A total of 60 inpatients recovering from recent CTS and early admitted to a pulmonary rehabilitation program. Interventions. Chest physiotherapy plus EMT with a resistive load of 30 cm H2O for active group and chest physiotherapy plus EMT with a sham load for control group. MEASURES: Changes in maximal expiratory pressure (MEP) were considered as primary outcome, while maximal inspiratory pressures (MIP), dynamic and static lung volumes, oxygenation, perceived symptoms of dyspnoea, thoracic pain, and well being (evaluated by visual analogic scale-VAS) and general health status were considered secondary outcomes. RESULTS: All outcomes recorded showed significant improvements in both groups; however, the change of MEP (+34.2 mmHg, P < 0.001 and +26.1%, P < 0.001 for absolute and % of predicted, resp.) was significantly higher in active group. Also VAS dyspnoea improved faster and more significantly (P < 0.05) at day 12, and 14 in active group when compared with control. The drop-out rate was 6%, without differences between groups. Conclusions. In patients recovering from recent CTS, specific EMT by Respilift is feasible and effective. This trial is registered with ClinicalTrials.gov NCT01510275.