A Comparison Of Wearable Physiological Sensors That Estimate Core Temperature In Simulated Hot Work Conditions

Methods: Eight men and eight women, regular cyclists (Age: 33.9±8 years; VO2max: 53.6±7.0 mL.kg−1.min−1) underwent two similar cycling trials in a controlled environment at 32°C, relative humidity 60%. The protocol consisted of an initial 10-minute rest, followed by an exercise protocol (60 minutes) in a hot environment, comprising 10 minutes at 20% of maximal aerobic power, 25 minutes at 55%, and 25 minutes at 75%, with an additional 25 minutes of post-exercise recovery. Core temperature was recorded simultaneously every minute using a gastrointestinal capsule (TCAPSULE) and the CORE Sensor (TCORE). Bland–Altman analysis was performed to calculate bias, upper (LCS) and lower (LCI) concordance limits, and the 95% confidence interval (CI95%). The maximum acceptable difference between the two devices was ±0.4°C. A mixed linear model was used to model the paired differences between the two measurement systems, considering the subjects, reliability and environmental conditions as random effects and the activities as a fixed effect. Results: The CORE Sensor recorded an ICC value of 0.98. A non-significant bias value of 0.01, LCS of 0.38°C, LCI of -0.35°C and 95% CI of ±0.36°C were found. The Exergen TAT-5000 recorded an ICC value of 0.90. A significant bias value of -0.59, LCS of 0.82°C, LCI of -2.05°C and 95% CI of ±1.44°C were found. Conclusion: Compared to TCAPSULE, the CORE Sensor was considered valid and reliable in estimating core temperature during cycling exercise in a hot environment. Compared to TCAPSULE, the Exergen TAT-5000 was considered reliable but invalid in estimating core temperature during cycling exercise in a hot environment.KEYWORDS Body Temperature; Cycling Exercise; Heat Wave; Thermometers; Heat Illnesses.

Wearable physiological sensors can provide a faithful record of a patient's physiological states without constant attention of caregivers. A computer program that can infer human activities from physiological recordings will be an valuable tool for physicians. In this paper we investigate to what extent current machine learning algorithms can correctly identify human activities from physiological sensors. We further identify two challenges that developers need to address. The first problem is that the labels of training data are inevitably noisy due to difficulties of annotating thousands hours of data. The second problem lies in the continuous nature of human activities, which violates the independence assumption made by many learning algorithms. We approach the first problem of noisy labeling in the multiple-label framework, and develop a conditional Markov models to take temporal context into consideration. We evaluate the proposed methods on 12,000 hours of the physiological recordings. The results show that support vector machines are effective to identify human activities from physiological signals, and efforts of disambiguating noisy labels are worthwhile

Stressful conversation is a frequently occurring stressor in our daily life. Stressors not only adversely affect our physical and mental health but also our relationships with family, friends, and coworkers. In this paper, we present a model to automatically detect stressful conversations using wearable physiological and inertial sensors. We conducted a lab and a field study with cohabiting couples to collect ecologically valid sensor data with temporally-precise labels of stressors. We introduce the concept of stress cycles, i.e., the physiological arousal and recovery, within a stress event. We identify several novel features from stress cycles and show that they exhibit distinguishing patterns during stressful conversations when compared to physiological response due to other stressors. We observe that hand gestures also show a distinct pattern when stress occurs due to stressful conversations. We train and test our model using field data collected from 38 participants. Our model can determine whether a detected stress event is due to a stressful conversation with an F1-score of 0.83, using features obtained from only one stress cycle, facilitating intervention delivery within 3.9 minutes since the start of a stressful conversation.

Validation of heart rate responses in wearable technology devices is generally composed of laboratory-based protocols that are steady state in nature and as a result, high accuracy measures are returned. However, there is a need to understand device validity in applied settings that include varied intensities of exercise. The purpose was to determine concurrent heart rate validity during trail running. Twenty-one healthy participants volunteered (female n = 10, [mean (SD)]: age = 31 [11] years, height = 173.0 [7] cm, mass = 75.6 [13] kg). Participants were outfitted with wearable technology devices (Garmin Fenix 5 wristwatch, Jabra Elite Sport earbuds, Motiv ring, Scosche Rhythm+ forearm band, Suunto Spartan Sport watch with accompanying chest strap) and completed a self-paced 3.22 km trail run while concurrently wearing a criterion heart rate strap (Polar H7 heart rate monitor). The trail runs were out-and-back with the first 1.61 km in an uphill direction, and the 1.61 return being downhill in nature. Validity was determined through three methods: Mean Absolute Percent Error (MAPE), Bland-Altman Limits of Agreement (LOA), and Lin’s Concordance Coefficient (rC). Validity measures overall are as follows: Garmin Fenix 5 (MAPE = 13%, LOA = -32 to 162, rC = 0.32), Jabra Elite Sport (MAPE = 23%, LOA = -464 to 503, rC = 0.38), Motiv ring (MAPE = 16%, LOA = -52 to 96, rC = 0.29), Scosche Rhythm+ (MAPE = 6%, LOA = -114 to 120, rC = 0.79), Suunto Spartan Sport (MAPE = 2%, LOA = -62 to 61, rC = 0.96). All photoplethysmography-based (PPG) devices displayed poor heart rate agreement during variable intensity trail running. Until technological advances occur in PPG-based devices allowing for acceptable agreement, heart rate in outdoor environments should be obtained using an ECG-based chest strap that can be connected to a wristwatch or other comparable receiver.

Regular physical activity is effective in reducing visceral white adipose tissue (AT) inflammation and oxidative stress, and these changes are commonly associated with reduced adiposity. However, the impact of multiple periods of physical activity, intercalated by periods of inactivity, i.e., intermittent physical activity, on markers of AT inflammation and oxidative stress is unknown. In the present study, 5-wk-old male C57BL/6 mice were randomized into three groups (n = 10/group): sedentary, regular physical activity, and intermittent physical activity, for 24 wk. All animals were singly housed and fed a diet containing 45% kcal from fat. Regularly active mice had access to voluntary running wheels throughout the study period, whereas intermittently active mice had access to running wheels for 3-wk intervals (i.e., 3 wk on/3 wk off) throughout the study. At death, regular and intermittent physical activity was associated with similar reductions in visceral AT mass (approximately -24%, P < 0.05) relative to sedentary. However, regularly, but not intermittently, active mice exhibited decreased expression of visceral AT genes related to inflammation (e.g., monocyte chemoattractant protein 1), immune cell infiltration (e.g., CD68, CD11c, F4/80, CD11b/CD18), oxidative stress (e.g., p47 phagocyte oxidase), and endoplasmic reticulum stress (e.g., CCAAT enhancer-binding protein homologous protein; all P < 0.05). Furthermore, regular, but not intermittent, physical activity was associated with a trend toward improvement in glucose tolerance (P = 0.059). Collectively, these findings suggest that intermittent physical activity over a prolonged period of time may lead to a reduction in adiposity but with retention of a sedentary obese white AT and metabolic phenotype.

IntroductionContinuous, ambulatory perioperative monitoring using wearable devices has shown promise for earlier detection of physiological deterioration and postoperative complications, preventing ‘failure-to-rescue’. This study aimed to compare the accuracy of vital signs measured by wrist-based wearables with gold standard measurements from vital signs monitors or nurse assessments in major abdominal surgery. MethodsAdult patients were eligible for inclusion in this prospective observational study validating the Empatica E4 wrist sensor intraoperatively and postoperatively. The primary outcomes were the 95% limits of agreement (LoA) between manual and device recordings of heart rate (HR) and temperature evaluated via Bland-Altman analysis. Secondary analysis was conducted using Clarke-Error grid analysis. ResultsOverall, 31 patients were recruited, and 27 patients completed the study. The median duration of recording per patient was 70.3 h, and a total of 2112 h of data recording were completed. Wrist-based HR measurement was accurate and moderately precise (bias: 0.3 bpm; 95% LoA −15.5 to 17.1), but temperature measurement was neither accurate nor precise (bias −2.2°C; 95% LoA −6.0 to 1.6). On Clarke-Error grid analysis, 74.5% and 29.6% of HR and temperature measurements, respectively, fell within the acceptable range of reference standards. ConclusionsContinuous perioperative monitoring of HR and temperature after major abdominal surgery using wrist-based sensors is feasible but was limited in this study by low precision. While wrist-based devices offer promise for the continuous monitoring of high-risk surgical patients, current technology is inadequate. Ongoing device hardware and software innovation with robust validation is required before such technologies can be routinely adopted in clinical practice.

Concerned with understanding emotion and visitor responses through both self-report and wearable physiological sensors, this paper examines the affective relationships enacted at a historic house within two distinct visitor routes. First, it looks at the growing interest in emotion and covers a compatible approach to physiology, emotion and affect drawing on the work of psychologists, neuroscientists and social psychologists. Subsequently, the paper offers an analysis and exploratory discussion of physiological and self-report survey data in determining the emotional and affective component of heritage visitation. This is presented within the context of research conducted in collaboration with the National Trust, UK. Through a discussion of the limitations of such a method, it offers a nuanced understanding of both emotional experience and the potential of technological tools.

Exploring the effect of urban features and immediate environment on body responses

The recent trend and parallel development/adoption of Virtual Reality, Brain Sensing Measures and associated technology such as Augmented Reality by large corporations, and the rise in the interests in the consumer market have set a positive tone for research in these disciplines. An important human factors area that is a catalyst to broad VR applications is the measure of perception, mental workload, and immersion amongst other issues, which are determining factors in the experience of using virtual environments. Traditional approaches in studying these issues use well-developed subjective measures via questionnaires. A new opportunity in the parallel developments in wearable physiological sensors such as brain scanners could potentially be an objective approach in resolving many subjective uncertainties amongst other prospects. Here, we propose the integration of these two emerging fields in order to provide a continuous, objective, physiological measure of an individual's VR experience for the purposes of enhancing user experience and improving performance. This positional paper attempts to merge two complementary field of work, and discusses implications which could potentially open up avenues of research which were traditionally difficult due to the limitations of equipment, or the lack of quantified approach.

Given the close coupling between wearable devices and the human body, a natural application for these devices is to inform the user of their physiological status. An important point is the potential for wearables to empower the user to monitor his/her own health rather than rely solely on medical professionals or sophisticated medical equipment. This makes it more convenient for the user to monitor certain vital signs more often and this continuous monitoring can prove crucial in diagnosing certain conditions. Pervasive monitoring allows for collection of a large amount of data, which in turn can allow treatments in medicine that are anticipatory rather than reactionary. We emphasize the importance of the design of convenient, wearable physiological sensors by looking in-depth at one specific wearable device called BioWatch, which can be used to non-invasively and continuously measure blood pressure from the wrist.

Seating location in the classroom can affect student engagement, attention and academic performance by providing better visibility, improved movement, and participation in discussions. Existing studies typically explore how traditional seating arrangements (e.g. grouped tables or traditional rows) influence students' perceived engagement, without considering group seating behaviours under more flexible seating arrangements. Furthermore, survey-based measures of student engagement are prone to subjectivity and various response bias. Therefore, in this research, we investigate how individual and group-wise classroom seating experiences affect student engagement using wearable physiological sensors. We conducted a field study at a high school and collected survey and wearable data from 23 students in 10 courses over four weeks. We aim to answer the following research questions: 1. How does the seating proximity between students relate to their perceived learning engagement? 2. How do students' group seating behaviours relate to their physiologically-based measures of engagement (i.e. physiological arousal and physiological synchrony)? Experiment results indicate that the individual and group-wise classroom seating experience is associated with perceived student engagement and physiologically-based engagement measured from electrodermal activity. We also find that students who sit close together are more likely to have similar learning engagement and tend to have high physiological synchrony. This research opens up opportunities to explore the implications of flexible seating arrangements and has great potential to maximize student engagement by suggesting intelligent seating choices in the future.

Functional seizure therapy via transauricular vagus nerve stimulation

Introduction/Objective Programmed death ligand 1 (PD-L1) is an immune checkpoint protein expressed on activated immune cells and tumor cells. This ligand is the FDA approved companion diagnostic marker to Pembrolizumab in non-small cell lung carcinomas (NSCLC). Many institutions score PD-L1 at the time of initial diagnosis on small biopsies or fine-needle aspiration (FNA) cytology specimens to provide tumor characteristics for clinical management decisions. PD-L1 also labels macrophages, which can cause difficulties in accurately assessing the PD-L1 score on cytology specimens. Dual staining with TTF-1/PD-L1 or p40/PD-L1 could help to accurately identify PD-L1 positivity in tumor cells without including macrophages in the final score. Our goal was to study if dual staining would change PD-L1 results. Methods/Case Report A database search was performed for cytology cases (pleural fluids and lung FNAs) diagnosed as NSCLC between the years 2021 and 2024. 10 cases of lung adenocarcinoma (AC) and 10 cases of squamous cell carcinoma (SCC) were identified in the archive. Dual staining with TTF-1/PD-L1 for AC or with p40/PD-L1 for SCC was performed in cases having an adequate cell block. Consensus review and scoring of the original PD-L1 and dual stains was performed by 4 pathologists, blinded to the original PD-L1 score result. Statistical analysis was performed to compare the original PD-L1 score and consensus PD-L1 score using a paired two tailed T-test after evaluation for normality. Similarly, the consensus PD-L1 score and dual PD-L1 score were compared. A Bland-Altman plot was then generated for both comparisons to evaluate for agreement of the methods within the upper and lower 95% limits of agreement (LoA). Results Upon consensus review, the mean difference between original PD-L1 score and consensus PD-L1 score was 2.5% (range: -41.5% to 47%; standard deviation (SD): 20.2). The paired two tailed T-test showed no significant difference between groups (p = 0.67). A Bland-Altman plot for this pair was constructed with upper 95% LoA 42.0 (confidence interval (CI): 19.5 to 64.6) and lower 95% LoA -37.0 (CI: -59.6 to -14.5). The mean difference was 2.5 (CI: -10.3 to 15.3) The Bland-Altman plot showed the majority of score differences fell within the upper and lower 95% LoA’s. The mean difference between consensus PD-L1 score and dual PD-L1 score was -2.5% (range: -40% to 30%; SD: 19.0). The paired two tailed T-test showed no significant difference between groups (p = 0.66). A Bland-Altman plot for this pair was constructed with upper 95% LoA 34.8 (CI: 12.2 to 57.3) and lower 95% LoA -39.9 (CI: -62.4 to -17.3). The mean difference was -2.5 (CI: -15.3 to 10.2). The Bland-Altman plot showed that all score differences fell within the upper and lower 95% LoA’s. Conclusion Our study demonstrates no significant difference in performance between mono PD-L1 staining and dual TTF-1/PD-L1 or p40/PD-L1 staining. These results indicate that dual staining would likely not have significant impact on PD-L1 scoring and may not be worth the additional cost. However, the dual stain was easier to interpret compared to the mono PD-L1 stain as it distinguished PD-L1 positive tumor cells from macrophages, which can often be challenging. We also found no significant difference between original PD-L1 and consensus PD-L1. This finding suggests that PD-L1 scoring by consensus opinion is not superior to interpretation by a single pathologist at our institution.

[Figure: see text]

Left atrial strain (LAS) on transthoracic echocardiogram (TTE) is increasingly recognised to have clinical utility in cardiovascular disease. Differences in LAS measurements between vendors remains a barrier for clinical use. We sought to compare LAS between two commonly used software platforms; the layer-specific endocardial and mid-myocardial measurements of LAS on General Electric (GE) Echopac were compared to TomTec strain. LAS was measured in 88 individuals with no previous cardiac history and 40 paroxysmal AF (PAF) patients, in sinus rhythm at TTE. Conventionally, LAS measured using GE Echopac is mid-myocardial strain (GE-mid); additionally, endocardial (GE-endo) LAS was evaluated. Both LAS measurements by GE were compared to TomTec-Arena (v2.30.02) measurements. Reservoir (ƐR), contractile (ƐCT) and conduit (ƐCD) phasic strain were evaluated. Both GE-mid and GE-endo LAS correlated well with TomTec LAS. On Bland-Altman analysis, GE-mid LAS measurements were systematically lower than TomTec LAS (ƐR: mean difference (MD) - 6.08%, limits of agreement (LOA) - 12%, 0%, ƐCT: MD - 0.8%, LOA - 7%, 5%, ƐCD: MD - 5.2% LOA - 12%, 1%). GE-endo LAS demonstrated no systematic difference from TomTec LAS, but had wider limits of agreement (ƐR: MD 0.41%, LOA - 7%, 8%, ƐCT: MD 0.50%, LOA - 6%, 7%, ƐCD: MD - 0.08%, LOA - 7%, 7%). ƐR had the best reproducibility. Mid-myocardial LAS, routinely evaluated by GE Echopac software, systematically underestimates LAS compared to TomTec. Using GE endocardial LAS eliminated this bias, but introduced greater variation between measurements. Serial measurements of LAS should therefore be performed on the same vendor system.