Wireless Technologies in Flexible and Wearable Sensing: from Materials Design, System Integration to Applications

Wireless and wearable sensors attract considerable interest in personalized healthcare by providing a unique approach for remote, non-contact, and continuous monitoring of various health-related signals without interference with daily life. Recent advances in wireless technologies and wearable sensors have promoted practical applications due to their significantly improved characteristics, such as reduction in size and thickness, enhancement in flexibility and stretchability, and improved conformability to the human body. Currently, most researches focus on active materials and structural designs for wearable sensors, with just a few exceptions reflecting on the technologies for wireless data transmission. This review provides a comprehensive overview of the state-of-the-art wireless technologies and related studies on empowering wearable sensors. The emerging functional nanomaterials utilized for designing unique wireless modules are highlighted, which include metals, carbons, and MXenes. Additionally, the review outlines the system-level integration of wireless modules with flexible sensors, spanning from novel design strategies for enhanced conformability to efficient transmitting data wirelessly. Furthermore, the review introduces representative applications for remote and non-invasive monitoring of physiological signals through on-skin and implantable wireless flexible sensing systems. Finally, the challenges, perspectives, and unprecedented opportunities for wireless and wearable sensors are discussed. This article is protected by copyright. All rights reserved.

Objective assessment of postural ergonomics in neurosurgery: integrating wearable technology in the operating room

OBJECTIVE

Physical stress associated with the static posture of neurosurgeons over prolonged periods can result in fatigue and musculoskeletal disorders. Objective assessment of surgical ergonomics may contribute to postural awareness and prevent further complications. This pilot study examined the feasibility of using wearable technology as a biofeedback tool to address this gap.

METHODS

Ten neurosurgeons, including 5 attendings (all faculty) and 5 trainees (1 fellow, 4 residents), were recruited and equipped with two wearable sensors attached to the back of their head and their upper back. The sensors collected the average time spent in extended (≤ -10°), neutral (> -10° and < 10°), and flexed (≥ 10°) static postures (undetected activity for more than 10 seconds) during spine and cranial procedures. Feasibility outcomes aimed for more than 70% of accurate data collection. Exploratory outcomes included the comparison of postural variability within and between participants adjusted to their demographics excluding nonrelated surgical activities, and postoperative self-assessment surveys.

RESULTS

Sixteen (80%) of 20 possible recordings were successfully collected and analyzed from 11 procedures (8 spine, 3 cranial). Surgeons maintained a static posture during 52.7% of the active surgical time (mean 1.58 hrs). During spine procedures, all surgeons used an exoscope while standing, leading to a significantly longer time spent in a neutral static posture (p < 0.001, partial η2 = 0.14): attendings remained longer in a neutral static posture (36.4% ± 15.3%) than in the extended (9% ± 6.3%) and flexed (5.7% ± 3.4%) static postures; trainees also remained longer in a neutral static posture (30.2% ± 13.8%) than in the extended (11.1% ± 6.3%) and flexed (11.9% ± 6.6%) static postures. During cranial procedures, surgeons intermittently transitioned between standing/exoscope use and sitting/microscope use, with trainees spending a shorter time in a neutral static posture (16.3% vs 48.5%, p < 0.001) and a longer time in a flexed static posture (18.5% vs 2.7%, p < 0.001) compared with attendings. Additionally, longer cranial procedures correlated with surgeons spending a longer time (r = 0.94) in any static posture (extended, flexed, and neutral), with taller surgeons exhibiting longer periods in flexed and extended static postures (r = 0.86). Postoperative self-assessment revealed that attendings perceived spine procedures as more difficult than trainees (p = 0.029), while trainees found cranial procedures to be of greater difficulty than spine procedures (p = 0.012). Attendings felt more stressed (p = 0.048), less calmed (p = 0.024), less relaxed (p = 0.048), and experienced greater stiffness in their upper body (p = 0.048) and more shoulder pain (p = 0.024) during cranial versus spine procedures.

CONCLUSIONS

Wearable technology is feasible to assess postural ergonomics and provide objective biofeedback to neurosurgeons during spine and cranial procedures. This study showed reproducibility for future comparative protocols focused on correcting posture and surgical ergonomic education.

Aerosol-Synthesized Surfactant-Free Single-Walled Carbon Nanotube-Based NO2 Sensors: Unprecedentedly High Sensitivity and Fast Recovery

This study pioneers a chemical sensor based on surfactant-free aerosol-synthesized single-walled carbon nanotube (SWCNT) films for detecting nitrogen dioxide (NO2). Unlike conventional CNTs, the SWCNTs used in this study exhibit one of the highest surface-to-volume ratios. They show minimal bundling without the need for surfactants and have the lowest number of defects among reported CNTs. Furthermore, the dry-transferrable and facile one-step lamination results in promising industrial viability. When applied to devices, the sensor shows excellent sensitivity (41.6% at 500 ppb), rapid response/recovery time (14.2/120.8 s), a remarkably low limit of detection (below ≈0.161 ppb), minimal noise, repeatability for more than 50 cycles without fluctuation, and long-term stability for longer than 6 months. This is the best performance reported for a pure CNT-based sensor. In addition, the aerosol SWCNTs demonstrate consistent gas-sensing performance even after 5000 bending cycles, indicating their suitability for wearable applications. Based on experimental and theoretical analyses, the proposed aerosol CNTs are expected to overcome the limitations associated with conventional CNT-based sensors, thereby offering a promising avenue for various sensor applications.

Digital Measures of Postural Sway Quantify Balance Deficits in Spinocerebellar Ataxia

BACKGROUND

Maintaining balance is crucial for independence and quality of life. Loss of balance is a hallmark of spinocerebellar ataxia (SCA).

OBJECTIVE

The aim of this study was to identify which standing balance conditions and digital measures of body sway were most discriminative, reliable, and valid for quantifying balance in SCA.

METHODS

Fifty-three people with SCA (13 SCA1, 13 SCA2, 14 SCA3, and 13 SCA6) and Scale for Assessment and Rating of Ataxia (SARA) scores 9.28 ± 4.36 and 31 healthy controls were recruited. Subjects stood in six test conditions (natural stance, feet together and tandem, each with eyes open [EO] and eyes closed [EC]) with an inertial sensor on their lower back for 30 seconds (×2). We compared test completion rate, test-retest reliability, and areas under the receiver operating characteristic curve (AUC) for seven digital sway measures. Pearson's correlations related sway with the SARA and the Patient-Reported Outcome Measure of Ataxia (PROM ataxia).

RESULTS

Most individuals with SCA (85%-100%) could stand for 30 seconds with natural stance EO or EC, and with feet together EO. The most discriminative digital sway measures (path length, range, area, and root mean square) from the two most reliable and discriminative conditions (natural stance EC and feet together EO) showed intraclass correlation coefficients from 0.70 to 0.91 and AUCs from 0.83 to 0.93. Correlations of sway with SARA were significant (maximum r = 0.65 and 0.73). Correlations with PROM ataxia were mild to moderate (maximum r = 0.56 and 0.34).

CONCLUSION

Inertial sensor measures of extent of postural sway in conditions of natural stance EC and feet together stance EO were discriminative, reliable, and valid for monitoring SCA. © 2024 International Parkinson and Movement Disorder Society.

Soft Polymer Optical Fiber Sensors for Intelligent Recognition of Elastomer Deformations and Wearable Applications

In recent years, soft robotic sensors have rapidly advanced to endow robots with the ability to interact with the external environment. Here, we propose a polymer optical fiber (POF) sensor with sensitive and stable detection performance for strain, bending, twisting, and pressing. Thus, we can map the real-time output light intensity of POF sensors to the spatial morphology of the elastomer. By leveraging the intrinsic correlations of neighboring sensors and machine learning algorithms, we realize the spatially resolved detection of the pressing and multi-dimensional deformation of elastomers. Specifically, the developed intelligent sensing system can effectively recognize the two-dimensional indentation position with a prediction accuracy as large as ~99.17%. The average prediction accuracy of combined strain and twist is ~98.4% using the random forest algorithm. In addition, we demonstrate an integrated intelligent glove for the recognition of hand gestures with a high recognition accuracy of 99.38%. Our work holds promise for applications in soft robots for interactive tasks in complex environments, providing robots with multidimensional proprioceptive perception. And it also can be applied in smart wearable sensing, human prosthetics, and human-machine interaction interfaces.

Accelerometer-based measures in Friedreich ataxia: a longitudinal study on real-life activity

Quantitative measurement of physical activity may complement neurological evaluation and provide valuable information on patients' daily life. We evaluated longitudinal changes of physical activity in patients with Friedreich ataxia (FRDA) using remote monitoring with wearable sensors. We performed an observational study in 26 adult patients with FRDA and 13 age-sex matched healthy controls (CTR). Participants were asked to wear two wearable sensors, at non-dominant wrist and at waist, for 7 days during waking hours. Evaluations were performed at baseline and at 1-year follow-up. We analysed the percentage of time spent in sedentary or physical activities, the Vector Magnitude on the 3 axes (VM3), and average number of steps/min. Study participants were also evaluated with ataxia clinical scales and functional tests for upper limbs dexterity and walking capability. Baseline data showed that patients had an overall reduced level of physical activity as compared to CTR. Accelerometer-based measures were highly correlated with clinical scales and disease duration in FRDA. Significantly changes from baseline to l-year follow-up were observed in patients for the following measures: (i) VM3; (ii) percentage of sedentary and light activity, and (iii) percentage of Moderate-Vigorous Physical Activity (MVPA). Reduction in physical activity corresponded to worsening in gait score of the Scale for Assessment and Rating of Ataxia. Real-life activity monitoring is feasible and well tolerated by patients. Accelerometer-based measures can quantify disease progression in FRDA over 1 year, providing objective information about patient's motor activities and supporting the usefulness of these data as complementary outcome measure in interventional trials.

Standing Balance Conditions and Digital Sway Measures for Clinical Trials of Friedreich's Ataxia

BACKGROUND

Progressive loss of standing balance is a feature of Friedreich's ataxia (FRDA).

OBJECTIVES

This study aimed to identify standing balance conditions and digital postural sway measures that best discriminate between FRDA and healthy controls (HC). We assessed test-retest reliability and correlations between sway measures and clinical scores.

METHODS

Twenty-eight subjects with FRDA and 20 HC completed six standing conditions: feet apart, feet together, and feet tandem, both with eyes opened (EO) and eyes closed. Sway was measured using a wearable sensor on the lumbar spine for 30 seconds. Test completion rate, test-retest reliability with intraclass correlation coefficients, and areas under the receiver operating characteristic curves (AUCs) for each measure were compared to identify distinguishable FRDA sway characteristics from HC. Pearson correlations were used to evaluate the relationships between discriminative measures and clinical scores.

RESULTS

Three of the six standing conditions had completion rates over 70%. Of these three conditions, natural stance and feet together with EO showed the greatest completion rates. All six of the sway measures' mean values were significantly different between FRDA and HC. Four of these six measures discriminated between groups with >0.9 AUC in all three conditions. The Friedreich Ataxia Rating Scale Upright Stability and Total scores correlated with sway measures with P-values <0.05 and r-values (0.63-0.86) and (0.65-0.81), respectively.

CONCLUSION

Digital postural sway measures using wearable sensors are discriminative and reliable for assessing standing balance in individuals with FRDA. Natural stance and feet together stance with EO conditions suggest use in clinical trials for FRDA. © 2024 International Parkinson and Movement Disorder Society.

Wearable Sensor Based on a Tough Conductive Gel for Real-Time and Remote Human Motion Monitoring

The usage of a conductive hydrogel in wearable sensors has been thoroughly researched recently. Nonetheless, hydrogel-based sensors cannot simultaneously have excellent mechanical property, high sensitivity, comfortable wearability, and rapid self-healing performance, which result in poor durability and reusability. Herein, a robust conductive hydrogel derived from one-pot polymerization and subsequent solvent replacement is developed as a wearable sensor. Owing to the reversible hydrogen bonds cross-linked between polymer chains and clay nanosheets, the resulting conductive hydrogel-based sensor exhibits outstanding flexibility, self-repairing, and fatigue resistance performances. The embedding of graphene oxide nanosheets offers an enhanced hydrogel network and easy release of wearable sensor from the target position through remote irradiation, while Li+ ions incorporated by solvent replacement endow the wearable sensor with low detection limit (sensing strain: 1%), high conductivity (4.3 S m-1) and sensitivity (gauge factor: 3.04), good freezing resistance, and water retention. Therefore, the fabricated wearable sensor is suitable to monitor small and large human motions on the site and remotely under subzero (-54 °C) or room temperature, indicating lots of promising applications in human-motion monitoring, information encryption and identification, and electronic skins.

IMUs Can Estimate Hip and Knee Range of Motion during Walking Tasks but Are Not Sensitive to Changes in Load or Grade

The ability to estimate lower-extremity mechanics in real-world scenarios may untether biomechanics research from a laboratory environment. This is particularly important for military populations where outdoor ruck marches over variable terrain and the addition of external load are cited as leading causes of musculoskeletal injury As such, this study aimed to examine (1) the validity of a minimal IMU sensor system for quantifying lower-extremity kinematics during treadmill walking and running compared with optical motion capture (OMC) and (2) the sensitivity of this IMU system to kinematic changes induced by load, grade, or a combination of the two. The IMU system was able to estimate hip and knee range of motion (ROM) with moderate accuracy during walking but not running. However, SPM analyses revealed IMU and OMC kinematic waveforms were significantly different at most gait phases. The IMU system was capable of detecting kinematic differences in knee kinematic waveforms that occur with added load but was not sensitive to changes in grade that influence lower-extremity kinematics when measured with OMC. While IMUs may be able to identify hip and knee ROM during gait, they are not suitable for replicating lab-level kinematic waveforms.

Wearable sensors to measure the influence of sonic seasoning on wine consumers in a live context: a preliminary proof-of-concept study

BACKGROUND

Any action capable of creating expectations about product quality would be able to modulate experienced pleasantness. In this context, during the 2022 edition of the Internet Festival (Pisa, Italy) a 'social experiment' was promoted to set up an affordable and reliable methodology based on wearable sensors to measure the emotions aroused in a live context on consumers by different kinds of wines. Therefore, five wines (two faulty ones and three high-quality samples) were proposed to 50 non-selected consumers in an arousing context with live jazz music as background. Both explicit (questionnaires) and two different approaches for implicit methods (electrocardiogram (ECG) recorded by wearable sensors vs. smartphones), the latter performed on a subgroup of 16, to measure the emotions aroused by wines and music were utilized synergistically.

RESULTS

According to our findings: (i) wine undoubtedly generates a significant emotional response on consumers; (ii) this answer is multifaceted and attributable to the quality level of the wine tasted. In fact, all things being equal, while drinking wine even untrained consumers can perfectly recognize good wines compared to products of lower quality; (iii) high-quality wines are able to induce a spectrum of positive emotions, as observed by the analysis of ECG signals, especially when they are coupled with background music.

CONCLUSION

The framework has certainly played to the advantage of good-quality wines, fostering their positive emotional characteristics on the palate even of some less experienced consumers, thanks to a dragging effect towards a positive mood generated by the surrounding conditions (good music in a beautiful location). © 2024 Society of Chemical Industry.

Long-term continuous instrumented insole-based gait analyses in daily life have advantages over longitudinal gait analyses in the lab to monitor healing of tibial fractures

Introduction: Monitoring changes in gait during rehabilitation allows early detection of complications. Laboratory-based gait analyses proved valuable for longitudinal monitoring of lower leg fracture healing. However, continuous gait data recorded in the daily life may be superior due to a higher temporal resolution and differences in behavior. In this study, ground reaction force-based gait data of instrumented insoles from longitudinal intermittent laboratory assessments were compared to monitoring in daily life. Methods: Straight walking data of patients were collected during clinical visits and in between those visits the instrumented insoles recorded all stepping activities of the patients during daily life. Results: Out of 16 patients, due to technical and compliance issues, only six delivered sufficient datasets of about 12 weeks. Stance duration was longer (p = 0.004) and gait was more asymmetric during daily life (asymmetry of maximal force p < 0.001, loading slope p = 0.001, unloading slope p < 0.001, stance duration p < 0.001). Discussion: The differences between the laboratory assessments and the daily-life monitoring could be caused by a different and more diverse behavior during daily life. The daily life gait parameters significantly improved over time with union. One of the patients developed an infected non-union and showed worsening of force-related gait parameters, which was earlier detectable in the continuous daily life gait data compared to the lab data. Therefore, continuous gait monitoring in the daily life has potential to detect healing problems early on. Continuous monitoring with instrumented insoles has advantages once technical and compliance problems are solved.

Data Augmentation Techniques for Accurate Action Classification in Stroke Patients with Hemiparesis

Stroke survivors with hemiparesis require extensive home-based rehabilitation. Deep learning-based classifiers can detect actions and provide feedback based on patient data; however, this is difficult owing to data sparsity and heterogeneity. In this study, we investigate data augmentation and model training strategies to address this problem. Three transformations are tested with varying data volumes to analyze the changes in the classification performance of individual data. Moreover, the impact of transfer learning relative to a pre-trained one-dimensional convolutional neural network (Conv1D) and training with an advanced InceptionTime model are estimated with data augmentation. In Conv1D, the joint training data of non-disabled (ND) participants and double rotationally augmented data of stroke patients is observed to outperform the baseline in terms of F1-score (60.9% vs. 47.3%). Transfer learning pre-trained with ND data exhibits 60.3% accuracy, whereas joint training with InceptionTime exhibits 67.2% accuracy under the same conditions. Our results indicate that rotational augmentation is more effective for individual data with initially lower performance and subset data with smaller numbers of participants than other techniques, suggesting that joint training on rotationally augmented ND and stroke data enhances classification performance, particularly in cases with sparse data and lower initial performance.

Ultrastretchable E-Skin Based on Conductive Hydrogel Microfibers for Wearable Sensors

Conductive microfibers play a significant role in the flexibility, stretchability, and conductivity of electronic skin (e-skin). Currently, the fabrication of conductive microfibers suffers from either time-consuming and complex operations or is limited in complex fabrication environments. Thus, it presents a one-step method to prepare conductive hydrogel microfibers based on microfluidics for the construction of ultrastretchable e-skin. The microfibers are achieved with conductive MXene cores and hydrogel shells, which are solidified with the covalent cross-linking between sodium alginate and calcium chloride, and mechanically enhanced by the complexation reaction of poly(vinyl alcohol) and sodium hydroxide. The microfiber conductivities are tailorable by adjusting the flow rate and concentration of core and shell fluids, which is essential to more practical applications in complex scenarios. More importantly, patterned e-skin based on conductive hydrogel microfibers can be constructed by combining microfluidics with 3D printing technology. Because of the great advantages in mechanical and electrical performance of the microfibers, the achieved e-skin shows impressive stretching and sensitivity, which also demonstrate attractive application values in motion monitoring and gesture recognition. These characteristics indicate that the ultrastretchable e-skin based on conductive hydrogel microfibers has great potential for applications in health monitoring, wearable devices, and smart medicine.

Objective Determination of Eating Occasion Timing: Combining Self-Report, Wrist Motion, and Continuous Glucose Monitoring to Detect Eating Occasions in Adults With Prediabetes and Obesity

BACKGROUND

Accurately identifying eating patterns, specifically the timing, frequency, and distribution of eating occasions (EOs), is important for assessing eating behaviors, especially for preventing and managing obesity and type 2 diabetes (T2D). However, existing methods to study EOs rely on self-report, which may be prone to misreporting and bias and has a high user burden. Therefore, objective methods are needed.

METHODS

We aim to compare EO timing using objective and subjective methods. Participants self-reported EO with a smartphone app (self-report [SR]), wore the ActiGraph GT9X on their dominant wrist, and wore a continuous glucose monitor (CGM, Abbott Libre Pro) for 10 days. EOs were detected from wrist motion (WM) using a motion-based classifier and from CGM using a simulation-based system. We described EO timing and explored how timing identified with WM and CGM compares with SR.

RESULTS

Participants (n = 39) were 59 ± 11 years old, mostly female (62%) and White (51%) with a body mass index (BMI) of 34.2 ± 4.7 kg/m2. All had prediabetes or moderately controlled T2D. The median time-of-day first EO (and interquartile range) for SR, WM, and CGM were 08:24 (07:00-09:59), 9:42 (07:46-12:26), and 06:55 (04:23-10:03), respectively. The median last EO for SR, WM, and CGM were 20:20 (16:50-21:42), 20:12 (18:30-21:41), and 21:43 (20:35-22:16), respectively. The overlap between SR and CGM was 55% to 80% of EO detected with tolerance periods of ±30, 60, and 120 minutes. The overlap between SR and WM was 52% to 65% EO detected with tolerance periods of ±30, 60, and 120 minutes.

CONCLUSION

The continuous glucose monitor and WM detected overlapping but not identical meals and may provide complementary information to self-reported EO.

Referent data for investigations of upper limb accelerometry: harmonized data from three cohorts of typically-developing children

Aim

The rise of wearable sensing technology shows promise for addressing the challenges of measuring motor behavior in pediatric populations. The current pediatric wearable sensing literature is highly variable with respect to the number of sensors used, sensor placement, wearing time, and how data extracted from the sensors are analyzed. Many studies derive conceptually similar variables via different calculation methods, making it hard to compare across studies and clinical populations. In hopes of moving the field forward, this report provides referent upper limb wearable sensor data from accelerometers on 25 variables in typically-developing children, ages 3-17 years.

Methods

This is a secondary analysis of data from three pediatric cohorts of children 3-17 years of age. Participants (n = 222) in the cohorts wore bilateral wrist accelerometers for 2-4 days for a total of 622 recording days. Accelerometer data were reprocessed to compute 25 variables that quantified upper limb movement duration, intensity, symmetry, and complexity. Analyses examined the influence of hand dominance, age, gender, reliability, day-to-day stability, and the relationships between variables.

Results

The majority of variables were similar on the dominant and non-dominant sides, declined slightly with age, and were not different between boys and girls. ICC values were moderate to excellent. Variation within individuals across days generally ranged from 3% to 32%. A web-based R shiny object is available for data viewing.

Interpretation

With the use of wearable movement sensors increasing rapidly, these data provide key, referent information for researchers as they design studies, and analyze and interpret data from neurodevelopmental and other pediatric clinical populations. These data may be of particularly high value for pediatric rare diseases.

Multisensor Integrated Platform Based on MEMS Charge Variation Sensing Technology for Biopotential Acquisition

We propose a new methodology for long-term biopotential recording based on an MEMS multisensor integrated platform featuring a commercial electrostatic charge-transfer sensor. This family of sensors was originally intended for presence tracking in the automotive industry, so the existing setup was engineered for the acquisition of electrocardiograms, electroencephalograms, electrooculograms, and electromyography, designing a dedicated front-end and writing proper firmware for the specific application. Systematic tests on controls and nocturnal acquisitions from patients in a domestic environment will be discussed in detail. The excellent results indicate that this technology can provide a low-power, unexplored solution to biopotential acquisition. The technological breakthrough is in that it enables adding this type of functionality to existing MEMS boards at near-zero additional power consumption. For these reasons, it opens up additional possibilities for wearable sensors and strengthens the role of MEMS technology in medical wearables for the long-term synchronous acquisition of a wide range of signals.

Missing Data Statistics Provide Causal Insights into Data Loss in Diabetes Health Monitoring by Wearable Sensors

BACKGROUND

Data loss in wearable sensors is an inevitable problem that leads to misrepresentation during diabetes health monitoring. We systematically investigated missing wearable sensors data to get causal insight into the mechanisms leading to missing data.

METHODS

Two-week-long data from a continuous glucose monitor and a Fitbit activity tracker recording heart rate (HR) and step count in free-living patients with type 2 diabetes mellitus were used. The gap size distribution was fitted with a Planck distribution to test for missing not at random (MNAR) and a difference between distributions was tested with a Chi-squared test. Significant missing data dispersion over time was tested with the Kruskal-Wallis test and Dunn post hoc analysis.

RESULTS

Data from 77 subjects resulted in 73 cleaned glucose, 70 HR and 68 step count recordings. The glucose gap sizes followed a Planck distribution. HR and step count gap frequency differed significantly (p < 0.001), and the missing data were therefore MNAR. In glucose, more missing data were found in the night (23:00-01:00), and in step count, more at measurement days 6 and 7 (p < 0.001). In both cases, missing data were caused by insufficient frequency of data synchronization.

CONCLUSIONS

Our novel approach of investigating missing data statistics revealed the mechanisms for missing data in Fitbit and CGM data.

Optimization of Torque-Control Model for Quasi-Direct-Drive Knee Exoskeleton Robots Based on Regression Forecasting

The choice of torque curve in lower-limb enhanced exoskeleton robots is a key problem in the control of lower-limb exoskeleton robots. As a human-machine coupled system, mapping from sensor data to joint torque is complex and non-linear, making it difficult to accurately model using mathematical tools. In this research study, the knee torque data of an exoskeleton robot climbing up stairs were obtained using an optical motion-capture system and three-dimensional force-measuring tables, and the inertial measurement unit (IMU) data of the lower limbs of the exoskeleton robot were simultaneously collected. Nonlinear approximations can be learned using machine learning methods. In this research study, a multivariate network model combining CNN and LSTM was used for nonlinear regression forecasting, and a knee joint torque-control model was obtained. Due to delays in mechanical transmission, communication, and the bottom controller, the actual torque curve will lag behind the theoretical curve. In order to compensate for these delays, different time shifts of the torque curve were carried out in the model-training stage to produce different control models. The above model was applied to a lightweight knee exoskeleton robot. The performance of the exoskeleton robot was evaluated using surface electromyography (sEMG) experiments, and the effects of different time-shifting parameters on the performance were compared. During testing, the sEMG activity of the rectus femoris (RF) decreased by 20.87%, while the sEMG activity of the vastus medialis (VM) increased by 17.45%. The experimental results verify the effectiveness of this control model in assisting knee joints in climbing up stairs.

Relationship of Knee Abduction Moment to Trunk and Lower Extremity Segment Acceleration during Sport-Specific Movements

The knee abduction moment (KAM) has been identified as a significant predictor of anterior cruciate ligament (ACL) injury risk; however, the cost and time demands associated with collecting three-dimensional (3D) kinetic data have prompted the need for alternative solutions. Wearable inertial measurement units (IMUs) have been explored as a potential solution for quantitative on-field assessment of injury risk. Most previous work has focused on angular velocity data, which are highly susceptible to bias and noise relative to acceleration data. The purpose of this pilot study was to assess the relationship between KAM and body segment acceleration during sport-specific movements. Three functional tasks were selected to analyze peak KAM using optical motion capture and force plates as well as peak triaxial segment accelerations using IMUs. Moderate correlations with peak KAM were observed for peak shank acceleration during single-leg hop; peak trunk, thigh, and shank accelerations during a deceleration task; and peak trunk, pelvis, and shank accelerations during a 45° cut. These findings provide preliminary support for the use of wearable IMUs to identify peak KAM during athletic tasks.

Rapid Self-Healing Hydrogel with Ultralow Electrical Hysteresis for Wearable Sensing

Self-healing hydrogels are in high demand for wearable sensing applications due to their remarkable deformability, high ionic and electrical conductivity, self-adhesiveness to human skin, as well as resilience to both mechanical and electrical damage. However, these hydrogels face challenges such as delayed healing times and unavoidable electrical hysteresis, which limit their practical effectiveness. Here, we introduce a self-healing hydrogel that exhibits exceptionally rapid healing with a recovery time of less than 0.12 s and an ultralow electrical hysteresis of less than 0.64% under cyclic strains of up to 500%. This hydrogel strikes an ideal balance, without notable trade-offs, between properties such as softness, deformability, ionic and electrical conductivity, self-adhesiveness, response and recovery times, durability, overshoot behavior, and resistance to nonaxial deformations such as twisting, bending, and pressing. Owing to this unique combination of features, the hydrogel is highly suitable for long-term, durable use in wearable sensing applications, including monitoring body movements and electrophysiological activities on the skin.

Detecting Psychological Interventions Using Bilateral Electromyographic Wearable Sensors

This study investigated the impact of auditory stimuli on muscular activation patterns using wearable surface electromyography (EMG) sensors. Employing four key muscles (Sternocleidomastoid Muscle (SCM), Cervical Erector Muscle (CEM), Quadricep Muscles (QMs), and Tibialis Muscle (TM)) and time domain features, we differentiated the effects of four interventions: silence, music, positive reinforcement, and negative reinforcement. The results demonstrated distinct muscle responses to the interventions, with the SCM and CEM being the most sensitive to changes and the TM being the most active and stimulus dependent. Post hoc analyses revealed significant intervention-specific activations in the CEM and TM for specific time points and intervention pairs, suggesting dynamic modulation and time-dependent integration. Multi-feature analysis identified both statistical and Hjorth features as potent discriminators, reflecting diverse adaptations in muscle recruitment, activation intensity, control, and signal dynamics. These features hold promise as potential biomarkers for monitoring muscle function in various clinical and research applications. Finally, muscle-specific Random Forest classification achieved the highest accuracy and Area Under the ROC Curve for the TM, indicating its potential for differentiating interventions with high precision. This study paves the way for personalized neuroadaptive interventions in rehabilitation, sports science, ergonomics, and healthcare by exploiting the diverse and dynamic landscape of muscle responses to auditory stimuli.

Resorbable conductive materials for optimally interfacing medical devices with the living

Implantable and wearable bioelectronic systems are arising growing interest in the medical field. Linking the microelectronic (electronic conductivity) and biological (ionic conductivity) worlds, the biocompatible conductive materials at the electrode/tissue interface are key components in these systems. We herein focus more particularly on resorbable bioelectronic systems, which can safely degrade in the biological environment once they have completed their purpose, namely, stimulating or sensing biological activity in the tissues. Resorbable conductive materials are also explored in the fields of tissue engineering and 3D cell culture. After a short description of polymer-based substrates and scaffolds, and resorbable electrical conductors, we review how they can be combined to design resorbable conductive materials. Although these materials are still emerging, various medical and biomedical applications are already taking shape that can profoundly modify post-operative and wound healing follow-up. Future challenges and perspectives in the field are proposed.

Experiences with Wearable Sensors in Oncology during Treatment: Lessons Learned from Feasibility Research Projects in Denmark

BACKGROUND

The fraction of elderly people in the population is growing, the incidence of some cancers is increasing, and the number of available cancer treatments is evolving, causing a challenge to healthcare systems. New healthcare tools are needed, and wearable sensors could partly be potential solutions. The aim of this case report is to describe the Danish research experience with wearable sensors in oncology reporting from three oncological wearable research projects.

CASE STUDIES

Three planned case studies investigating the feasibility of different wearable sensor solutions during cancer treatment are presented, focusing on study design, population, device, aim, and planned outcomes. Further, two actual case studies performed are reported, focusing on patients included, data collected, results achieved, further activities planned, and strengths and limitations.

RESULTS

Only two of the three planned studies were performed. In general, patients found the technical issues of wearable sensors too challenging to deal with during cancer treatment. However, at the same time it was demonstrated that a large amount of data could be collected if the framework worked efficiently.

CONCLUSION

Wearable sensors have the potential to help solve challenges in clinical oncology, but for successful research projects and implementation, a setup with minimal effort on the part of patients is requested.

Incorporating Wearable Technology for Enhanced Rehabilitation Monitoring after Hip and Knee Replacement

Osteoarthritis (OA) poses a growing challenge for the aging population, especially in the hip and knee joints, contributing significantly to disability and societal costs. Exploring the integration of wearable technology, this study addresses the limitations of traditional rehabilitation assessments in capturing real-world experiences and dynamic variations. Specifically, it focuses on continuously monitoring physical activity in hip and knee OA patients using automated unsupervised evaluations within the rehabilitation process. We analyzed data from 1144 patients who used a mobile health application after surgery; the activity data were collected using the Garmin Vivofit 4. Several parameters, such as the total number of steps per day, the peak 6-minute consecutive cadence (P6MC) and peak 1-minute cadence (P1M), were computed and analyzed on a daily basis. The results indicated that cadence-based measurements can effectively, and earlier, differ among patients with hip and knee conditions, as well as in the recovery process. Comparisons based on recovery status and type of surgery reveal distinctive trajectories, emphasizing the effectiveness of P6MC and P1M in detecting variations earlier than total steps per day. Furthermore, cadence-based measurements showed a lower inter-day variability (40%) compared to the total number of steps per day (80%). Automated assessments, including P1M and P6MC, offer nuanced insights into the patients' dynamic activity profiles.

Urban environment influences on stress, autonomic reactivity and circadian rhythm: protocol for an ambulatory study of mental health and sleep

Introduction

Converging evidence suggests that urban living is associated with an increased likelihood of developing mental health and sleep problems. Although these aspects have been investigated in separate streams of research, stress, autonomic reactivity and circadian misalignment can be hypothesized to play a prominent role in the causal pathways underlining the complex relationship between the urban environment and these two health dimensions. This study aims at quantifying the momentary impact of environmental stressors on increased autonomic reactivity and circadian rhythm, and thereby on mood and anxiety symptoms and sleep quality in the context of everyday urban living.

Method

The present article reports the protocol for a feasibility study that aims at assessing the daily environmental and mobility exposures of 40 participants from the urban area of Jerusalem over 7 days. Every participant will carry a set of wearable sensors while being tracked through space and time with GPS receivers. Skin conductance and heart rate variability will be tracked to monitor participants' stress responses and autonomic reactivity, whereas electroencephalographic signal will be used for sleep quality tracking. Light exposure, actigraphy and skin temperature will be used for ambulatory circadian monitoring. Geographically explicit ecological momentary assessment (GEMA) will be used to assess participants' perception of the environment, mood and anxiety symptoms, sleep quality and vitality. For each outcome variable (sleep quality and mental health), hierarchical mixed models including random effects at the individual level will be used. In a separate analysis, to control for potential unobserved individual-level confounders, a fixed effect at the individual level will be specified for case-crossover analyses (comparing each participant to oneself).

Conclusion

Recent developments in wearable sensing methods, as employed in our study or with even more advanced methods reviewed in the Discussion, make it possible to gather information on the functioning of neuro-endocrine and circadian systems in a real-world context as a way to investigate the complex interactions between environmental exposures, behavior and health. Our work aims to provide evidence on the health effects of urban stressors and circadian disruptors to inspire potential interventions, municipal policies and urban planning schemes aimed at addressing those factors.

Digital Mobility Measures: A Window into Real-World Severity and Progression of Parkinson's Disease

BACKGROUND

Real-world monitoring using wearable sensors has enormous potential for assessing disease severity and symptoms among persons with Parkinson's disease (PD). Many distinct features can be extracted, reflecting multiple mobility domains. However, it is unclear which digital measures are related to PD severity and are sensitive to disease progression.

OBJECTIVES

The aim was to identify real-world mobility measures that reflect PD severity and show discriminant ability and sensitivity to disease progression, compared to the Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS) scale.

METHODS

Multicenter real-world continuous (24/7) digital mobility data from 587 persons with PD and 68 matched healthy controls were collected using an accelerometer adhered to the lower back. Machine learning feature selection and regression algorithms evaluated associations of the digital measures using the MDS-UPDRS (I-III). Binary logistic regression assessed discriminatory value using controls, and longitudinal observational data from a subgroup (n = 33) evaluated sensitivity to change over time.

RESULTS

Digital measures were only moderately correlated with the MDS-UPDRS (part II-r = 0.60 and parts I and III-r = 0.50). Most associated measures reflected activity quantity and distribution patterns. A model with 14 digital measures accurately distinguished recently diagnosed persons with PD from healthy controls (81.1%, area under the curve: 0.87); digital measures showed larger effect sizes (Cohen's d: [0.19-0.66]), for change over time than any of the MDS-UPDRS parts (Cohen's d: [0.04-0.12]).

CONCLUSIONS

Real-world mobility measures are moderately associated with clinical assessments, suggesting that they capture different aspects of motor capacity and function. Digital mobility measures are sensitive to early-stage disease and to disease progression, to a larger degree than conventional clinical assessments, demonstrating their utility, primarily for clinical trials but ultimately also for clinical care. © 2023 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Reliability of Xsens IMU-Based Lower Extremity Joint Angles during In-Field Running

The Xsens Link motion capture suit has become a popular tool in investigating 3D running kinematics based on wearable inertial measurement units outside of the laboratory. In this study, we investigated the reliability of Xsens-based lower extremity joint angles during unconstrained running on stable (asphalt) and unstable (woodchip) surfaces within and between five different testing days in a group of 17 recreational runners (8 female, 9 male). Specifically, we determined the within-day and between-day intraclass correlation coefficients (ICCs) and minimal detectable changes (MDCs) with respect to discrete ankle, knee, and hip joint angles. When comparing runs within the same day, the investigated Xsens-based joint angles generally showed good to excellent reliability (median ICCs > 0.9). Between-day reliability was generally lower than the within-day estimates: Initial hip, knee, and ankle angles in the sagittal plane showed good reliability (median ICCs > 0.88), while ankle and hip angles in the frontal plane showed only poor to moderate reliability (median ICCs 0.38-0.83). The results were largely unaffected by the surface. In conclusion, within-day adaptations in lower-extremity running kinematics can be captured with the Xsens Link system. Our data on between-day reliability suggest caution when trying to capture longitudinal adaptations, specifically for ankle and hip joint angles in the frontal plane.

Avoiding Potential Pitfalls in Designing Wired Glucose Biosensors

Glucose sensing has been studied for more than half a century, leading many to believe that further progress comes mainly from engineering efforts. Our society requires robust, reliable, compact, and easy-to-use sensing solutions for decentralized applications such as wearables, and engineering solutions are essential. However, true progress is only possible by understanding and improving the underlying working principles and fundamental limitations. This Perspective discusses the delicate relationship between the observed current and glucose concentration when using wired enzyme biosensors. Some of the potential pitfalls often encountered in the recent literature are discussed. These include the need to suppress the influence of enzyme turnover kinetics on the sensor signal and the undesired faradaic charging of the electron transfer mediator that gives a continuously decaying baseline signal. These fundamental issues must be carefully evaluated and resolved for the realization of continuously operating enzyme biosensor systems.

Fall Risk Assessment in Stroke Survivors: A Machine Learning Model Using Detailed Motion Data from Common Clinical Tests and Motor-Cognitive Dual-Tasking

BACKGROUND

Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy.

OBJECTIVE

Develop machine learning models using inertial sensors to objectively classify fall risk in stroke survivors. Determine optimal sensor configurations and clinical test protocols.

METHODS

21 stroke survivors performed balance, Timed Up and Go, 10 Meter Walk, and Sit-to-Stand tests with and without dual-tasking. A total of 8 motion sensors captured lower limb and trunk kinematics, and 92 spatiotemporal gait and clinical features were extracted. Supervised models-Support Vector Machine, Logistic Regression, and Random Forest-were implemented to classify high vs. low fall risk. Sensor setups and test combinations were evaluated.

RESULTS

The Random Forest model achieved 91% accuracy using dual-task balance sway and Timed Up and Go walk time features. Single thorax sensor models performed similarly to multi-sensor models. Balance and Timed Up and Go best-predicted fall risk.

CONCLUSION

Machine learning models using minimal inertial sensors during clinical assessments can accurately quantify fall risk in stroke survivors. Single thorax sensor setups are effective. Findings demonstrate a feasible objective fall screening approach to assist rehabilitation.

Validity of the CALERA Research Sensor to Assess Body Core Temperature during Maximum Exercise in Patients with Heart Failure

(1) Background: It is important to monitor the body core temperature (Tc) of individuals with chronic heart failure (CHF) during rest or exercise, as they are susceptible to complications. Gastrointestinal capsules are a robust indicator of the Tc at rest and during exercise. A practical and non-invasive sensor called CALERA Research was recently introduced, promising accuracy, sensitivity, continuous real-time analysis, repeatability, and reproducibility. This study aimed to assess the validity of the CALERA Research sensor when monitoring patients with CHF during periods of rest, throughout brief cardiopulmonary exercise testing, and during their subsequent recovery. (2) Methods: Twelve male CHF patients volunteered to participate in a 70-min protocol in a laboratory at 28 °C and 39% relative humidity. After remaining calm for 20 min, they underwent a symptom-limited stress test combined with ergospirometry on a treadmill, followed by 40 min of seated recovery. The Tc was continuously monitored by both Tc devices. (3) Results: The Tc values from the CALERA Research sensor and the gastrointestinal sensor showed no associations at rest (r = 0.056, p = 0.154) and during exercise (r = -0.015, p = 0.829) and a weak association during recovery (r = 0.292, p < 0.001). The Cohen's effect size of the differences between the two Tc assessment methods for rest, exercise, and recovery was 1.04 (large), 0.18 (none), and 0.45 (small), respectively. The 95% limit of agreement for the CALERA Research sensor was -0.057 ± 1.03 °C. (4) Conclusions: The CALERA sensor is a practical and, potentially, promising device, but it does not provide an accurate Tc estimation in CHF patients at rest, during brief exercise testing, and during recovery.

Femtosecond Laser Direct Writing of Flexible Electronic Devices: A Mini Review

By virtue of its narrow pulse width and high peak power, the femtosecond pulsed laser can achieve high-precision material modification, material additive or subtractive, and other forms of processing. With additional good material adaptability and process compatibility, femtosecond laser-induced application has achieved significant progress in flexible electronics in recent years. These advancements in the femtosecond laser fabrication of flexible electronic devices are comprehensively summarized here. This review first briefly introduces the physical mechanism and characteristics of the femtosecond laser fabrication of various electronic microdevices. It then focuses on effective methods of improving processing efficiency, resolution, and size. It further highlights the typical progress of applications, including flexible energy storage devices, nanogenerators, flexible sensors, and detectors, etc. Finally, it discusses the development tendency of ultrashort pulse laser processing. This review should facilitate the precision manufacturing of flexible electronics using a femtosecond laser.

Radiation Detector Front-End Readout Chip with Nonbinary Successive Approximation Register Analog-to-Digital Converter for Wearable Healthcare Monitoring Applications

A 16-channel front-end readout chip for a radiation detector is designed for portable or wearable healthcare monitoring applications. The proposed chip reads the signal of the radiation detector and converts it into digital serial-out data by using a nonbinary successive approximation register (SAR) analog-to-digital converter (ADC) that has a 1-MS/s sampling rate and 10-b resolution. The minimum-to-maximum differential and integral nonlinearity are measured as -0.32 to 0.33 and -0.43 to 0.37 least significant bits, respectively. The signal-to-noise-and-distortion ratio and effective number of bits are 57.41 dB and 9.24 bits, respectively, for an input frequency of 500 kHz and a sampling rate of 1 MS/s. The SAR ADC has a 38.9-fJ/conversion step figure of merit at the sampling rate of 1 MS/s. The proposed chip can read input signals with peak currents ranging from 20 to 750 μA and convert the analog signal into a 10-bit serial-output digital signal. The input dynamic range is 2-75 pC. The resolution of the peak current is 208.3 nA. The chip, which has an area of 1.444 mm × 10.568 mm, is implemented using CMOS 0.18-μm 1P6M technology, and the power consumption of each channel is 19 mW. This design is suitable for wearable devices, especially biomedical devices.

The Importance of Digital Health Literacy in an Evolving Parkinson's Disease Care System

Digital health technologies are growing at a rapid pace and changing the healthcare landscape. Our current understanding of digital health literacy in Parkinson's disease (PD) is limited. In this review, we discuss the potential challenges of low digital health literacy in PD with particular attention to telehealth, deep brain stimulation, wearable sensors, and smartphone applications. We also highlight inequities in access to digital health technologies. Future research is needed to better understand digital health literacy among individuals with PD and to develop effective solutions. We must invest resources to evaluate, understand, and enhance digital health literacy for individuals with PD.

Wearable Sensors as a Preoperative Assessment Tool: A Review

Surgery is a common first-line treatment for many types of disease, including cancer. Mortality rates after general elective surgery have seen significant decreases whilst postoperative complications remain a frequent occurrence. Preoperative assessment tools are used to support patient risk stratification but do not always provide a precise and accessible assessment. Wearable sensors (WS) provide an accessible alternative that offers continuous monitoring in a non-clinical setting. They have shown consistent uptake across the perioperative period but there has been no review of WS as a preoperative assessment tool. This paper reviews the developments in WS research that have application to the preoperative period. Accelerometers were consistently employed as sensors in research and were frequently combined with photoplethysmography or electrocardiography sensors. Pre-processing methods were discussed and missing data was a common theme; this was dealt with in several ways, commonly by employing an extraction threshold or using imputation techniques. Research rarely processed raw data; commercial devices that employ internal proprietary algorithms with pre-calculated heart rate and step count were most commonly employed limiting further feature extraction. A range of machine learning models were used to predict outcomes including support vector machines, random forests and regression models. No individual model clearly outperformed others. Deep learning proved successful for predicting exercise testing outcomes but only within large sample-size studies. This review outlines the challenges of WS and provides recommendations for future research to develop WS as a viable preoperative assessment tool.

Associations Between Coordination and Wearable Sensor Variables Vary by Recording Context but Not Assessment Type

Motor coordination is an important driver of development and improved coordination assessments could facilitate better screening, diagnosis, and intervention for children at risk of developmental disorders. Wearable sensors could provide data that enhance the characterization of coordination and the clinical utility of that data may vary depending on how sensor variables from different recording contexts relate to coordination. We used wearable sensors at the wrists to capture upper-limb movement in 85 children aged 6-12. Sensor variables were extracted from two recording contexts. Structured recordings occurred in the lab during a unilateral throwing task. Unstructured recordings occurred during free-living activity. The objective was to determine the influence of recording context (unstructured versus structured) and assessment type (direct vs. indirect) on the association between sensor variables and coordination. The greatest associations were between six sensor variables from the structured context and the direct measure of coordination. Worse coordination scores were associated with upper-limb movements that had higher peak magnitudes, greater variance, and less smoothness. The associations were consistent across both arms, even though the structured task was unilateral. This finding suggests that wearable sensors could be paired with a simple, structured task to yield clinically informative variables that relate to motor coordination.

Real-world characterization of vestibular contributions during locomotion

Introduction

The vestibular system, which encodes our head movement in space, plays an important role in maintaining our balance as we navigate the environment. While in-laboratory research demonstrates that the vestibular system exerts a context-dependent influence on the control of balance during locomotion, differences in whole-body and head kinematics between indoor treadmill and real-world locomotion challenge the generalizability of these findings. Thus, the goal of this study was to characterize vestibular-evoked balance responses in the real world using a fully portable system.

Methods

While experiencing stochastic electrical vestibular stimulation (0-20 Hz, amplitude peak ± 4.5 mA, root mean square 1.25 mA) and wearing inertial measurement units (IMUs) on the head, low back, and ankles, 10 participants walked outside at 52 steps/minute (∼0.4 m/s) and 78 steps/minute (∼0.8 m/s). We calculated time-dependent coherence (a measure of correlation in the frequency domain) between the applied stimulus and the mediolateral back, right ankle, and left ankle linear accelerations to infer the vestibular control of balance during locomotion.

Results

In all participants, we observed vestibular-evoked balance responses. These responses exhibited phasic modulation across the stride cycle, peaking during the middle of the single-leg stance in the back and during the stance phase for the ankles. Coherence decreased with increasing locomotor cadence and speed, as observed in both bootstrapped coherence differences (p < 0.01) and peak coherence (low back: 0.23 ± 0.07 vs. 0.16 ± 0.14, p = 0.021; right ankle: 0.38 ± 0.12 vs. 0.25 ± 0.10, p < 0.001; left ankle: 0.33 ± 0.09 vs. 0.21 ± 0.09, p < 0.001).

Discussion

These results replicate previous in-laboratory studies, thus providing further insight into the vestibular control of balance during naturalistic movements and validating the use of this portable system as a method to characterize real-world vestibular responses. This study will help support future work that seeks to better understand how the vestibular system contributes to balance in variable real-world environments.

Wearable sensor supports in-situ and continuous monitoring of plant health in precision agriculture era

Plant health is intricately linked to crop quality, food security and agricultural productivity. Obtaining accurate plant health information is of paramount importance in the realm of precision agriculture. Wearable sensors offer an exceptional avenue for investigating plant health status and fundamental plant science, as they enable real-time and continuous in-situ monitoring of physiological biomarkers. However, a comprehensive overview that integrates and critically assesses wearable plant sensors across various facets, including their fundamental elements, classification, design, sensing mechanism, fabrication, characterization and application, remains elusive. In this study, we provide a meticulous description and systematic synthesis of recent research progress in wearable sensor properties, technology and their application in monitoring plant health information. This work endeavours to serve as a guiding resource for the utilization of wearable plant sensors, empowering the advancement of plant health within the precision agriculture paradigm.

Informing Acceptability and Feasibility of Digital Phenotyping for Personalized HIV Prevention among Marginalized Populations Presenting to the Emergency Department

For marginalized populations with ongoing HIV epidemics, alternative methods are needed for understanding the complexities of HIV risk and delivering prevention interventions. Due to lack of engagement in ambulatory care, such groups have high utilization of drop-in care. Therefore, emergency departments represent a location with those at highest risk for HIV and in highest need of novel prevention methods. Digital phenotyping via data collected from smartphones and other wearable sensors could provide the innovative vehicle for examining complex HIV risk and assist in delivering personalized prevention interventions. However, there is paucity in exploring if such methods are an option. This study aimed to fill this gap via a cross-sectional psychosocial assessment with a sample of N=85 emergency department patients with HIV risk. Findings demonstrate that although potentially feasible, acceptability of digital phenotyping is questionable. Technology-assisted HIV prevention needs to be designed with the target community and address key ethical considerations.

Prediction of stroke patients' bedroom-stay duration: machine-learning approach using wearable sensor data

Background: The importance of being physically active and avoiding staying in bed has been recognized in stroke rehabilitation. However, studies have pointed out that stroke patients admitted to rehabilitation units often spend most of their day immobile and inactive, with limited opportunities for activity outside their bedrooms. To address this issue, it is necessary to record the duration of stroke patients staying in their bedrooms, but it is impractical for medical providers to do this manually during their daily work of providing care. Although an automated approach using wearable devices and access points is more practical, implementing these access points into medical facilities is costly. However, when combined with machine learning, predicting the duration of stroke patients staying in their bedrooms is possible with reduced cost. We assessed using machine learning to estimate bedroom-stay duration using activity data recorded with wearable devices. Method: We recruited 99 stroke hemiparesis inpatients and conducted 343 measurements. Data on electrocardiograms and chest acceleration were measured using a wearable device, and the location name of the access point that detected the signal of the device was recorded. We first investigated the correlation between bedroom-stay duration measured from the access point as the objective variable and activity data measured with a wearable device and demographic information as explanatory variables. To evaluate the duration predictability, we then compared machine-learning models commonly used in medical studies. Results: We conducted 228 measurements that surpassed a 90% data-acquisition rate using Bluetooth Low Energy. Among the explanatory variables, the period spent reclining and sitting/standing were correlated with bedroom-stay duration (Spearman's rank correlation coefficient (R) of 0.56 and -0.52, p < 0.001). Interestingly, the sum of the motor and cognitive categories of the functional independence measure, clinical indicators of the abilities of stroke patients, lacked correlation. The correlation between the actual bedroom-stay duration and predicted one using machine-learning models resulted in an R of 0.72 and p < 0.001, suggesting the possibility of predicting bedroom-stay duration from activity data and demographics. Conclusion: Wearable devices, coupled with machine learning, can predict the duration of patients staying in their bedrooms. Once trained, the machine-learning model can predict without continuously tracking the actual location, enabling more cost-effective and privacy-centric future measurements.

Unrevealing the connection between real sample analysis and analytical method. The case of cytokines

Cytokines are compounds that belong to a special class of signaling biomolecules that are responsible for several functions in the human body, being involved in cell growth, inflammatory, and neoplastic processes. Thus, they represent valuable biomarkers for diagnosing and drug therapy monitoring certain medical conditions. Because cytokines are secreted in the human body, they can be detected in both conventional samples, such as blood or urine, but also in samples less used in medical practice such as sweat or saliva. As the importance of cytokines was identified, various analytical methods for their determination in biological fluids were reported. The gold standard in cytokine detection is considered the enzyme-linked immunosorbent assay method and the most recent ones have been considered and compared in this study. It is known that the conventional methods are accompanied by a few disadvantages that new methods of analysis, especially electrochemical sensors, are trying to overcome. Electrochemical sensors proved to be suited for the elaboration of integrated, portable, and wearable sensing devices, which could also facilitate cytokines determination in medical practice.

A novel measurement approach to dynamic change of limb length discrepancy using deep learning and wearable sensors

The accurate identification of dynamic change of limb length discrepancy (LLD) in non-clinical settings is of great significance for monitoring gait function change in people's everyday lives. How to search for advanced techniques to measure LLD changes in non-clinical settings has always been a challenging endeavor in recent related research. In this study, we have proposed a novel approach to accurately measure the dynamic change of LLD outdoors by using deep learning and wearable sensors. The basic idea is that the measurement of dynamic change of LLD was considered as a multiple gait classification task based on LLD change that is clearly associated with its gait pattern. A hybrid deep learning model of convolutional neural network and long short-term memory (CNN-LSTM) was developed to precisely classify LLD gait patterns by discovering the most representative spatial-temporal LLD dynamic change features. Twenty-three healthy subjects were recruited to simulate four levels of LLD by wearing a shoe lift with different heights. The Delsys TrignoTM system was implemented to simultaneously acquire gait data from six sensors positioned on the hip, knee and ankle joint of two lower limbs respectively. The experimental results showed that the developed CNN-LSTM model could reach a higher accuracy of 93.24% and F1-score of 93.48% to classify four different LLD gait patterns when compared with CNN, LSTM, and CNN-gated recurrent unit(CNN-GRU), and gain better recall and precision (more than 92%) to detect each LLD gait pattern accurately. Our model could achieve excellent learning ability to discover the most representative LLD dynamic change features for classifying LLD gait patterns accurately. Our technical solution would help not only to accurately measure LLD dynamic change in non-clinical settings, but also to potentially find out lower limb joints with more abnormal compensatory change caused by LLD.

Examining the influence of the Get aHEAD Safely in Soccer™ program on head impact kinematics and neck strength in female youth soccer players

The objective was to examine the efficacy of the Get aHEAD Safely in Soccer™ intervention on head impact kinematics and neck strength in female youth soccer players. The control group (CG) consisted of 13 players (age: 11.0 ± 0.4 yrs), while the experimental group (EG) consisted of 14 players (age: 10.6 ± 0.5 yrs). Head impact kinematics included peak linear acceleration (PLA), peak rotational acceleration (PRA), and peak rotational velocity (PRV). Pre- and post-season measures included strength measures of neck/torso flexion (NF/TF) and extension (NE/TE). Data were analysed using a multilevel linear model and ANOVA techniques. No differences in PLA, PRA, or PRV were observed between groups. The EG showed significant improvement in NF strength while the CG showed significant improvement in NE strength. Both groups significantly improved in TF pre- to post-season. The foundational strength components of the Get aHEAD Safely in Soccer program appear to show a benefit in youth soccer players beginning to learn the skill of purposeful heading.

Environment Endurable, Self-Healing, Super-Adhesive, and Mechanically Strong Ionogels for Reliable Sensing

Ionogels possess high conductivity, stretchability, and adhesion, making them promising as flexible sensors. However, it remains challenging to fabricate an ionogel which integrates excellent environment endurance, superior mechanical strength, high self-healing efficiency, and super adhesion. Herein, a supramolecular ionic liquid is synthesized using calcium chloride and 1-butyl-3-methylimidazolium chloride. An advanced ionogel based on this supramolecular ionic liquid is conveniently constructed by a one-pot method with acrylamide and acrylic acid as monomers. The supramolecular cross-linking network, formed by affluent coordination interactions, hydrogen bonds, and electrostatic interactions, provides the ionogel with ideal mechanical strength (tensile strength up to 1.7 MPa), high self-healing efficiency (up to 149%), super adhesion (up to 358 kPa on aluminum), excellent solvent tolerance (less than 10% weight increase, high mechanical and sensing performance retention after being soaked in organic solvents), and low-temperature endurance (breaking elongation can reach 87% at -30 °C). The supramolecular ionogels can function as multi-mode sensors, capable of monitoring strain and different amplitudes of human movements in real-time. Moreover, the sensing performance of ionogels remains unaffected even after being self-healed or exposure to organic solvents. It is expected that this study could offer valuable design ideas to construct advanced gel materials applicable in complicated environment.

Healthcare Wearable Sensors Adhesion to Human Fingernails and Toenails

A novel adhesion method of a sensor to a fingernail is described. Wearable sensors can provide health insights to humans for a wide variety of benefits, such as continuous wellness monitoring and disease monitoring throughout a patient's daily life. While there are many locations to place these wearable sensors on the body, we will focus on the fingertip, one significant way that people interact with the world. Like artificial fingernails used for aesthetics, wearable healthcare sensors can be attached to the fingernail for short or long time periods with minimal irritation and disruption to daily life. In this study the structure and methods of healthcare sensors' attachment and removal have been explored to support (1) the sensor functional requirements, (2) biological and environmentally compatible solutions and (3) ease of attachment and removal for short- and long-term user applications. Initial fingernail sensors were attached using a thin adhesive layer of commonly available cosmetic nail glue. While this approach allowed for easy application and strong adhesion to the nail, the removal could expose the fingernail and finger to a commercially available cosmetic nail removal (acetone-based chemical) for extended times measured in minutes. Therefore, a novel structure and method were developed for rapid healthcare sensor attachment and removal in seconds, which supported both the sensor functional objectives and the biologically and environmentally safe use objectives.

Advances in Non-Electrochemical Sensing of Human Sweat Biomarkers: From Sweat Sampling to Signal Reading

Sweat, commonly referred to as the ultrafiltrate of blood plasma, is an essential physiological fluid in the human body. It contains a wide range of metabolites, electrolytes, and other biologically significant markers that are closely linked to human health. Compared to other bodily fluids, such as blood, sweat offers distinct advantages in terms of ease of collection and non-invasive detection. In recent years, considerable attention has been focused on wearable sweat sensors due to their potential for continuous monitoring of biomarkers. Electrochemical methods have been extensively used for in situ sweat biomarker analysis, as thoroughly reviewed by various researchers. This comprehensive review aims to provide an overview of recent advances in non-electrochemical methods for analyzing sweat, including colorimetric methods, fluorescence techniques, surface-enhanced Raman spectroscopy, and more. The review covers multiple aspects of non-electrochemical sweat analysis, encompassing sweat sampling methodologies, detection techniques, signal processing, and diverse applications. Furthermore, it highlights the current bottlenecks and challenges faced by non-electrochemical sensors, such as limitations and interference issues. Finally, the review concludes by offering insights into the prospects for non-electrochemical sensing technologies. By providing a valuable reference and inspiring researchers engaged in the field of sweat sensor development, this paper aspires to foster the creation of innovative and practical advancements in this domain.

Comparison of state-of-the-art deep learning architectures for detection of freezing of gait in Parkinson's disease

Introduction

Freezing of gait (FOG) is one of the most debilitating motor symptoms experienced by patients with Parkinson's disease (PD). FOG detection is possible using acceleration data from wearable sensors, and a convolutional neural network (CNN) is often used to determine the presence of FOG epochs. We compared the performance of a standard CNN for the detection of FOG with two more complex networks, which are well suited for time series data, the MiniRocket and the InceptionTime.

Methods

We combined acceleration data of people with PD across four studies. The final data set was split into a training (80%) and hold-out test (20%) set. A fifth study was included as an unseen test set. The data were windowed (2 s) and five-fold cross-validation was applied. The CNN, MiniRocket, and InceptionTime models were evaluated using a receiver operating characteristic (ROC) curve and its area under the curve (AUC). Multiple sensor configurations were evaluated for the best model. The geometric mean was subsequently calculated to select the optimal threshold. The selected model and threshold were evaluated on the hold-out and unseen test set.

Results

A total of 70 participants (23.7 h, 9% FOG) were included in this study for training and testing, and in addition, 10 participants provided an unseen test set (2.4 h, 11% FOG). The CNN performed best (AUC = 0.86) in comparison to the InceptionTime (AUC = 0.82) and MiniRocket (AUC = 0.76) models. For the CNN, we found a similar performance for a seven-sensor configuration (lumbar, upper and lower legs and feet; AUC = 0.86), six-sensor configuration (upper and lower legs and feet; AUC = 0.87), and two-sensor configuration (lower legs; AUC = 0.86). The optimal threshold of 0.45 resulted in a sensitivity of 77% and a specificity of 58% for the hold-out set (AUC = 0.72), and a sensitivity of 85% and a specificity of 68% for the unseen test set (AUC = 0.90).

Conclusion

We confirmed that deep learning can be used to detect FOG in a large, heterogeneous dataset. The CNN model outperformed more complex networks. This model could be employed in future personalized interventions, with the ultimate goal of using automated FOG detection to trigger real-time cues to alleviate FOG in daily life.

Empowering Participatory Research in Urban Health: Wearable Biometric and Environmental Sensors for Activity Recognition

Participatory exposure research, which tracks behaviour and assesses exposure to stressors like air pollution, traditionally relies on time-activity diaries. This study introduces a novel approach, employing machine learning (ML) to empower laypersons in human activity recognition (HAR), aiming to reduce dependence on manual recording by leveraging data from wearable sensors. Recognising complex activities such as smoking and cooking presents unique challenges due to specific environmental conditions. In this research, we combined wearable environment/ambient and wrist-worn activity/biometric sensors for complex activity recognition in an urban stressor exposure study, measuring parameters like particulate matter concentrations, temperature, and humidity. Two groups, Group H (88 individuals) and Group M (18 individuals), wore the devices and manually logged their activities hourly and minutely, respectively. Prioritising accessibility and inclusivity, we selected three classification algorithms: k-nearest neighbours (IBk), decision trees (J48), and random forests (RF), based on: (1) proven efficacy in existing literature, (2) understandability and transparency for laypersons, (3) availability on user-friendly platforms like WEKA, and (4) efficiency on basic devices such as office laptops or smartphones. Accuracy improved with finer temporal resolution and detailed activity categories. However, when compared to other published human activity recognition research, our accuracy rates, particularly for less complex activities, were not as competitive. Misclassifications were higher for vague activities (resting, playing), while well-defined activities (smoking, cooking, running) had few errors. Including environmental sensor data increased accuracy for all activities, especially playing, smoking, and running. Future work should consider exploring other explainable algorithms available on diverse tools and platforms. Our findings underscore ML's potential in exposure studies, emphasising its adaptability and significance for laypersons while also highlighting areas for improvement.

Assessing the Effects of Mild Traumatic Brain Injury on Vestibular Home Exercise Performance with Wearable Sensors

After a mild traumatic brain injury (mTBI), dizziness and balance problems are frequently reported, affecting individuals' daily lives and functioning. Vestibular rehabilitation is a standard treatment approach for addressing these issues, but its efficacy in this population remains inconclusive. A potential reason for suboptimal outcomes is the lack of objective monitoring of exercise performance, which is crucial for therapeutic success. This study utilized wearable inertial measurement units (IMUs) to quantify exercise performance in individuals with mTBI during home-based vestibular rehabilitation exercises. Seventy-three people with mTBI and fifty healthy controls were enrolled. Vestibular exercises were performed, and IMUs measured forehead and sternum velocities and range of motions. The mTBI group demonstrated a slower forehead peak angular velocity in all exercises, which may be a compensatory strategy to manage balance issues or symptom exacerbation. Additionally, the mTBI group exhibited a larger forehead range of motion during specific exercises, potentially linked to proprioceptive deficits. These findings emphasize the usefulness of utilizing IMUs to monitor the quality of home-based vestibular exercises for individuals with mTBI and the potential for IMUs improving rehabilitation outcomes.

Physiological sensor data cleaning with autoencoders

Objective.Physiological sensor data (e.g. photoplethysmograph) is important for remotely monitoring patients' vital signals, but is often affected by measurement noise. Existing feature-based models for signal cleaning can be limited as they might not capture the full signal characteristics.Approach.In this work we present a deep learning framework for sensor signal cleaning based on dilated convolutions which capture the coarse- and fine-grained structure in order to classify whether a signal is noisy or clean. However, since obtaining annotated physiological data is costly and time-consuming we propose an autoencoder-based semi-supervised model which is able to learn a representation of the sensor signal characteristics, also adding an element of interpretability.Main results.Our proposed models are over 8% more accurate than existing feature-based approaches with half the false positive/negative rates. Finally, we show that with careful tuning (that can be improved further), the semi-supervised model outperforms supervised approaches suggesting that incorporating the large amounts of available unlabeled data can be advantageous for achieving high accuracy (over 90%) and minimizing the false positive/negative rates.Significance.Our approach enables us to reliably separate clean from noisy physiological sensor signal that can pave the development of reliable features and eventually support decisions regarding drug efficacy in clinical trials.

A Sensorised Glove to Detect Scratching for Patients with Atopic Dermatitis

In this work, a lightweight compliant glove that detects scratching using data from microtubular stretchable sensors on each finger and an inertial measurement unit (IMU) on the palm through a machine learning model is presented: the SensorIsed Glove for Monitoring Atopic Dermatitis (SIGMA). SIGMA provides the user and clinicians with a quantifiable way of assaying scratch as a proxy to itch. With the quantitative information detailing scratching frequency and duration, the clinicians would be able to better classify the severity of itch and scratching caused by atopic dermatitis (AD) more objectively to optimise treatment for the patients, as opposed to the current subjective methods of assessments that are currently in use in hospitals and research settings. The validation data demonstrated an accuracy of 83% of the scratch prediction algorithm, while a separate 30 min validation trial had an accuracy of 99% in a controlled environment. In a pilot study with children (n = 6), SIGMA accurately detected 94.4% of scratching when the glove was donned. We believe that this simple device will empower dermatologists to more effectively measure and quantify itching and scratching in AD, and guide personalised treatment decisions.