Stride Velocity 95th Centile Detects Decline in Ambulatory Function Over Shorter Intervals than the 6-Minute Walk Test or North Star Ambulatory Assessment in Duchenne Muscular Dystrophy.

Background: Stride Velocity 95th Centile (SV95C) is the first wearable device-derived clinical outcome assessment (COA) to receive European Medicines Agency (EMA) qualification as a primary endpoint in ambulant patients with Duchenne muscular dystrophy (DMD) aged ≥4 years. Objective: To compare SV95C-in its first-ever clinical trial application as a secondary endpoint-with established motor function COAs used in the trial (Four-Stair Climb [4SC] velocity, North Star Ambulatory Assessment [NSAA], and Six-Minute Walk Distance [6MWD]). Methods: SV95C was a secondary endpoint in a subset (n = 47) of participants in the SPITFIRE/WN40227 trial of taldefgrobep alfa, which was discontinued due to lack of clinical benefit. Participants in the ≤48-week SV95C sub-study were 6-11 years old and received corticosteroids for ≥6 months pre-treatment. Pearson correlations were used to compare SV95C with the other COAs. Responsiveness and changes over time were respectively assessed via standardized response means (SRMs) based on absolute changes and mixed models for repeated measures. Results: SV95C change at Week 24 was -0.07 m/s, with limited variability (standard deviation: 0.16, n = 27). The SRM for SV95C indicated moderate responsiveness to clinical change at the earliest timepoint (Week 12, n = 46), while those of the other COAs did not indicate moderate responsiveness until Week 36 (6MWD, n = 33) or Week 48 (4SC velocity, n = 20; NSAA total score, n = 20). Baseline correlations between SV95C and other COAs were strong (r = 0.611-0.695). Correlations between SV95C change from baseline to Week 48 and changes in other COAs were moderate to strong (r = 0.443-0.678).∥. Conclusions: Overall, SV95C demonstrated sensitivity to ambulatory decline over short intervals, low variability, and correlation with established COAs. Although the negative trial precluded demonstration of SV95C's sensitivity to drug effect, these findings support the continued use of SV95C in DMD clinical trials.

The Difference in the Assessment of Knee Extension/Flexion Angles during Gait between Two Calibration Methods for Wearable Goniometer Sensors.

Frontal and axial knee motion can affect the accuracy of the knee extension/flexion motion measurement using a wearable goniometer. The purpose of this study was to test the hypothesis that calibrating the goniometer on an individual's body would reduce errors in knee flexion angle during gait, compared to bench calibration. Ten young adults (23.2 ± 1.3 years) were enrolled. Knee flexion angles during gait were simultaneously assessed using a wearable goniometer sensor and an optical three-dimensional motion analysis system, and the absolute error (AE) between the two methods was calculated. The mean AE across a gait cycle was 2.4° (0.5°) for the on-body calibration, and the AE was acceptable (<5°) throughout a gait cycle (range: 1.5-3.8°). The mean AE for the on-bench calibration was 4.9° (3.4°) (range: 1.9-13.6°). Statistical parametric mapping (SPM) analysis revealed that the AE of the on-body calibration was significantly smaller than that of the on-bench calibration during 67-82% of the gait cycle. The results indicated that the on-body calibration of a goniometer sensor had acceptable and better validity compared to the on-bench calibration, especially for the swing phase of gait.

Wearable smart contact lenses: A critical comparison of three physiological signals outputs for health monitoring.

Smart contact lenses (SCLs) have been considered as novel wearable devices for out-of-hospital and self-monitoring applications. They are capable of non-invasively and continuously monitoring physiological signals in the eyes, including vital biophysical (e.g., intraocular of pressure, temperature, and electrophysiological signal) and biochemical signals (e.g., pH, glucose, protein, nitrite, lactic acid, and ions). Recent progress mainly focuses on the rational design of wearable SCLs for physiological signal monitoring, while also facilitating the treatment of various ocular diseases. It covers contact lens materials, fabrication technologies, and integration methods. We also highlight and discuss a critical comparison of SCLs with electrical, microfluidic, and optical signal outputs in health monitoring. Their advantages and disadvantages could help researchers to make decisions when developing SCLs with desired properties for physiological signal monitoring. These unique capabilities make SCLs promising diagnostic and therapeutic tools. Despite the extensive research in SCLs, new technologies are still in their early stages of development and there are a few challenges to be addressed before these SCLs technologies can be successfully commercialized particularly in the form of rigorous clinical trials.

Age related disparities in sleep apnea diagnosis using a wearable device: Implications of 4% vs. 3% hypopnea scoring criteria.

STUDY OBJECTIVES: Obstructive sleep apnea (OSA) diagnosis relies on the Apnea-Hypopnea Index (AHI), with discrepancies arising from the 3% and 4% desaturation criteria. This study investigates age-related variations in OSA severity classification, utilizing data from 1201 adult patients undergoing Home Sleep Apnea Testing (HSAT) with SleepImage Ring@. METHODS: The study employs Bland-Altman analysis to compare AHI values obtained with the 3% and 4% desaturation criteria. Age-stratified analysis explores discrepancies across different age groups. RESULTS: The analysis reveals a systematic bias favoring the 3% criterion, impacting the quantification of apnea events. Age-specific patterns demonstrate diminishing agreement between criteria with increasing age. CONCLUSION: This comprehensive study underscores the importance of standardized criteria in OSA diagnosis. The findings emphasize age-specific considerations and ethical concerns, providing crucial insights for optimizing patient care and advancing sleep medicine practices.

A Low-Cost Wearable Device to Estimate Body Temperature Based on Wrist Temperature.

The remote monitoring of vital signs and healthcare provision has become an urgent necessity due to the impact of the COVID-19 pandemic on the world. Blood oxygen level, heart rate, and body temperature data are crucial for managing the disease and ensuring timely medical care. This study proposes a low-cost wearable device employing non-contact sensors to monitor, process, and visualize critical variables, focusing on body temperature measurement as a key health indicator. The wearable device developed offers a non-invasive and continuous method to gather wrist and forehead temperature data. However, since there is a discrepancy between wrist and actual forehead temperature, this study incorporates statistical methods and machine learning to estimate the core forehead temperature from the wrist. This research collects 2130 samples from 30 volunteers, and both the statistical least squares method and machine learning via linear regression are applied to analyze these data. It is observed that all models achieve a significant fit, but the third-degree polynomial model stands out in both approaches. It achieves an R2 value of 0.9769 in the statistical analysis and 0.9791 in machine learning.

Wearable mobile health device for monitoring postoperative ambulation among patients with colorectal cancer undergoing minimally invasive surgery: A prospective comparison study.

BACKGROUND: This study aimed to evaluate the feasibility of using mHealth devices for monitoring postoperative ambulation among patients with colorectal cancer undergoing minimally invasive surgery (MIS). METHODS: Patients with colorectal cancer undergoing MIS were prospectively recruited to wear mHealth devices for recording postoperative ambulation between October 2018 and January 2021. The primary outcome was the compliance by evaluating the weekly submission rate of step counts. The secondary outcome was the association of weekly step counts and postoperative length of stay. RESULTS: Of 107 eligible patients, 53 patients wore mHealth devices, whereas 54 patients did not. The average weekly submission rate was 72.6% for the first month after surgery. The total step counts <4000 or >10 000 in the postoperative week one were negatively associated with postoperative length of stay (ß = -2.874, p = 0.038). CONCLUSIONS: mHealth devices provide an objective assessment of postoperative ambulation among patients with colorectal cancer undergoing MIS. CLINICAL TRIAL REGISTRATION: NCT03277235.

Facile and cost-effective fabrication of wearable alpha-naphtholphthalein-based halochromic sensor for wound pH monitoring.

The sensor, designed to be worn directly on the skin, is suitable for real-time monitoring of the recovery level of not only general wounds, but also difficult-to-heal wounds, such as those with chronic inflammation. Notably, healthy skin has a pH range of 4-6. When a wound occurs, the pH is known to be approximately 7.4. In this study, alpha-naphtholphthalein (Naph) was immersed in a cotton-blended textile to produce a wearable halochromic sensor that clearly changed color depending on the pH of the skin in the range 6-9, including pH 7.4, which is the skin infection state. The coating was performed without using an organic solvent by dissolving it in micelle form using cetyltrimethylammonium bromide, a surfactant, in water. Naph-based halochromic sensor shows light yellow, which is the dye's own color, at pH 6, which is a healthy skin condition, and gradually showed a clear color change to light green-green-blue as pH increased. Even after washing and drying by rubbing with regular tap water, the color change due to pH was maintained more than 10 times. Naph-based halochromic sensors use a simple solution production and coating method and are not only reusable sensors that can be washed with water but also use environmentally friendly water, making them very suitable for developing commercial products for wound pH monitoring. In addition, it can be easily applied to medical supplies, such as medical gauze, patient clothes, and compression bandages, as well as everyday wear, such as clothing, gloves, and socks. Therefore, it is expected to be widely used as a wound pH sensor, allowing real-time monitoring of the skin condition of individuals with chronic skin inflammation, including patients requiring wound recovery.

Large-Scale Wearable Textile-Based Sweat Sensor with High Sensitivity, Rapid Response, and Stable Electrochemical Performance.

Textile-based sweat sensors display great potential to enhance wearable comfort and health monitoring; however, their widespread application is severely hindered by the intricate manufacturing process and electrochemical characteristics. To address this challenge, we combined both impregnation coating technology and conjugated electrospinning technology to develop an electro-assisted impregnation core-spinning technology (EAICST), which enables us to simply construct a sheath-core electrochemical sensing yarn (TPFV/CPP yarn) via coating PEDOT:PSS-coated carbon fibers (CPP) with polyurethane (TPU)/polyacrylonitrile (PAN)/poloxamer (F127)/valinomycin as shell. The TPFV/CPP yarn was sewn into the fabric and integrated with a sensor to achieve a detachable feature and efficiently monitor K+ levels in sweat. By introducing EAICST, a speed of 10 m/h can be realized in the continuous preparation of the TPFV/CPP yarn, while the interconnected pores in the yarn sheath enable it to quickly capture and diffuse sweat. Besides, the sensor exhibited excellent sensitivity (54.26 mV/decade), fast response (1.7 s), anti-interference, and long-term stability (5000 s or more). Especially, it also possesses favorable washability and wear resistance properties. Taken together, this study provides a crucial technical foundation for the development of advanced wearable devices designed for sweat analysis.

Validation of a Wearable Virtual Reality Perimeter for Glaucoma Staging, The NOVA Trial: Novel Virtual Reality Field Assessment.

Purpose: Compare estimated sensitivities of SITA-Standard to the RATA-Standard algorithm of the Radius virtual reality perimeter (VRP), and measure concordance in glaucoma staging. Methods: One hundred adult glaucoma patients-half with suspect or mild glaucoma, and half with moderate or severe-from five clinics performed four 24-2 visual field tests during a single visit, two with the Humphrey Field Analyzer (HFA) and two with Radius, in randomized order: HRHR or RHRH. Only one eye was tested per participant. We used the Wilcoxon rank sum test with Bonferroni correction to compare distributions of estimated sensitivities across all 54 test locations over the 15 to 40 dB measurement range of the Radius. Weighted kappa measured concordance in glaucoma staging between two masked glaucoma experts using Medicare definitions of severity. Results: A total of 62 OD and 38 OS eyes were tested. Estimated sensitivities for SITA-Standard and RATA-Standard were not significantly different for OD, but were for OS-likely because of SITA-Standard OD and OS being significantly different in our sample, but not for RATA-Standard. Low agreement was observed between 15 to 22 dB. Concordance in glaucoma staging was high for both graders: kappa = 0.91 and kappa = 0.93. Average test duration was 298 seconds for RATA-Standard and 341 seconds for SITA-Standard. The correlation in mean deviation was 0.94. Conclusions: Estimated sensitivities of RATA-Standard are comparable to SITA-Standard between 23 to 40 dB with high concordance in glaucoma staging. Translational Relevance: Radius VRP is statistically noninferior to HFA when staging glaucoma using Medicare definitions.

Advancements in optical fiber-based wearable sensors for smart health monitoring.

Healthcare system is undergoing a significant transformation from a traditional hospital-centered to an individual-centered one, as a result of escalating chronic diseases, ageing populations, and ever-increasing healthcare costs,. Wearable sensors have become widely used in health monitoring systems since the COVID-19 pandemic. They enable continuous measurement of important health indicators like body temperature, wrist pulse, respiration rate, and non-invasive bio fluids like saliva and perspiration. Over the last few decades, the development has mostly concentrated on electrochemical and electrical wearable sensors. However, due to the drawbacks of such sensors, such as electronic waste, electromagnetic interference, non-electrical security, and poor performance, researchers are exhibiting a strong interest in optical principle-based systems. Fiber-based optical wearables are among the most promising healthcare systems because of advancements in high-sensitivity, durable, multiplexed sensing, and simple integration with flexible materials to improve wearability and simplicity. We present an overview of recent developments in optical fiber-based wearable sensors, focusing on two mechanisms: wavelength interrogation and intensity modulation for the detection of body temperature, pulse rate, respiration rate, body movements, and biomedical noninvasive fluids, with a thorough examination of their benefits and drawbacks. This review also focuses on improving working performance and application techniques for healthcare systems, including the integration of nanomaterials and the usage of the Internet of Things (IoT) with signal processing. Finally, the review concludes with a discussion of the future possibilities and problems for optical fiber-based wearables.

Exploring wearable technology use and importance of health monitoring in the hazardous occupations of first responders and professional drivers.

OBJECTIVES: Hazardous materials (HAZMAT) pose risks to the health and safety of professionals involved with transportation and emergency responses. Two distinct occupational groups that encounter HAZMAT events are first responders and professional drivers. Wearable technology is a tool that can assist with monitoring the health of professionals involved in HAZMAT events. The aim of this study was to compare and evaluate the perceptions of first responders and professional drivers on wearable technology and attitudes toward health monitoring. METHODS: A survey was administered to first responders (n = 112) and professional drivers (n = 218). Statistical approaches included bivariate analysis, latent class analysis, logistic regression analysis, and path analysis for the variables of interest. RESULTS: There were significant differences between the groups in perceptions of the benefits of monitoring certain health indicators. Professional drivers were more likely to have a history of wearable technology use compared with first responders (odds ratio [OR] = 10.1; 95% CI, 4.42-22.9), reported greater exposure to HAZMAT (OR = 4.32; 95% CI, 2.24-8.32), and were more willing to have their health data monitored by someone other than themselves (OR = 9.27; 95% CI, 3.67-23.4). A multinomial regression model revealed that occupation was not a significant predictor of class preference for acceptance of monitoring specific health indicators. CONCLUSIONS: Occupation appeared to be important but further analysis uncovered that characteristics of individuals within the occupations were more salient to the use of wearable technology. HAZMAT exposure, someone else monitoring health data, and experience with wearable technology use were found to be important factors for perceptions about benefits of health monitoring with wearable technology.

Recent advances in smart wearable sensors for continuous human health monitoring.

In recent years, the biochemical and biological research areas have shown great interest in a smart wearable sensor because of its increasing prevalence and high potential to monitor human health in a non-invasive manner by continuous screening of biomarkers dispersed throughout the biological analytes, as well as real-time diagnostic tools and time-sensitive information compared to conventional hospital-centered system. These smart wearable sensors offer an innovative option for evaluating and investigating human health by incorporating a portion of recent advances in technology and engineering that can enhance real-time point-of-care-testing capabilities. Smart wearable sensors have emerged progressively with a mixture of multiplexed biosensing, microfluidic sampling, and data acquisition systems incorporated with flexible substrate and bodily attachments for enhanced wearability, portability, and reliability. There is a good chance that smart wearable sensors will be relevant to the early detection and diagnosis of disease management and control. Therefore, pioneering smart wearable sensors into reality seems extremely promising despite possible challenges in this cutting-edge technology for a better future in the healthcare domain. This review presents critical viewpoints on recent developments in wearable sensors in the upcoming smart digital health monitoring in real-time scenarios. In addition, there have been proactive discussions in recent years on materials selection, design optimization, efficient fabrication tools, and data processing units, as well as their continuous monitoring and tracking strategy with system-level integration such as internet-of-things, cyber-physical systems, and machine learning algorithms.

A 3D-printed hollow microneedle-based electrochemical sensing device for in situ plant health monitoring.

Plant health monitoring is devised as a new concept to elucidate in situ physiological processes. The need for increased food production to nourish the growing global population is inconsistent with the dramatic impact of climate change, which hinders crop health and exacerbates plant stress. In this context, wearable sensors play a crucial role in assessing plant stress. Herein, we present a low-cost 3D-printed hollow microneedle array (HMA) patch as a sampling device coupled with biosensors based on screen-printing technology, leading to affordable analysis of biomarkers in the plant fluid of a leaf. First, a refinement of the 3D-printing method showed a tip diameter of 25.9 ± 3.7 µm with a side hole diameter on the microneedle of 228.2 ± 18.6 µm using an affordable 3D printer (<500 EUR). Notably, the HMA patch withstanded the forces exerted by thumb pressing (i.e. 20-40 N). Subsequently, the holes of the HMA enabled the fluid extraction tested in vitro and in vivo in plant leaves (i.e. 13.5 ± 1.1 µL). A paper-based sampling strategy adapted to the HMA allowed the collection of plant fluid. Finally, integrating the sampling device onto biosensors facilitated the in situ electrochemical analysis of plant health biomarkers (i.e. H2O2, glucose, and pH) and the electrochemical profiling of plants in five plant species. Overall, this electrochemical platform advances precise and versatile sensors for plant health monitoring. The wearable device can potentially improve precision farming practices, addressing the critical need for sustainable and resilient agriculture in changing environmental conditions.

Investigating Receptivity and Affect Using Machine Learning: Ecological Momentary Assessment and Wearable Sensing Study.

BACKGROUND: As mobile health (mHealth) studies become increasingly productive owing to the advancements in wearable and mobile sensor technology, our ability to monitor and model human behavior will be constrained by participant receptivity. Many health constructs are dependent on subjective responses, and without such responses, researchers are left with little to no ground truth to accompany our ever-growing biobehavioral data. This issue can significantly impact the quality of a study, particularly for populations known to exhibit lower compliance rates. To address this challenge, researchers have proposed innovative approaches that use machine learning (ML) and sensor data to modify the timing and delivery of surveys. However, an overarching concern is the potential introduction of biases or unintended influences on participants' responses when implementing new survey delivery methods. OBJECTIVE: This study aims to demonstrate the potential impact of an ML-based ecological momentary assessment (EMA) delivery system (using receptivity as the predictor variable) on the participants' reported emotional state. We examine the factors that affect participants' receptivity to EMAs in a 10-day wearable and EMA-based emotional state-sensing mHealth study. We study the physiological relationships indicative of receptivity and affect while also analyzing the interaction between the 2 constructs. METHODS: We collected data from 45 healthy participants wearing 2 devices measuring electrodermal activity, accelerometer, electrocardiography, and skin temperature while answering 10 EMAs daily, containing questions about perceived mood. Owing to the nature of our constructs, we can only obtain ground truth measures for both affect and receptivity during responses. Therefore, we used unsupervised and supervised ML methods to infer affect when a participant did not respond. Our unsupervised method used k-means clustering to determine the relationship between physiology and receptivity and then inferred the emotional state during nonresponses. For the supervised learning method, we primarily used random forest and neural networks to predict the affect of unlabeled data points as well as receptivity. RESULTS: Our findings showed that using a receptivity model to trigger EMAs decreased the reported negative affect by >3 points or 0.29 SDs in our self-reported affect measure, scored between 13 and 91. The findings also showed a bimodal distribution of our predicted affect during nonresponses. This indicates that this system initiates EMAs more commonly during states of higher positive emotions. CONCLUSIONS: Our results showed a clear relationship between affect and receptivity. This relationship can affect the efficacy of an mHealth study, particularly those that use an ML algorithm to trigger EMAs. Therefore, we propose that future work should focus on a smart trigger that promotes EMA receptivity without influencing affect during sampled time points.

Wearable-Sensor-Based Weakly Supervised Parkinson's Disease Assessment with Data Augmentation.

Parkinson's disease (PD) is the second most prevalent dementia in the world. Wearable technology has been useful in the computer-aided diagnosis and long-term monitoring of PD in recent years. The fundamental issue remains how to assess the severity of PD using wearable devices in an efficient and accurate manner. However, in the real-world free-living environment, there are two difficult issues, poor annotation and class imbalance, both of which could potentially impede the automatic assessment of PD. To address these challenges, we propose a novel framework for assessing the severity of PD patient's in a free-living environment. Specifically, we use clustering methods to learn latent categories from the same activities, while latent Dirichlet allocation (LDA) topic models are utilized to capture latent features from multiple activities. Then, to mitigate the impact of data imbalance, we augment bag-level data while retaining key instance prototypes. To comprehensively demonstrate the efficacy of our proposed framework, we collected a dataset containing wearable-sensor signals from 83 individuals in real-life free-living conditions. The experimental results show that our framework achieves an astounding 73.48% accuracy in the fine-grained (normal, mild, moderate, severe) classification of PD severity based on hand movements. Overall, this study contributes to more accurate PD self-diagnosis in the wild, allowing doctors to provide remote drug intervention guidance.

Automatic assessment of infant carrying and holding using at-home wearable recordings.

Assessing infant carrying and holding (C/H), or physical infant-caregiver interaction, is important for a wide range of contexts in development research. An automated detection and quantification of infant C/H is particularly needed in long term at-home studies where development of infants' neurobehavior is measured using wearable devices. Here, we first developed a phenomenological categorization for physical infant-caregiver interactions to support five different definitions of C/H behaviors. Then, we trained and assessed deep learning-based classifiers for their automatic detection from multi-sensor wearable recordings that were originally used for mobile assessment of infants' motor development. Our results show that an automated C/H detection is feasible at few-second temporal accuracy. With the best C/H definition, the automated detector shows 96% accuracy and 0.56 kappa, which is slightly less than the video-based inter-rater agreement between trained human experts (98% accuracy, 0.77 kappa). The classifier performance varies with C/H definition reflecting the extent to which infants' movements are present in each C/H variant. A systematic benchmarking experiment shows that the widely used actigraphy-based method ignores the normally occurring C/H behaviors. Finally, we show proof-of-concept for the utility of the novel classifier in studying C/H behavior across infant development. Particularly, we show that matching the C/H detections to individuals' gross motor ability discloses novel insights to infant-parent interaction.

Longitudinal assessment of sweat-based TNF-alpha in inflammatory bowel disease using a wearable device.

Wearable devices can non-invasively monitor patients with chronic diseases. Sweat is an easily accessible biofluid for continuous sampling of analytes, including inflammatory markers and cytokines. We evaluated a sweat sensing wearable device in subjects with and without inflammatory bowel disease (IBD), a chronic inflammatory condition of the gastrointestinal tract. Participants with an IBD related hospital admission and a C-reactive protein level above 5 mg/L wore a sweat sensing wearable device for up to 5 days. Tumor necrosis factor-alpha (TNF-α) levels were continually assessed in the sweat via the sensor, and daily in the blood. A second cohort of healthy subjects without chronic diseases wore the device for up to 48 h. Twenty-eight subjects were enrolled. In the 16 subjects with IBD, a moderate linear relationship between serum and sweat TNF-α levels was observed (R2 = 0.72). Subjects with IBD were found to have a mean sweat TNF-α level of 2.11 pg/mL, compared to a mean value of 0.19 pg/mL in 12 healthy controls (p < 0.0001). Sweat TNF-α measurements differentiated subjects with active IBD from healthy subjects with an AUC of 0.962 (95% CI 0.894-1.000). A sweat sensing wearable device can longitudinally measure key sweat-based markers of IBD. TNF-α levels in the sweat of subjects with IBD correlate with serum values, suggesting feasibility in non-invasive disease monitoring.

Early Prediction of Poststroke Rehabilitation Outcomes Using Wearable Sensors.

OBJECTIVE: Inpatient rehabilitation represents a critical setting for stroke treatment, providing intensive, targeted therapy and task-specific practice to minimize a patient's functional deficits and facilitate their reintegration into the community. However, impairment and recovery vary greatly after stroke, making it difficult to predict a patient's future outcomes or response to treatment. In this study, the authors examined the value of early-stage wearable sensor data to predict 3 functional outcomes (ambulation, independence, and risk of falling) at rehabilitation discharge. METHODS: Fifty-five individuals undergoing inpatient stroke rehabilitation participated in this study. Supervised machine learning classifiers were retrospectively trained to predict discharge outcomes using data collected at hospital admission, including patient information, functional assessment scores, and inertial sensor data from the lower limbs during gait and/or balance tasks. Model performance was compared across different data combinations and was benchmarked against a traditional model trained without sensor data. RESULTS: For patients who were ambulatory at admission, sensor data improved the predictions of ambulation and risk of falling (with weighted F1 scores increasing by 19.6% and 23.4%, respectively) and maintained similar performance for predictions of independence, compared to a benchmark model without sensor data. The best-performing sensor-based models predicted discharge ambulation (community vs household), independence (high vs low), and risk of falling (normal vs high) with accuracies of 84.4%, 68.8%, and 65.9%, respectively. Most misclassifications occurred with admission or discharge scores near the classification boundary. For patients who were nonambulatory at admission, sensor data recorded during simple balance tasks did not offer predictive value over the benchmark models. CONCLUSION: These findings support the continued investigation of wearable sensors as an accessible, easy-to-use tool to predict the functional recovery after stroke. IMPACT: Accurate, early prediction of poststroke rehabilitation outcomes from wearable sensors would improve our ability to deliver personalized, effective care and discharge planning in the inpatient setting and beyond.