Whole-Body Movements Increase Arm Use Outcomes of Wrist-Worn Accelerometers in Stroke Patients

Factors associated with upper extremity use after stroke: a scoping review of accelerometry studies

BACKGROUND

A discrepancy between the level of impairment at the upper extremity (UE) and its use in activities of daily life is frequently observed in individuals who have experienced a stroke. Wrist-worn accelerometers allow an objective and valid measure of UE use in everyday life. Accelerometer studies have shown that a wide range of factors beyond UE impairment can influence UE use. This scoping review aims to identify factors associated with UE use and to investigate the influence of different types of accelerometry metrics on these associations.

METHOD

A search using CINHAL, Embase, MEDLINE, Compendex, and Web of Science Core Collection databases was performed. Studies that assessed the association between UE use quantified with accelerometers and factors related to the person or their environment in individuals with stroke were included. Data related to study design, participants characteristics, accelerometry methodology (absolute vs. relative UE use metrics), and associations with personal and environmental factors were extracted.

RESULTS

Fifty-four studies were included. Multiple studies consistently reported associations between relative UE use and stroke severity, UE motor impairment, unimanual capacity, bimanual capacity, and mobility. In contrast, there were inconsistent associations with factors such as neglect and concordance between dominance and side of paresis and a consistent lack of association between relative UE use and time since stroke, sex, and age. Metrics of absolute paretic UE use yielded different results regarding their association with personal and environmental factors, as they were more influenced by factors related to physical activity and less associated with factors related to UE capacity.

CONCLUSION

Healthcare providers should recognize the complexity of the relationship between UE use and impairment and consider additional factors when selecting assessments during rehabilitation to identify patients at risk of underutilizing their paretic arm in daily life. Future research in this domain should preconize relative UE use metrics or multi-sensors method to control for the effect of physical activity.

Construct validity and responsiveness of clinical upper limb measures and sensor-based arm use within the first year after stroke: a longitudinal cohort study

BACKGROUND

Construct validity and responsiveness of upper limb outcome measures are essential to interpret motor recovery poststroke. Evaluating the associations between clinical upper limb measures and sensor-based arm use (AU) fosters a coherent understanding of motor recovery. Defining sensor-based AU metrics for intentional upper limb movements could be crucial in mitigating bias from walking-related activities. Here, we investigate the measurement properties of a comprehensive set of clinical measures and sensor-based AU metrics when gait and non-functional upper limb movements are excluded.

METHODS

In this prospective, longitudinal cohort study, individuals with motor impairment were measured at days 3 ± 2 (D3), 10 ± 2 (D10), 28 ± 4 (D28), 90 ± 7 (D90), and 365 ± 14 (D365) after their first stroke. Using clinical measures, upper limb motor function (Fugl-Meyer Assessment), capacity (Action Research Arm Test, Box & Block Test), and perceived performance (14-item Motor Activity Log) were assessed. Additionally, individuals wore five movement sensors (trunk, wrists, and ankles) for three days. Thirteen AU metrics were computed based on functional movements during non-walking periods. Construct validity across clinical measures and AU metrics was determined by Spearman's rank correlations for each time point. Criterion responsiveness was examined by correlating patient-reported Global Rating of Perceived Change (GRPC) scores and observed change in upper limb measures and AU metrics. Optimal cut-off values for minimal important change (MIC) were estimated by ROC curve analysis.

RESULTS

Ninety-three individuals participated. At D3 and D10, correlations between clinical measures and AU metrics showed variability (range rs: 0.44-0.90). All following time points showed moderate-to-high positive correlations between clinical measures and affected AU metrics (range rs: 0.57-0.88). Unilateral nonaffected AU duration was negatively correlated with clinical measures (range rs: -0.48 to -0.77). Responsiveness across outcomes was highest between D10-D28 within moderate to strong relations between GRPC and clinical measures (rs: range 0.60-0.73), whereas relations were weaker for AU metrics (range rs: 0.28-0.43) Eight MIC values were estimated for clinical measures and nine for AU metrics, showing moderate to good accuracy (66-87%).

CONCLUSIONS

We present reference data on the construct validity and responsiveness of clinical upper limb measures and specified sensor-based AU metrics within the first year after stroke. The MIC values can be used as a benchmark for clinical stroke rehabilitation.

TRIAL REGISTRATION

This trial was registered on clinicaltrials.gov; registration number NCT03522519.

Near-zero stiffness accelerometer with buckling of tunable electrothermal microbeams

Pre-shaped microbeams, curved or inclined, are widely used in MEMS for their interesting stiffness properties. These mechanisms allow a wide range of positive and negative stiffness tuning in their direction of motion. A mechanism of pre-shaped beams with opposite curvature, connected in a parallel configuration, can be electrothermally tuned to reach a near-zero or negative stiffness behavior at the as-fabricated position. The simple structure helps incorporate the tunable spring mechanism in different designs for accelerometers, even with different transduction technologies. The sensitivity of the accelerometer can be considerably increased or tuned for different applications by electrothermally changing the stiffness of the spring mechanism. Opposite inclined beams are implemented in a capacitive micromachined accelerometer. The measurements on fabricated prototypes showed more than 55 times gain in sensitivity compared to their initial sensitivity. The experiments showed promising results in enhancing the resolution of acceleration sensing and the potential to reach unprecedent performance in micromachined accelerometers.

Optimizing Epoch Length and Activity Count Threshold Parameters in Accelerometry: Enhancing Upper Extremity Use Quantification in Cerebral Palsy

Various accelerometry protocols have been used to quantify upper extremity (UE) activity, encompassing diverse epoch lengths and thresholding methods. However, there is no consensus on the most effective approach. The aim of this study was to delineate the optimal parameters for analyzing accelerometry data to quantify UE use in individuals with unilateral cerebral palsy (CP).

METHODS

A group of adults with CP (n = 15) participated in six activities of daily living, while a group of children with CP (n = 14) underwent the Assisting Hand Assessment. Both groups performed the activities while wearing ActiGraph GT9X-BT devices on each wrist, with concurrent video recording. Use ratio (UR) derived from accelerometry and video analysis and accelerometer data were compared for different epoch lengths (1, 1.5, and 2 s) and activity count (AC) thresholds (between 2 and 150).

RESULTS

In adults, results are comparable across epoch lengths, with the best AC thresholds being ≥ 100. In children, results are similar across epoch lengths of 1 and 1.5 (optimal AC threshold = 50), while the optimal threshold is higher with an epoch length of 2 (AC = 75).

CONCLUSIONS

The combination of epoch length and AC thresholds should be chosen carefully as both influence the validity of the quantification of UE use.

The Application of Wearable Sensors and Machine Learning Algorithms in Rehabilitation Training: A Systematic Review

The integration of wearable sensor technology and machine learning algorithms has significantly transformed the field of intelligent medical rehabilitation. These innovative technologies enable the collection of valuable movement, muscle, or nerve data during the rehabilitation process, empowering medical professionals to evaluate patient recovery and predict disease development more efficiently. This systematic review aims to study the application of wearable sensor technology and machine learning algorithms in different disease rehabilitation training programs, obtain the best sensors and algorithms that meet different disease rehabilitation conditions, and provide ideas for future research and development. A total of 1490 studies were retrieved from two databases, the Web of Science and IEEE Xplore, and finally 32 articles were selected. In this review, the selected papers employ different wearable sensors and machine learning algorithms to address different disease rehabilitation problems. Our analysis focuses on the types of wearable sensors employed, the application of machine learning algorithms, and the approach to rehabilitation training for different medical conditions. It summarizes the usage of different sensors and compares different machine learning algorithms. It can be observed that the combination of these two technologies can optimize the disease rehabilitation process and provide more possibilities for future home rehabilitation scenarios. Finally, the present limitations and suggestions for future developments are presented in the study.

A Sensor-Based Feedback Device Stimulating Daily Life Upper Extremity Activity in Stroke Patients: A Feasibility Study

This study aims to evaluate the feasibility and explore the efficacy of the Arm Activity Tracker (AAT). The AAT is a device based on wrist-worn accelerometers that provides visual and tactile feedback to stimulate daily life upper extremity (UE) activity in stroke patients.

METHODS

A randomised, crossover within-subject study was conducted in sub-acute stroke patients admitted to a rehabilitation centre. Feasibility encompassed (1) adherence: the dropout rate and the number of participants with insufficient AAT data collection; (2) acceptance: the technology acceptance model (range: 7-112) and (3) usability: the system usability scale (range: 0-100). A two-way ANOVA was used to estimate the difference between the baseline, intervention and control conditions for (1) paretic UE activity and (2) UE activity ratio.

RESULTS

Seventeen stroke patients were included. A 29% dropout rate was observed, and two participants had insufficient data collection. Participants who adhered to the study reported good acceptance (median (IQR): 94 (77-111)) and usability (median (IQR): 77.5 (75-78.5)-). We found small to medium effect sizes favouring the intervention condition for paretic UE activity (η²G = 0.07, p = 0.04) and ratio (η²G = 0.11, p = 0.22).

CONCLUSION

Participants who adhered to the study showed good acceptance and usability of the AAT and increased paretic UE activity. Dropouts should be further evaluated, and a sufficiently powered trial should be performed to analyse efficacy.

Responsiveness and trajectory of changes in the rating of everyday arm-use in the community and home (REACH) scale over the first-year post-stroke

OBJECTIVE

To examine the trajectory of the Rating of Everyday Arm-use in the Community and Home (REACH) scores over the first-year post-stroke, determine if REACH scores are modified by baseline impairment level and explore the responsiveness of the REACH scale through hypothesis testing.

DESIGN

Consecutive sample longitudinal study.

SETTING

Participants were recruited from an acute stroke unit and followed up at three, six, and 12 months post-stroke.

PARTICIPANTS

Seventy-three participants with upper limb weakness (Shoulder Abduction and Finger Extension score ≤ 8).

MAIN MEASURES

The REACH scale is a six-level self-report classification scale that captures how the affected upper limb is being used in one's own environment. The Fugl-Meyer Upper Limb Assessment (FMA-UL), Stroke Upper Limb Capacity Scale (SULCS), accelerometer-based activity count ratio and Global Rating of Change Scale (GRCS) were used to capture upper limb impairment, capacity, and use.

RESULTS

The following proportions of participants improved at least one REACH level: 64% from baseline to three months, 37% from three to six months and 13% from six to 12 months post-stroke. The trajectory of REACH scores over time was associated with baseline impairment. Change in REACH had a moderate correlation to change in SULCS and the GRCS but not FMA-UL or the activity count ratio.

CONCLUSIONS

Results of hypothesis testing provide preliminary evidence of the responsiveness of the REACH scale. On average, individuals with severe impairment continued to show improvement in use over the first year, while those with mild/moderate impairment plateaued and a small proportion decreased in the early chronic phase.

Concurrent validity of machine learning-classified functional upper extremity use from accelerometry in chronic stroke

Objective: This study aims to investigate the validity of machine learning-derived amount of real-world functional upper extremity (UE) use in individuals with stroke. We hypothesized that machine learning classification of wrist-worn accelerometry will be as accurate as frame-by-frame video labeling (ground truth). A second objective was to validate the machine learning classification against measures of impairment, function, dexterity, and self-reported UE use. Design: Cross-sectional and convenience sampling. Setting: Outpatient rehabilitation. Participants: Individuals (>18 years) with neuroimaging-confirmed ischemic or hemorrhagic stroke >6-months prior (n = 31) with persistent impairment of the hemiparetic arm and upper extremity Fugl-Meyer (UEFM) score = 12-57. Methods: Participants wore an accelerometer on each arm and were video recorded while completing an "activity script" comprising activities and instrumental activities of daily living in a simulated apartment in outpatient rehabilitation. The video was annotated to determine the ground-truth amount of functional UE use. Main outcome measures: The amount of real-world UE use was estimated using a random forest classifier trained on the accelerometry data. UE motor function was measured with the Action Research Arm Test (ARAT), UEFM, and nine-hole peg test (9HPT). The amount of real-world UE use was measured using the Motor Activity Log (MAL). Results: The machine learning estimated use ratio was significantly correlated with the use ratio derived from video annotation, ARAT, UEFM, 9HPT, and to a lesser extent, MAL. Bland-Altman plots showed excellent agreement between use ratios calculated from video-annotated and machine-learning classification. Factor analysis showed that machine learning use ratios capture the same construct as ARAT, UEFM, 9HPT, and MAL and explain 83% of the variance in UE motor performance. Conclusion: Our machine learning approach provides a valid measure of functional UE use. The accuracy, validity, and small footprint of this machine learning approach makes it feasible for measurement of UE recovery in stroke rehabilitation trials.

Classification of functional and non-functional arm use by inertial measurement units in individuals with upper limb impairment after stroke

Background: Arm use metrics derived from wrist-mounted movement sensors are widely used to quantify the upper limb performance in real-life conditions of individuals with stroke throughout motor recovery. The calculation of real-world use metrics, such as arm use duration and laterality preferences, relies on accurately identifying functional movements. Hence, classifying upper limb activity into functional and non-functional classes is paramount. Acceleration thresholds are conventionally used to distinguish these classes. However, these methods are challenged by the high inter and intra-individual variability of movement patterns. In this study, we developed and validated a machine learning classifier for this task and compared it to methods using conventional and optimal thresholds. Methods: Individuals after stroke were video-recorded in their home environment performing semi-naturalistic daily tasks while wearing wrist-mounted inertial measurement units. Data were labeled frame-by-frame following the Taxonomy of Functional Upper Limb Motion definitions, excluding whole-body movements, and sequenced into 1-s epochs. Actigraph counts were computed, and an optimal threshold for functional movement was determined by receiver operating characteristic curve analyses on group and individual levels. A logistic regression classifier was trained on the same labels using time and frequency domain features. Performance measures were compared between all classification methods. Results: Video data (6.5 h) of 14 individuals with mild-to-severe upper limb impairment were labeled. Optimal activity count thresholds were ≥20.1 for the affected side and ≥38.6 for the unaffected side and showed high predictive power with an area under the curve (95% CI) of 0.88 (0.87,0.89) and 0.86 (0.85, 0.87), respectively. A classification accuracy of around 80% was equivalent to the optimal threshold and machine learning methods and outperformed the conventional threshold by ∼10%. Optimal thresholds and machine learning methods showed superior specificity (75-82%) to conventional thresholds (58-66%) across unilateral and bilateral activities. Conclusion: This work compares the validity of methods classifying stroke survivors' real-life arm activities measured by wrist-worn sensors excluding whole-body movements. The determined optimal thresholds and machine learning classifiers achieved an equivalent accuracy and higher specificity than conventional thresholds. Our open-sourced classifier or optimal thresholds should be used to specify the intensity and duration of arm use.

Accuracy of gait and posture classification using movement sensors in individuals with mobility impairment after stroke

Background: Stroke leads to motor impairment which reduces physical activity, negatively affects social participation, and increases the risk of secondary cardiovascular events. Continuous monitoring of physical activity with motion sensors is promising to allow the prescription of tailored treatments in a timely manner. Accurate classification of gait activities and body posture is necessary to extract actionable information for outcome measures from unstructured motion data. We here develop and validate a solution for various sensor configurations specifically for a stroke population.

Methods: Video and movement sensor data (locations: wrists, ankles, and chest) were collected from fourteen stroke survivors with motor impairment who performed real-life activities in their home environment. Video data were labeled for five classes of gait and body postures and three classes of transitions that served as ground truth. We trained support vector machine (SVM), logistic regression (LR), and k-nearest neighbor (kNN) models to identify gait bouts only or gait and posture. Model performance was assessed by the nested leave-one-subject-out protocol and compared across five different sensor placement configurations.

Results: Our method achieved very good performance when predicting real-life gait versus non-gait (Gait classification) with an accuracy between 85% and 93% across sensor configurations, using SVM and LR modeling. On the much more challenging task of discriminating between the body postures lying, sitting, and standing as well as walking, and stair ascent/descent (Gait and postures classification), our method achieves accuracies between 80% and 86% with at least one ankle and wrist sensor attached unilaterally. The Gait and postures classification performance between SVM and LR was equivalent but superior to kNN.

Conclusion: This work presents a comparison of performance when classifying Gait and body postures in post-stroke individuals with different sensor configurations, which provide options for subsequent outcome evaluation. We achieved accurate classification of gait and postures performed in a real-life setting by individuals with a wide range of motor impairments due to stroke. This validated classifier will hopefully prove a useful resource to researchers and clinicians in the increasingly important field of digital health in the form of remote movement monitoring using motion sensors.