Improving Accelerometry-Based Measurement of Functional Use of the Upper Extremity After Stroke: Machine Learning Versus Counts Threshold Method

A data-driven approach to detect upper limb functional use during daily life in breast cancer survivors using wrist-worn sensors

To gain insights into the impact of upper limb (UL) dysfunctions after breast cancer treatment, this study aimed to develop a temporal convolutional neural network (TCN) to detect functional daily UL use in breast cancer survivors using data from a wrist-worn accelerometer. A pre-existing dataset of 10 breast cancer survivors was used that contained raw 3-axis acceleration data and simultaneously recorded video data, captured during four daily life activities. The input of our TCN consists of a 3-axis acceleration sequence with a receptive field of 243 samples. The 4 ResNet TCN blocks perform dilated temporal convolutions with a kernel of size 3 and a dilation rate that increases by a factor of 3 after each iteration. Outcomes of interest were functional UL use (minutes) and percentage UL use. We found strong agreement between the video and predicted data for functional UL use (ICC = 0.975) and moderately strong agreement for %UL use (ICC = 0.794). The TCN model overestimated the functional UL use by 0.71 min and 3.06%. Model performance showed good accuracy, f1, and AUPRC scores (0.875, 0.909, 0.954, respectively). In conclusion, using wrist-worn accelerometer data, the TCN model effectively identified functional UL use in daily life among breast cancer survivors.

Prognostic factors for the development of upper limb dysfunctions after breast cancer: the UPLIFT-BC prospective longitudinal cohort study protocol

INTRODUCTION

Upper limb (UL) dysfunctions are highly prevalent in people after breast cancer and have a great impact on performing activities in daily living. To improve care, a more comprehensive understanding of the development and persistence of UL dysfunctions is needed. Therefore, the UPLIFT-BC study will primarily examine the prognostic value of different factors at the body functions and structures, environmental and personal level of the International Classification of Functioning, Disability and Health (ICF) framework at 1-month post-surgery for persisting UL dysfunctions at 6 months after finishing cancer treatment.

METHODS AND ANALYSIS

A prospective longitudinal cohort study, running from 1-week pre-surgery to 6 months post-local cancer treatment, is performed in a cohort of 250 women diagnosed with primary breast cancer. Different potentially prognostic factors to UL dysfunctions, covering body functions and structures, environmental and personal factors of the ICF, are assessed pre-surgically and at different time points post-surgery. The primary aim is to investigate the prognostic value of these factors at 1-month post-surgery for subjective UL function (ie, QuickDASH) at 6 months post-cancer treatment, that is, 6 months post-radiotherapy or post-surgery (T3), depending on the individuals' cancer treatment trajectory. In this, factors with relevant prognostic value pre-surgery are considered as well. Similar analyses are performed with an objective measure for UL function (ie, accelerometry) and a composite score of the combination of subjective and objective UL function. Second, in the subgroup of participants who receive radiotherapy, the prognostic value of the same factors is explored at 1-month post-radiotherapy and 6 months post-surgery. A forward stepwise selection strategy is used to obtain these multivariable prognostic models.

ETHICS AND DISSEMINATION

The study protocol was approved by the Ethics Committee of UZ/KU Leuven (reference number s66248). The results of this study will be published in peer-reviewed journals and will be presented at several research conferences.

TRIAL REGISTRATION NUMBER

NCT05297591.

Measuring functional hand use in children with unilateral cerebral palsy using accelerometry and machine learning

AIM

To investigate wearable sensors for measuring functional hand use in children with unilateral cerebral palsy (CP).

METHOD

Dual wrist-worn accelerometry data were collected from three females and seven males with unilateral CP (mean age = 10 years 2 months [SD 3 years]) while performing hand tasks during video-recorded play sessions. Video observers labelled instances of functional and non-functional hand use. Machine learning was compared to the conventional activity count approach for identifying unilateral hand movements as functional or non-functional. Correlation and agreement analyses compared the functional usage metrics derived from each method.

RESULTS

The best-performing machine learning approach had high precision and recall when trained on an individual basis (F1  = 0.896 [SD 0.043]). On an individual basis, the best-performing classifier showed a significant correlation (r = 0.990, p < 0.001) and strong agreement (bias = 0.57%, 95% confidence interval = -4.98 to 6.13) with video observations. When validated in a leave-one-subject-out scenario, performance decreased significantly (F1  = 0.584 [SD 0.076]). The activity count approach failed to detect significant differences in non-functional or functional hand activity and showed no significant correlation or agreement with the video observations.

INTERPRETATION

With further development, wearable accelerometry combined with machine learning may enable quantitative monitoring of everyday functional hand use in children with unilateral CP.

Assessing Upper Limb Function in Breast Cancer Survivors Using Wearable Sensors and Machine Learning in a Free-Living Environment

(1) Background: Being able to objectively assess upper limb (UL) dysfunction in breast cancer survivors (BCS) is an emerging issue. This study aims to determine the accuracy of a pre-trained lab-based machine learning model (MLM) to distinguish functional from non-functional arm movements in a home situation in BCS. (2) Methods: Participants performed four daily life activities while wearing two wrist accelerometers and being video recorded. To define UL functioning, video data were annotated and accelerometer data were analyzed using a counts threshold method and an MLM. Prediction accuracy, recall, sensitivity, f1-score, 'total minutes functional activity' and 'percentage functionally active' were considered. (3) Results: Despite a good MLM accuracy (0.77-0.90), recall, and specificity, the f1-score was poor. An overestimation of the 'total minutes functional activity' and 'percentage functionally active' was found by the MLM. Between the video-annotated data and the functional activity determined by the MLM, the mean differences were 0.14% and 0.10% for the left and right side, respectively. For the video-annotated data versus the counts threshold method, the mean differences were 0.27% and 0.24%, respectively. (4) Conclusions: An MLM is a better alternative than the counts threshold method for distinguishing functional from non-functional arm movements. However, the abovementioned wrist accelerometer-based assessment methods overestimate UL functional activity.

Quantitative Assessment of Upper Limb Movement in Post-Stroke Adults for Identification of Sensitive Measures in Reaching and Lifting Activities

BACKGROUND

The assumption of this work is the achievement of objective results of the movement structure, which forms the basis for in-depth analysis and, consequently, for determining the upper limb movements that are most affected by stroke compared to healthy people.

METHODS

An analysis of relevant and systematically identified features of upper limb movement in post-stroke adults is presented based on scalable hypothesis tests. The basic features were calculated using movements defined by the x, y, and z coordinates (i.e., 3D trajectory time series) and compared to the results of post-stroke patients with healthy controls of similar age.

RESULTS

After automatic feature selection, out of the 1004 common features of upper limb movement, the most differentiated were the upper arm movements in reaching kinematics. In terms of movement type, movements in the frontal plane (shoulder abduction and adduction) were the most sensitive to changes. The largest number of discriminating features was determined on the basis of acceleration time series.

CONCLUSIONS

In the 3D assessment of functional activities of the upper limb, the upper arm turned out to be the most differentiated body segment, especially during abduction and adduction movements. The results indicate a special need to pay attention to abduction and adduction movements to improve the activities of daily living of the upper limbs after a stroke.

Comparison of wrist actimetry variables of paretic upper limb use in post stroke patients for ecological monitoring

BACKGROUND

To date, many wrist actimetric variables dedicated to measuring the upper limbs (UL) in post-stroke patients have been developed but very few comparisons have been made between them. The objective of this study was to compare different actimetric variables of the ULs between a stroke and healthy population.

METHODS

Accelerometers were worn continuously for a period of 7 days on both wrists of 19 post-stroke hemiparetic patients as well as 11 healthy subjects. Various wrist actimetry variables were calculated, including the Jerk ratio 50 (JR50, cumulative probability that the Jerk Ratio is between 1 and 2), absolute (FuncUse30) and relative (FuncUseRatio30) amounts of functional use of movements of the ULs with angular amplitude greater than 30°, and absolute (UH) and relative (UseHoursRatio) use hours.

RESULTS

FuncUse30, FuncUseRatio30, UH, UseHoursRatio and JR50 of the paretic UL of stroke patients were significantly lower than in the non-dominant UL of healthy subjects. Comparing the ratio variables in stroke patients, FuncUseRatio30 was significantly lower than UseHoursRatio and JR50, suggesting a more clinically sensitive variable to monitor. In an exploratory analysis, FuncUseRatio tends to decrease with angular range of motion for stroke patients while it remains stable and close to 1 for healthy subjects. UseHoursRatio, FuncUseRatio30 and JR50 show linear correlation with Fugl-Meyer score (FM), with r² equal to 0.53, 0.35 and 0.21, respectively.

CONCLUSION

This study determined that the FuncUseRatio30 variable provides the most sensitive clinical biomarker of paretic UL use in post-stroke patients, and that FuncUseHours-angular range of motion relationship allows the identification of the UL behaviour of each patient. This ecological information on the level of functional use of the paretic UL can be used to improve follow-up and develop patient-specific therapy.

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.

Measurement of Functional Use in Upper Extremity Prosthetic Devices Using Wearable Sensors and Machine Learning

Trials for therapies after an upper limb amputation (ULA) require a focus on the real-world use of the upper limb prosthesis. In this paper, we extend a novel method for identifying upper extremity functional and nonfunctional use to a new patient population: upper limb amputees. We videotaped five amputees and 10 controls performing a series of minimally structured activities while wearing sensors on both wrists that measured linear acceleration and angular velocity. The video data was annotated to provide ground truth for annotating the sensor data. Two different analysis methods were used: one that used fixed-size data chunks to create features to train a Random Forest classifier and one that used variable-size data chunks. For the amputees, the fixed-size data chunk method yielded good results, with 82.7% median accuracy (range of 79.3-85.8) on the 10-fold cross-validation intra-subject test and 69.8% in the leave-one-out inter-subject test (range of 61.4-72.8). The variable-size data method did not improve classifier accuracy compared to the fixed-size method. Our method shows promise for inexpensive and objective quantification of functional upper extremity (UE) use in amputees and furthers the case for use of this method in assessing the impact of UE rehabilitative treatments.

Validity of Novel Outcome Measures for Hand Function Performance After Stroke Using Egocentric Video

BACKGROUND

Evaluating upper limb (UL) interventions after stroke calls for outcome measures that describe impact on daily life in the community. UL use ratio has been used to quantify the performance domain of UL function, but generally focuses on arm use only. A hand use ratio could provide additional information about UL function after stroke. Additionally, a ratio based on the role of the more-affected hand in bilateral activities (stabilizer or manipulator) may also reflect hand function recovery. Egocentric video is a novel modality that can record both dynamic and static hand use and hand roles at home after stroke.

OBJECTIVE

To validate hand use and hand role ratios from egocentric video against standardized clinical UL assessments.

METHODS

Twenty-four stroke survivors recorded daily tasks in a home simulation laboratory and their daily routines at home using egocentric cameras. Spearman's correlation was used to compare the ratios with the Fugl-Meyer Assessment-Upper Extremity (FMA-UE), Action Research Arm Test (ARAT), and Motor Activity Log-30 (MAL, Amount of Use (AoU), and Quality of Movement (QoM)).

RESULTS

Hand use ratio significantly correlated with the FMA-UE (0.60, 95% CI: 0.26, 0.81), ARAT (0.44, CI: 0.04, 0.72), MAL-AoU (0.80, CI: 0.59, 0.91), and MAL-QoM (0.79, CI: 0.57, 0.91). Hand role ratio had no significant correlations with the assessments.

CONCLUSION

Hand use ratio automatically extracted from egocentric video, but not hand role ratio, was found to be a valid measure of hand function performance in our sample. Further investigation is necessary to interpret hand role information.

Comparing algorithms for assessing upper limb use with inertial measurement units

The various existing measures to quantify upper limb use from wrist-worn inertial measurement units can be grouped into three categories: 1) Thresholded activity counting, 2) Gross movement score and 3) machine learning. However, there is currently no direct comparison of all these measures on a single dataset. While machine learning is a promising approach to detecting upper limb use, there is currently no knowledge of the information used by machine learning measures and the data-related factors that influence their performance. The current study conducted a direct comparison of the 1) thresholded activity counting measures, 2) gross movement score,3) a hybrid activity counting and gross movement score measure (introduced in this study), and 4) machine learning measures for detecting upper-limb use, using previously collected data. Two additional analyses were also performed to understand the nature of the information used by machine learning measures and the influence of data on the performance of machine learning measures. The intra-subject random forest machine learning measure detected upper limb use more accurately than all other measures, confirming previous observations in the literature. Among the non-machine learning (or traditional) algorithms, the hybrid activity counting and gross movement score measure performed better than the other measures. Further analysis of the random forest measure revealed that this measure used information about the forearm's orientation and amount of movement to detect upper limb use. The performance of machine learning measures was influenced by the types of movements and the proportion of functional data in the training/testing datasets. The study outcomes show that machine learning measures perform better than traditional measures and shed some light on how these methods detect upper-limb use. However, in the absence of annotated data for training machine learning measures, the hybrid activity counting and gross movement score measure presents a reasonable alternative. We believe this paper presents a step towards understanding and optimizing measures for upper limb use assessment using wearable sensors.

Measuring Hand Use in the Home after Cervical Spinal Cord Injury Using Egocentric Video

Egocentric video has recently emerged as a potential solution for monitoring hand function in individuals living with tetraplegia in the community, especially for its ability to detect functional use in the home environment. The aim of this study was to develop and validate a wearable vision-based system for measuring hand use in the home among individuals living with tetraplegia. Several deep learning algorithms for detecting functional hand-object interactions were developed and compared. The most accurate algorithm was used to extract measures of hand function from 65 h of unscripted video recorded at home by 20 participants with tetraplegia. These measures were: the percentage of interaction time over total recording time (Perc); the average duration of individual interactions (Dur); and the number of interactions per hour (Num). To demonstrate the clinical validity of the technology, egocentric measures were correlated with validated clinical assessments of hand function and independence (Graded Redefined Assessment of Strength, Sensibility and Prehension [GRASSP], Upper Extremity Motor Score [UEMS], and Spinal Cord Independent Measure [SCIM]). Hand-object interactions were automatically detected with a median F1-score of 0.80 (0.67-0.87). Our results demonstrated that higher UEMS and better prehension were related to greater time spent interacting, whereas higher SCIM and better hand sensation resulted in a higher number of interactions performed during the egocentric video recordings. For the first time, measures of hand function automatically estimated in an unconstrained environment in individuals with tetraplegia have been validated against internationally accepted measures of hand function. Future work will necessitate a formal evaluation of the reliability and responsiveness of the egocentric-based performance measures for hand use.

Classification Models of Action Research Arm Test Activities in Post-Stroke Patients Based on Human Hand Motion

The Action Research Arm Test (ARAT) presents a ceiling effect that prevents the detection of improvements produced with rehabilitation treatments in stroke patients with mild finger joint impairments. The aim of this study was to develop classification models to predict whether activities with similar ARAT scores were performed by a healthy subject or by a subject post-stroke using the extension and flexion angles of 11 finger joints as features. For this purpose, we used three algorithms: Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN). The dataset presented class imbalance, and the classification models presented a low recall, especially in the stroke class. Therefore, we implemented class balance using Borderline-SMOTE. After data balancing the classification models showed significantly higher accuracy, recall, f1-score, and AUC. However, after data balancing, the SVM classifier showed a higher performance with a precision of 98%, a recall of 97.5%, and an AUC of 0.996. The results showed that classification models based on human hand motion features in combination with the oversampling algorithm Borderline-SMOTE achieve higher performance. Furthermore, our study suggests that there are differences in ARAT activities performed between healthy and post-stroke individuals that are not detected by the ARAT scoring process.

The use of wearable sensors to assess and treat the upper extremity after stroke: a scoping review

PURPOSE

To address the gap in the literature and clarify the expanding role of wearable sensor data in stroke rehabilitation, we summarized the methods for upper extremity (UE) sensor-based assessment and sensor-based treatment.

MATERIALS AND METHODS

The guideline outlined by the preferred reporting items for systematic reviews and meta-analysis extension for scoping reviews was used to complete this scoping review. Information pertaining to participant demographics, sensory information, data collection, data processing, data analysis, and study results were extracted from the studies for analysis and synthesis.

RESULTS

We included 43 articles in the final review. We organized the results into assessment and treatment categories. The included articles used wearable sensors to identify UE functional motion, categorize motor impairment/activity limitation, and quantify real-world use. Wearable sensors were also used to augment UE training by triggering sensory cues or providing instructional feedback about the affected UE.

CONCLUSIONS

Sensors have the potential to greatly expand assessment and treatment beyond traditional clinic-based approaches. This capability could support the quantification of rehabilitation dose, the nuanced assessment of impairment and activity limitation, the characterization of daily UE use patterns in real-world settings, and augment UE training adherence for home-based rehabilitation.IMPLICATIONS FOR REHABILITATIONSensor data have been used to assess UE functional motion, motor impairment/activity limitation, and real-world use.Sensor-assisted treatment approaches are emerging, and may be a promising tool to augment UE adherence in home-based rehabilitation.Wearable sensors may extend our ability to objectively assess UE motion beyond supervised clinical settings, and into home and community settings.

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.

Monitoring Arm Movements Post-Stroke for Applications in Rehabilitation and Home Settings

Optimal recovery of arm function following stroke requires patients to perform a large number of functional arm movements in clinical therapy sessions, as well as at home. Technology to monitor adherence to this activity would be helpful to patients and clinicians. Current approaches to monitoring arm movements are limited because of challenges in distinguishing between functional and non-functional movements. Here, we present an Arm Rehabilitation Monitor (ARM), a device intended to make such measurements in an unobtrusive manner. The ARM device is based on a single Inertial Measurement Unit (IMU) worn on the wrist and uses machine learning techniques to interpret the resulting signals. We characterized the ability of the ARM to detect reaching actions in a functional assessment dataset (functional assessment tasks) and an Activities-of-Daily-Living (ADL) dataset (pizza-making and walking task) from 12 participants with stroke. The Convolutional Neural Network (CNN) and Random Forests (RF) classifiers had a Matthews Correlation Coefficient score of 0.59 and 0.58 when trained and tested on the functional dataset, 0.50 and 0.49 when trained and tested on the ADL dataset, and 0.37 and 0.36 when trained on the functional dataset and tested on the ADL dataset, respectively. The latter is the most relevant scenario for the intended application of training during a clinical visit for monitoring movements in the in-home setting. The classifiers showed good performance in estimating the time spent reaching and number of reaching gestures and showed low sensitivity to irrelevant arm movements produced during walking. We conclude that the ARM has sufficient accuracy and robustness to merit being used in preliminary studies to monitor arm activity in rehabilitation or home applications.

Machine Learning Improves Functional Upper Extremity Use Capture in Distal Radius Fracture Patients

Current outcome measures, including strength/range of motion testing, patient-reported outcomes (PROs), and motor skill testing, may provide inadequate granularity in reflecting functional upper extremity (UE) use after distal radius fracture (DRF) repair. Accelerometry analysis also has shortcomings, namely, an inability to differentiate functional versus nonfunctional movements. The objective of this study was to evaluate the accuracy of machine learning (ML) analyses in capturing UE functional movements based on accelerometry data for patients after DRF repair. In this prospective study, six patients were enrolled 2-6 weeks after DRF open reduction and internal fixation (ORIF). They all performed standardized activities while wearing a wrist accelerometer, and the data were analyzed by an ML algorithm. These activities were also videotaped and evaluated by visual inspection. Our novel ML algorithm was able to predict from accelerometry data whether the limb was performing a movement rated as functional, with accuracy of 90.4% ± 3.6% for within-subject modeling and 79.8% ± 8.9% accuracy for between-subject modeling. The application of ML algorithms to accelerometry data allowed for capture of functional UE activity in patients after DRF open reduction and internal fixation and accurately predicts functional UE use. Such analyses could improve our understanding of recovery and enhance routine postoperative rehabilitation in DRF patients.

PrimSeq: A deep learning-based pipeline to quantitate rehabilitation training

Stroke rehabilitation seeks to accelerate motor recovery by training functional activities, but may have minimal impact because of insufficient training doses. In animals, training hundreds of functional motions in the first weeks after stroke can substantially boost upper extremity recovery. The optimal quantity of functional motions to boost recovery in humans is currently unknown, however, because no practical tools exist to measure them during rehabilitation training. Here, we present PrimSeq, a pipeline to classify and count functional motions trained in stroke rehabilitation. Our approach integrates wearable sensors to capture upper-body motion, a deep learning model to predict motion sequences, and an algorithm to tally motions. The trained model accurately decomposes rehabilitation activities into elemental functional motions, outperforming competitive machine learning methods. PrimSeq furthermore quantifies these motions at a fraction of the time and labor costs of human experts. We demonstrate the capabilities of PrimSeq in previously unseen stroke patients with a range of upper extremity motor impairment. We expect that our methodological advances will support the rigorous measurement required for quantitative dosing trials in stroke rehabilitation.

Stroke commonly damages motor function in the upper extremity (UE), leading to long-term disability and loss of independence in a majority of individuals. Rehabilitation seeks to restore function by training daily activities, which deliver repeated UE functional motions. The optimal number of functional motions necessary to boost recovery is unknown. This gap stems from the lack of measurement tools to feasibly count functional motions. We thus developed the PrimSeq pipeline to enable the accurate and rapid counting of building-block functional motions, called primitives. PrimSeq uses wearable sensors to capture rich motion information from the upper body, and custom-built algorithms to detect and count functional primitives in this motion data. We showed that our deep learning algorithm precisely counts functional primitives performed by stroke patients and outperformed other benchmark algorithms. We also showed patients tolerated the wearable sensors and that the approach is 366 times faster at counting primitives than humans. PrimSeq thus provides a precise and practical means of quantifying functional primitives, which promises to advance stroke research and clinical care and to improve the outcomes of individuals with stroke.

Pilot Test of Dosage Effects in HEXORR II for Robotic Hand Movement Therapy in Individuals With Chronic Stroke

Impaired use of the hand in functional tasks remains difficult to overcome in many individuals after a stroke. This often leads to compensation strategies using the less-affected limb, which allows for independence in some aspects of daily activities. However, recovery of hand function remains an important therapeutic goal of many individuals, and is often resistant to conventional therapies. In prior work, we developed HEXORR I, a robotic device that allows practice of finger and thumb movements with robotic assistance. In this study, we describe modifications to the device, now called HEXORR II, and a clinical trial in individuals with chronic stroke. Fifteen individuals with a diagnosis of chronic stroke were randomized to 12 or 24 sessions of robotic therapy. The sessions involved playing several video games using thumb and finger movement. The robot applied assistance to extension movement that was adapted based on task performance. Clinical and motion capture evaluations were performed before and after training and again at a 6-month followup. Fourteen individuals completed the protocol. Fugl-Meyer scores improved significantly at the 6 month time point compared to baseline, indicating reductions in upper extremity impairment. Flexor hypertonia (Modified Ashworth Scale) also decreased significantly due to the intervention. Motion capture found increased finger range of motion and extension ability after the intervention that continued to improve during the followup period. However, there was no change in a functional measure (Action Research Arm Test). At the followup, the high dose group had significant gains in hand displacement during a forward reach task. There were no other significant differences between groups. Future work with HEXORR II should focus on integrating it with functional task practice and incorporating grip and squeezing tasks.

Trial Registration:ClinicalTrials.gov, NCT04536987. Registered 3 September 2020 - Retrospectively registered, https://clinicaltrials.gov/ct2/show/record/NCT04536987.

Prediction of Upper Limb use Three Months after Stroke: A Prospective Longitudinal Study

BACKGROUND

A major goal of upper limb (UL) rehabilitation after stroke is to facilitate the use of the paretic arm in daily life activities.

PURPOSE

To examine if UL impairment two weeks after stroke can predict real-life UL use at three months. Furthermore, to identify additional factors which contribute to future UL use, and characteristics of patients who do not achieve normal UL use.

METHODS

This study included patients with stroke ≥ 18 years. UL impairment was assessed by Fugl-Meyer upper extremity motor assessment (FM). Use ratio between affected and unaffected UL was assessed with accelerometers at three months after stroke. The association between FM score and UL use ratio was investigated with linear regression models and adjusted for secondary variables. Non-normal use was examined by a logistic regression.

RESULTS

Eighty-seven patients were included. FM score two weeks after stroke predicted 38% of the variance in UL use ratio three months after stroke. A multivariate regression model predicted 55%, and the significant predictors were FM, motor-evoked potential (MEP) status, and neglect. Non-normal use could be predicted with a high accuracy based on MEP and/or neglect. In a logistic regression sensitivity for prediction of non-normal use was 0.93 and specificity was 0.75.

CONCLUSION

Better baseline capacity of the paretic UL predicted increased use of the arm and hand in daily life. Non-normal UL use could be predicted reliably based on the absence of MEPs and/or presence of neglect.

A Framework for Sensor-Based Assessment of Upper-Limb Functioning in Hemiparesis

The ultimate goal of any upper-limb neurorehabilitation procedure is to improve upper-limb functioning in daily life. While clinic-based assessments provide an assessment of what a patient can do, they do not completely reflect what a patient does in his/her daily life. The use of compensatory strategies such as the use of the less affected upper-limb or excessive use of trunk in daily life is a common behavioral pattern seen in patients with hemiparesis. To this end, there has been an increasing interest in the use of wearable sensors to objectively assess upper-limb functioning. This paper presents a framework for assessing upper-limb functioning using sensors by providing: (a) a set of definitions of important constructs associated with upper-limb functioning; (b) different visualization methods for evaluating upper-limb functioning; and (c) two new measures for quantifying how much an upper-limb is used and the relative bias in their use. The demonstration of some of these components is presented using data collected from inertial measurement units from a previous study. The proposed framework can help guide the future technical and clinical work in this area to realize valid, objective, and robust tools for assessing upper-limb functioning. This will in turn drive the refinement and standardization of the assessment of upper-limb functioning.

Quantification of the relative arm use in patients with hemiparesis using inertial measurement units

Introduction

Accelerometry-based activity counting for measuring arm use is prone to overestimation due to non-functional movements. In this paper, we used an inertial measurement unit (IMU)-based gross movement (GM) score to quantify arm use.

Methods

In this two-part study, we first characterized the GM by comparing it to annotated video recordings of 5 hemiparetic patients and 10 control subjects performing a set of activities. In the second part, we tracked the arm use of 5 patients and 5 controls using two wrist-worn IMUs for 7 and 3 days, respectively. The IMU data was used to develop quantitative measures (total and relative arm use) and a visualization method for arm use.

Results

From the characterization study, we found that GM detects functional activities with 50–60% accuracy and eliminates non-functional activities with >90% accuracy. Continuous monitoring of arm use showed that the arm use was biased towards the dominant limb and less paretic limb for controls and patients, respectively.

Conclusions

The gross movement score has good specificity but low sensitivity in identifying functional activity. The at-home study showed that it is feasible to use two IMU-watches to monitor relative arm use and provided design considerations for improving the assessment method.

Clinical trial registry number: CTRI/2018/09/015648

Perspectives and recommendations of individuals with tetraplegia regarding wearable cameras for monitoring hand function at home: Insights from a community-based study

CONTEXT

Wearable cameras have great potential for producing novel outcome measures of upper limb (UL) function and guiding care in individuals with cervical spinal cord injury (cSCI) living in the community. However, little is known about the perspectives of individuals with cSCI on the potential adoption of this technology.

OBJECTIVE

To analyze feedback from individuals with cSCI regarding the use of wearable cameras to record daily activities at home, in order to define guidelines for improving the design of this technology and fostering its implementation to optimize UL rehabilitation.

DESIGN

Mixed-methods study.

PARTICIPANTS

Thirteen adults with cSCI C3-C8 AIS A-D impairment.

MEASURES

Interview including survey and semi-structured questions.

RESULTS

Participants felt that this technology can provide naturalistic information regarding hand use to clinicians and researchers, which in turn can lead to better assessments of UL function and optimized therapies. Participants described the technology as easy-to-use but often reported discomfort that prevented them from conducting long recordings of fully natural activities. Privacy concerns included the possibility to capture household members and personal information displayed on objects (e.g. smartphones).

CONCLUSION

We provide the first set of guidelines to help researchers and therapists understand which steps need to be taken to translate wearable cameras into outpatient care and community-based research for UL rehabilitation. These guidelines include miniaturized and easy-to-wear cameras, as well as multiple measures for preventing privacy concerns such as avoiding public spaces and providing control over the recordings (e.g. start and stop the recordings at any time, keep or delete a recording).

Automatic Identification of Upper Extremity Rehabilitation Exercise Type and Dose Using Body-Worn Sensors and Machine Learning: A Pilot Study

BACKGROUND

Prior studies suggest that participation in rehabilitation exercises improves motor function poststroke; however, studies on optimal exercise dose and timing have been limited by the technical challenge of quantifying exercise activities over multiple days.

OBJECTIVES

The objectives of this study were to assess the feasibility of using body-worn sensors to track rehabilitation exercises in the inpatient setting and investigate which recording parameters and data analysis strategies are sufficient for accurately identifying and counting exercise repetitions.

METHODS

MC10 BioStampRC® sensors were used to measure accelerometer and gyroscope data from upper extremities of healthy controls (n = 13) and individuals with upper extremity weakness due to recent stroke (n = 13) while the subjects performed 3 preselected arm exercises. Sensor data were then labeled by exercise type and this labeled data set was used to train a machine learning classification algorithm for identifying exercise type. The machine learning algorithm and a peak-finding algorithm were used to count exercise repetitions in non-labeled data sets.

RESULTS

We achieved a repetition counting accuracy of 95.6% overall, and 95.0% in patients with upper extremity weakness due to stroke when using both accelerometer and gyroscope data. Accuracy was decreased when using fewer sensors or using accelerometer data alone.

CONCLUSIONS

Our exploratory study suggests that body-worn sensor systems are technically feasible, well tolerated in subjects with recent stroke, and may ultimately be useful for developing a system to measure total exercise "dose" in poststroke patients during clinical rehabilitation or clinical trials.