Upper limb joint angle tracking with inertial sensors

Motion Tracking Algorithms Based on Wearable Inertial Sensor: A Focus on Shoulder

Shoulder Range of Motion (ROM) has been studied with several devices and methods in recent years. Accurate tracking and assessment of shoulder movements could help us to understand the pathogenetic mechanism of specific conditions in quantifying the improvements after rehabilitation. The assessment methods can be classified as subjective and objective. However, self-reported methods are not accurate, and they do not allow the collection of specific information. Therefore, developing measurement devices that provide quantitative and objective data on shoulder function and range of motion is important. A comprehensive search of PubMed and IEEE Xplore was conducted. The sensor fusion algorithm used to analyze shoulder kinematics was described in all studies involving wearable inertial sensors. Eleven articles were included. The Quality Assessment of Diagnostic Accuracy Studies-2 was used to assess the risk of bias (QUADAS-2). The finding showed that the Kalman filter and its variants UKF and EKF are used in the majority of studies. Alternatives based on complementary filters and gradient descent algorithms have been reported as being more computationally efficient. Many approaches and algorithms have been developed to solve this problem. It is useful to fuse data from different sensors to obtain a more accurate estimation of the 3D position and 3D orientation of a body segment. The sensor fusion technique makes this integration reliable. This systematic review aims to redact an overview of the literature on the sensor fusion algorithms used for shoulder motion tracking.

Validation of the Perception Neuron system for full-body motion capture

Recent advancements in Inertial Measurement Units (IMUs) offers the possibility of its use as a cost effective and portable alternative to traditional optoelectronic motion capture systems in analyzing biomechanical performance. One such commercially available IMU is the Perception Neuron motion capture system (PNS). The accuracy of the PNS had been tested and was reported to be a valid method for assessing the upper body range of motion to within 5° RMSE. However, testing of the PNS was limited to upper body motion involving functional movement within a single plane. Therefore, the purpose of this study is to further validate the Perception Neuron system with reference to a conventional optoelectronic motion capture system (VICON) through the use of dynamic movements (e.g., walking, jogging and a multi-articular sports movement with object manipulation) and to determine its feasibility through full-body kinematic analysis. Validation was evaluated using Pearson’s R correlation, RMSE and Bland-Altman estimates. Present findings suggest that the PNS performed well against the VICON motion analysis system with most joint angles reporting a RMSE of < 4° and strong average Pearson’s R correlation of 0.85, with the exception of the shoulder abduction/adduction where RMSE was larger and Pearson’s R correlation at a moderate level. Bland-Altman analysis revealed that most joint angles across the different movements had a mean bias of less than 10°, except for the shoulder abduction/adduction and elbow flexion/extension measurements. It was concluded that the PNS may not be the best substitute for traditional motion analysis technology if there is a need to replicate raw joint angles. However, there was adequate sensitivity to measure changes in joint angles and would be suitable when normalized joint angles are compared and the focus of analysis is to identify changes in movement patterns.

Analysis of Gait Sub-Movements to Estimate Ataxia Severity using Ankle Inertial Data

Objective: Assessment of motor severity in cerebellar ataxia is critical for monitoring disease progression and evaluating the effectiveness of therapeutic interventions. Though wearable sensors have been used to monitor gait tasks in order to enable frequent assessment, existing solutions only estimate gait performance severity rather than comprehensive motor severity. In this study, we propose a new approach that analyzes sub-second movement profiles of the lower-limbs during gait to estimate overall motor severity in cerebellar ataxia. Methods: A total of 37 ataxia subjects and 12 healthy subjects performed a 5 m walk-and-turn task with two ankle-worn inertial sensors. Lower-limb movements were decomposed into one-dimensional sub-movements, namely movement elements. Supervised regression models trained on data features of movement elements estimated the Brief Ataxia Rating Scale (BARS) and its sub-scores evaluated by clinicians. The proposed models were also compared to models trained on widely-accepted spatiotemporal gait features. Results: Estimated total BARS showed strong agreement with clinician-evaluated scores with r² = 0.72 and a root mean square error of 2.6 BARS points. Movement element-based models significantly outperformed conventional, spatiotemporal gait feature-based models. Conclusion: The proposed algorithm accurately assessed overall motor severity in cerebellar ataxia using inertial data collected from bilaterally-placed ankle sensors during a simple walk-and-turn task. Significance: Our work could support fine-grained monitoring of disease progression and patients' responses to medical/clinical interventions.

Wrist-Based Accelerometers and Visual Analog Scales as Outcome Measures for Shoulder Activity During Daily Living in Patients With Rotator Cuff Tendinopathy: Instrument Validation Study

BACKGROUND

Shoulder pain secondary to rotator cuff tendinopathy affects a large proportion of patients in orthopedic surgery practices. Corticosteroid injections are a common intervention proposed for these patients. The clinical evaluation of a response to corticosteroid injections is usually based only on the patient's self-evaluation of his function, activity, and pain by multiple questionnaires with varying metrological qualities. Objective measures of upper extremity functions are lacking, but wearable sensors are emerging as potential tools to assess upper extremity function and activity.

OBJECTIVE

This study aimed (1) to evaluate and compare test-retest reliability and sensitivity to change of known clinical assessments of shoulder function to wrist-based accelerometer measures and visual analog scales (VAS) of shoulder activity during daily living in patients with rotator cuff tendinopathy convergent validity and (2) to determine the acceptability and compliance of using wrist-based wearable sensors.

METHODS

A total of 38 patients affected by rotator cuff tendinopathy wore wrist accelerometers on the affected side for a total of 5 weeks. Western Ontario Rotator Cuff (WORC) index; Short version of the Disability of the Arm, Shoulder, and Hand questionnaire (QuickDASH); and clinical examination (range of motion and strength) were performed the week before the corticosteroid injections, the day of the corticosteroid injections, and 2 and 4 weeks after the corticosteroid injections. Daily Single Assessment Numeric Evaluation (SANE) and VAS were filled by participants to record shoulder pain and activity. Accelerometer data were processed to extract daily upper extremity activity in the form of active time; activity counts; and ratio of low-intensity activities, medium-intensity activities, and high-intensity activities.

RESULTS

Daily pain measured using VAS and SANE correlated well with the WORC and QuickDASH questionnaires (r=0.564-0.815) but not with accelerometry measures, amplitude, and strength. Daily activity measured with VAS had good correlation with active time (r=0.484, P=.02). All questionnaires had excellent test-retest reliability at 1 week before corticosteroid injections (intraclass correlation coefficient [ICC]=0.883-0.950). Acceptable reliability was observed with accelerometry (ICC=0.621-0.724), apart from low-intensity activities (ICC=0.104). Sensitivity to change was excellent at 2 and 4 weeks for all questionnaires (standardized response mean=1.039-2.094) except for activity VAS (standardized response mean=0.50). Accelerometry measures had low sensitivity to change at 2 weeks, but excellent sensitivity at 4 weeks (standardized response mean=0.803-1.032).

CONCLUSIONS

Daily pain VAS and SANE had good correlation with the validated questionnaires, excellent reliability at 1 week, and excellent sensitivity to change at 2 and 4 weeks. Daily activity VAS and accelerometry-derived active time correlated well together. Activity VAS had excellent reliability, but moderate sensitivity to change. Accelerometry measures had moderate reliability and acceptable sensitivity to change at 4 weeks.

Wearable systems for shoulder kinematics assessment: a systematic review

BACKGROUND

Wearable sensors are acquiring more and more influence in diagnostic and rehabilitation field to assess motor abilities of people with neurological or musculoskeletal impairments. The aim of this systematic literature review is to analyze the wearable systems for monitoring shoulder kinematics and their applicability in clinical settings and rehabilitation.

METHODS

A comprehensive search of PubMed, Medline, Google Scholar and IEEE Xplore was performed and results were included up to July 2019. All studies concerning wearable sensors to assess shoulder kinematics were retrieved.

RESULTS

Seventy-three studies were included because they have fulfilled the inclusion criteria. The results showed that magneto and/or inertial sensors are the most used. Wearable sensors measuring upper limb and/or shoulder kinematics have been proposed to be applied in patients with different pathological conditions such as stroke, multiple sclerosis, osteoarthritis, rotator cuff tear. Sensors placement and method of attachment were broadly heterogeneous among the examined studies.

CONCLUSIONS

Wearable systems are a promising solution to provide quantitative and meaningful clinical information about progress in a rehabilitation pathway and to extrapolate meaningful parameters in the diagnosis of shoulder pathologies. There is a strong need for development of this novel technologies which undeniably serves in shoulder evaluation and therapy.

Wearable Monitoring of Joint Angle and Muscle Activity

Simultaneous tracking of muscle activity and joint rotation is of significant interest in rehabilitation, but gold-standard methods with optical motion tracking and wireless electromyography recording typically restricts this to the laboratory setting. There has been significant progress using wear-able inertial measurement units (IMUs) for motion tracking, but there are no systems that can easily be deployed to home and provide simultaneous electromyography. We addressed this gap by developing a flexible, wearable, Bluetooth-connected sensor that records both IMU and EMG activity. The sensor runs an efficient quaternion-based complementary filter that estimates the sensor orientation while correcting for estimate drift and constraining magnetometer estimates to only influence heading. The difference in two sensor orientations is used to estimate the joint angle, which can be further improved with joint axis estimation. We demonstrate successful tracking of joint angle and muscle activity in a home environment with just the sensors and a smartphone.

Measurement of Upper Limb Range of Motion Using Wearable Sensors: A Systematic Review

Background

Wearable sensors are portable measurement tools that are becoming increasingly popular for the measurement of joint angle in the upper limb. With many brands emerging on the market, each with variations in hardware and protocols, evidence to inform selection and application is needed. Therefore, the objectives of this review were related to the use of wearable sensors to calculate upper limb joint angle. We aimed to describe (i) the characteristics of commercial and custom wearable sensors, (ii) the populations for whom researchers have adopted wearable sensors, and (iii) their established psychometric properties.

Methods

A systematic review of literature was undertaken using the following data bases: MEDLINE, EMBASE, CINAHL, Web of Science, SPORTDiscus, IEEE, and Scopus. Studies were eligible if they met the following criteria: (i) involved humans and/or robotic devices, (ii) involved the application or simulation of wearable sensors on the upper limb, and (iii) calculated a joint angle.

Results

Of 2191 records identified, 66 met the inclusion criteria. Eight studies compared wearable sensors to a robotic device and 22 studies compared to a motion analysis system. Commercial (n = 13) and custom (n = 7) wearable sensors were identified, each with variations in placement, calibration methods, and fusion algorithms, which were demonstrated to influence accuracy.

Conclusion

Wearable sensors have potential as viable instruments for measurement of joint angle in the upper limb during active movement. Currently, customised application (i.e. calibration and angle calculation methods) is required to achieve sufficient accuracy (error <  5°). Additional research and standardisation is required to guide clinical application.

Trial Registration

This systematic review was registered with PROSPERO (CRD42017059935).

Predictive trajectory estimation during rehabilitative tasks in augmented reality using inertial sensors

This paper presents a wireless kinematic tracking framework used for biomechanical analysis during rehabilitative tasks in augmented and virtual reality. The framework uses low-cost inertial measurement units and exploits the rigid connections of the human skeletal system to provide egocentric position estimates of joints to centimeter accuracy. On-board sensor fusion combines information from three-axis accelerometers, gyroscopes, and magnetometers to provide robust estimates in real-time. Sensor precision and accuracy were validated using the root mean square error of estimated joint angles against ground truth goniometer measurements. The sensor network produced a mean estimate accuracy of 2.81° with 1.06° precision, resulting in a maximum hand tracking error of 7.06 cm. As an application, the network is used to collect kinematic information from an unconstrained object manipulation task in augmented reality, from which dynamic movement primitives are extracted to characterize natural task completion in N = 3 able-bodied human subjects. These primitives are then leveraged for trajectory estimation in both a generalized and a subject-specific scheme resulting in 0.187 cm and 0.161 cm regression accuracy, respectively. Our proposed kinematic tracking network is wireless, accurate, and especially useful for predicting voluntary actuation in virtual and augmented reality applications.

Survey of Motion Tracking Methods Based on Inertial Sensors: A Focus on Upper Limb Human Motion

Motion tracking based on commercial inertial measurements units (IMUs) has been widely studied in the latter years as it is a cost-effective enabling technology for those applications in which motion tracking based on optical technologies is unsuitable. This measurement method has a high impact in human performance assessment and human-robot interaction. IMU motion tracking systems are indeed self-contained and wearable, allowing for long-lasting tracking of the user motion in situated environments. After a survey on IMU-based human tracking, five techniques for motion reconstruction were selected and compared to reconstruct a human arm motion. IMU based estimation was matched against motion tracking based on the Vicon marker-based motion tracking system considered as ground truth. Results show that all but one of the selected models perform similarly (about 35 mm average position estimation error).

Human Arm Motion Tracking by Orientation-Based Fusion of Inertial Sensors and Kinect Using Unscented Kalman Filter

Due to various applications of human motion capture techniques, developing low-cost methods that would be applicable in nonlaboratory environments is under consideration. MEMS inertial sensors and Kinect are two low-cost devices that can be utilized in home-based motion capture systems, e.g., home-based rehabilitation. In this work, an unscented Kalman filter approach was developed based on the complementary properties of Kinect and the inertial sensors to fuse the orientation data of these two devices for human arm motion tracking during both stationary shoulder joint position and human body movement. A new measurement model of the fusion algorithm was obtained that can compensate for the inertial sensors drift problem in high dynamic motions and also joints occlusion in Kinect. The efficiency of the proposed algorithm was evaluated by an optical motion tracker system. The errors were reduced by almost 50% compared to cases when either inertial sensor or Kinect measurements were utilized.

Upper Limb Kinematics Using Inertial and Magnetic Sensors: Comparison of Sensor-to-Segment Calibrations

Magneto-Inertial Measurement Unit sensors (MIMU) display high potential for the quantitative evaluation of upper limb kinematics, as they allow monitoring ambulatory measurements. The sensor-to-segment calibration step, consisting of establishing the relation between MIMU sensors and human segments, plays an important role in the global accuracy of joint angles. The aim of this study was to compare sensor-to-segment calibrations for the MIMU-based estimation of wrist, elbow, and shoulder joint angles, by examining trueness (“close to the reference”) and precision (reproducibility) validity criteria. Ten subjects performed five sessions with three different operators. Three classes of calibrations were studied: segment axes equal to technical MIMU axes (TECH), segment axes generated during a static pose (STATIC), and those generated during functional movements (FUNCT). The calibrations were compared during the maximal uniaxial movements of each joint, plus an extra multi-joint movement. Generally, joint angles presented good trueness and very good precision in the range 5°–10°. Only small discrepancy between calibrations was highlighted, with the exception of a few cases. The very good overall accuracy (trueness and precision) of MIMU-based joint angle data seems to be more dependent on the level of rigor of the experimental procedure (operator training) than on the choice of calibration itself.

Real-Time Arm Tracking for HMI Applications

Limb tracking is an important aspect of human-machine interfaces (HMI). These systems, however, can often be limited by complex algorithms requiring significant processing power, obtrusive and immobile sensing techniques, and high costs. In this work, we utilize a sensor fusion algorithm implemented in commercial inertial measurement units (IMU) to combine accelerometer and gyroscope measurements in an effort to minimize computational requirements of the limb tracking system. In addition, previously developed methods were implemented to eliminate sensor drift by including information from a magnetometer. We tested the accuracy of our system by computing the root mean squared error (RMSE) of the true angle between the headings of two sensors and the estimate of that angle through quaternion-vector manipulations. An average RMSE of approximately 2.9° was achieved. Our limb tracking system is wearable, minimally complex, low-cost, and simple to use which has proven useful in multiple HMI applications discussed herein.

Human Joint Angle Estimation with Inertial Sensors and Validation with A Robot Arm

Traditionally, human movement has been captured primarily by motion capture systems. These systems are costly, require fixed cameras in a controlled environment, and suffer from occlusion. Recently, the availability of low-cost wearable inertial sensors containing accelerometers, gyroscopes, and magnetometers have provided an alternative means to overcome the limitations of motion capture systems. Wearable inertial sensors can be used anywhere, cannot be occluded, and are low cost. Several groups have described algorithms for tracking human joint angles. We previously described a novel approach based on a kinematic arm model and the Unscented Kalman Filter (UKF). Our proposed method used a minimal sensor configuration with one sensor on each segment. This paper reports significant improvements in both the algorithm and the assessment. The new model incorporates gyroscope and accelerometer random drift models, imposes physical constraints on the range of motion for each joint, and uses zero-velocity updates to mitigate the effect of sensor drift. A high-precision industrial robot arm precisely quantifies the performance of the tracker during slow, normal, and fast movements over continuous 15-min recording durations. The agreement between the estimated angles from our algorithm and the high-precision robot arm reference was excellent. On average, the tracker attained an RMS angle error of about 3(°) for all six angles. The UKF performed slightly better than the more common Extended Kalman Filter.

A simple calibration for upper limb motion tracking and reconstruction

This paper extends the work of inertial sensor based upper limb motion tracking by introducing a simple calibration method to automatically construct a global reference frame and estimate arm length. The method has effectively eliminated the requirement of manually aligning the sensors' local reference frames when multiple sensors are used to track the movements of the individual arm segments. The capacity of arm length estimation also makes it possible to reconstruct position trajectories of the elbow and the wrist joints in a reference frame with the shoulder joint as the origin. Verification of the algorithm has been done by comparing the estimated arm length with the Kinect captured pseudo ground truth. Effectiveness of the algorithm can be observed by visualizing the reconstructed position trajectories of the arm joints.

Arm hand skilled performance in cerebral palsy: activity preferences and their movement components

Background

Assessment of arm-hand use is very important in children with cerebral palsy (CP) who encounter arm-hand problems. To determine validity and reliability of new instruments to assess actual performance, a set of standardized test situations including activities of daily living (ADL) is required. This study gives information with which such a set for upper extremity skill research may be fine-tuned, relative to a specific research question. Aim of this study is to a) identify upper extremity related ADL children with CP want to improve on, b) determine the 10 most preferred goals of children with CP, and c) identify movement components of all goals identified.

Method

The Canadian Occupational Performance Measure was used to identify upper extremity-related ADL preferences (goals) of 53 children with CP encountering arm-hand problems (mean age 9 ± 4.5 year). Goals were ranked based on importance attributed to each goal and the number of times a goal was mentioned, resulting in a gross list with goals. Additionally, two studies were performed, i.e. study A to determine the 10 most preferred goals for 3 age groups (2.5-5 years; 6-11 years, 12-19 years), based on the total preference score, and study B to identify movement components, like reaching and grasping, of all goals identified for both the leading and the assisting arm-hand.

Results

Seventy-two goals were identified. The 10 most preferred goals differed with age, changing from dressing and leisure-related goals in the youngest children to goals regarding personal care and eating for children aged 6-11 years. The oldest children preferred goals regarding eating, personal care and computer use. The movement components ‘positioning’, ‘reach’, ‘grasp’, and ‘hold’ were present in most tasks. ‘Manipulating’ was more important for the leading arm-hand, whereas ‘fixating’ was more important for the assisting arm-hand.

Conclusion

This study gave insight into the preferences regarding ADL children with CP would like to improve on, and the movement components characterizing these activities. This information can be used to create a set of standardized test situations, which can be used to assess the validity and reliability of new measurement instruments to gauge actual arm-hand skilled performance.

Human-Scale Motion Capture with an Accelerometer-Based Gaming Controller

Gaming controllers are attractive devices for research due to their onboard sensing capabilities and low cost. However, a proper quantitative analysis regarding their suitability for motion capture has yet to be fully reported. In this paper, a detailed analysis of the accelerometers of the Nintendo Wiimote is presented. The gravity-compensated acceleration data from the accelerometers of theWiimote were plotted, compared and correlated with computed acceleration data derived from a six-camera motion capture system. The results show high correlation and low mean absolute error between the gravity-compensated data from the accelerometers of the controllers and computed acceleration from position data of the motion capture system. From the results obtained, it can be inferred that the Wiimote is well suited for motion capture applications where post-processing of data is practical.