In vivo estimation of the shoulder joint center of rotation using magneto-inertial sensors: MRI-based accuracy and repeatability assessment

Feasibility of reconstructing the glenohumeral center of rotation with a single camera setup

BACKGROUND

An accurate estimation of the glenohumeral joint center of rotation (CoR) is important during alignment of braces and exoskeletons, as a misalignment will introduce undesired forces on the human body. The aim of this research was to develop a new method to estimate the glenohumeral CoR and register the location to the body using a single camera and two printed markers.

METHODS

During shoulder anteflexion, the arm roughly describes an arc in the sagittal plane, with the glenohumeral joint in the center. Two binary square-fiducial ArUco markers were secured to the upper arm and the scapula, their position and orientation were obtained, and a sphere was fitted to the coordinates of the arm marker. The sphere center position was then registered on the skin. The accuracy was assessed with a test bench with a known rotational center. The repeatability was assessed in vivo with five healthy participants.

RESULTS

The mean absolute offset between the true CoR of the test bench and the fitted sphere centers across multiple trials was 2.7 mm at a velocity of 30 degrees/s, and 2.5 mm at 60 degrees/s. The root mean squared distance from the estimated sphere centers after each trial to the mean sphere center across all trials per participant was 5.1 mm on average for the novice examiner and 5.2 mm for the expert examiner.

CONCLUSIONS

The proposed method is able to accurately and precisely estimate the glenohumeral CoR.

Clinical methods of dynamic and quantitative evaluation of the shoulder and scapula complex: a scoping review

ABSTRACT The shoulder joint has the greatest range of motion and is the most susceptible to dysfunction. Dynamic and quantitative evaluations of this region provide better information for the clinic but the choice of the method depends on its measurement properties. This study aimed to identify the existing methods of quantitative dynamic evaluation of the shoulder and scapula complex, in a clinical context for the general population, identifying the measurement properties and outcomes of each method. The scoping review included in vivo studies, with samples without a specific clinical condition and involving applicable methods in a clinical context. We identified evaluated outcome, measurement method, and its measurement properties. We selected 29 studies that investigated 12 measurement methods, and evaluated their validity and reliability for 17 different outcomes. Most studies (n=21) addressed the position of the shoulder and the scapula and the derivative outcomes, using mainly the units of inertial measurement (n=5) and inertial magnetic measurement (n=6) as evaluation methods. The outcomes with valid and reliable methods were: shoulder joint range; scapula and shoulder motion range; muscle activity; shoulder joint center; humerus length; torque-time curve; functional performance; scapular dyskinesia; external shoulder rotators force; shoulder joint functionality and range; initial scapular movement; scapula and shoulder position; and shoulder angular velocity.

Métodos clínicos de avaliação dinâmica e quantitativa do complexo ombro e escápula: uma revisão de escopo

RESUMO A articulação do ombro possui a maior amplitude de movimento e está mais suscetível a disfunções. Avaliações dinâmicas e quantitativas dessa região fornecem melhores informações para a clínica, mas a escolha do método a ser utilizado depende de suas propriedades de medição. O objetivo deste estudo foi identificar os métodos existentes de avaliação dinâmica quantitativa do complexo ombro e escápula em um contexto clínico para a população em geral, identificando as propriedades de medição e os desfechos avaliados para cada método. A revisão de escopo incluiu estudos in vivo, com amostras sem uma condição clínica específica e envolvendo métodos aplicáveis em um contexto clínico. Foram identificados: desfecho avaliado, método de medição e suas propriedades de medição. Foram selecionados 29 estudos que investigaram 12 métodos de medição, sendo avaliadas sua validade e confiabilidade para 17 desfechos diferentes. A posição do ombro e da escápula e os desfechos derivados foram abordados pelo maior número de estudos (n=21), sendo seus principais métodos de avaliação as unidades de medição inercial (n=5) e unidades de medição magnética inercial (n=6). Os desfechos que apresentaram métodos válidos e confiáveis foram: amplitude articular de ombro; amplitude de movimento da escápula e do ombro; atividade muscular; centro articular do ombro; comprimento do úmero; curva torque-tempo; desempenho funcional; discinesia escapular; força de rotadores externos do ombro; funcionalidade e amplitude articular; movimento escapular inicial; posição da escápula e do ombro; e velocidade angular do ombro.

Rotation sequence and marker tracking method affects the humerothoracic kinematics of manual wheelchair propulsion

The literature on shoulder (humerothoracic) kinematics in manual wheelchair propulsion is growing. Inconsistencies in the reporting of which rotation sequence is used to compute three-dimensional (3D) angles complicates the interpretation and comparison between studies. The purpose of this study was to compare the effects of three often used and recommended rotation sequences (ZXY, XZY, and YXY) and two tracking methods (anatomical and cluster only) on the humerothoracic kinematics of manual wheelchair propulsion. Fourteen able-bodied participants performed manual wheelchair propulsion on a treadmill, while a motion capture system recorded the movements at 120 Hz. Humeral and thoracic segment coordinate systems were constructed according to ISB recommendations. Humerothoracic angles were calculated using each of the three rotation sequences. The ZXY and XZY sequences yielded similar angles in terms of both shape and amplitude, but, perhaps unsurprisingly, these differed substantially from the YXY sequence. Anatomical tracking showed neither gimbal locks nor phase angle discontinuities for any rotation sequence, while cluster tracking yielded phase angle discontinuities for the ZXY and YXY rotation sequences. The two tracking methods yielded similar joint angles for all sequences except for internal/external rotation, and the cluster-only method had larger variability than the anatomical method. These results highlight the importance of reporting which rotation sequence and tracking method are used when calculating humerothoracic angles in order to allow for straightforward interpretation of results and comparison across studies.

A functional calibration protocol for ankle plantar-dorsiflexion estimate using magnetic and inertial measurement units: Repeatability and reliability assessment

The ankle joint complex presents a tangled functional anatomy, which understanding is fundamental to effectively estimate its kinematics on the sagittal plane. Protocols based on the use of magnetic and inertial measurement units (MIMUs) currently do not take in due account this factor. To this aim, a joint coordinate system for the ankle joint complex is proposed, along with a protocol to perform its anatomical calibration using MIMUs, consisting in a combination of anatomical functional calibrations of the tibiotalar axis and static acquisitions. Protocol repeatability and reliability were tested according to the metrics proposed in Schwartz et al. (2004) involving three different operators performing the protocol three times on ten participants, undergoing instrumented gait analysis through both stereophotogrammetry and MIMUs. Instrumental reliability was evaluated comparing the MIMU-derived kinematic traces with the stereophotogrammetric ones, obtained with the same protocol, through the linear fit method. A total of 270 gait cycles were considered. Results showed that the protocol was repeatable and reliable for what concerned the operators (0.4 ± 0.4 deg and 0.8 ± 0.5 deg, respectively). Instrumental reliability analysis showed a mean RMSD of 3.0 ± 1.3 deg, a mean offset of 9.4 ± 8.4 deg and a mean linear relationship strength of R² = 0.88 ± 0.08. With due caution, the protocol can be considered both repeatable and reliable. Further studies should pay attention to the other ankle degrees of freedom as well as on the angular convention to compute them.

A database of physical therapy exercises with variability of execution collected by wearable sensors

This document introduces the PHYTMO database, which contains data from physical therapies recorded with inertial sensors, including information from an optical reference system. PHYTMO includes the recording of 30 volunteers, aged between 20 and 70 years old. A total amount of 6 exercises and 3 gait variations were recorded. The volunteers performed two series with a minimum of 8 repetitions in each one. PHYTMO includes magneto-inertial data, together with a highly accurate location and orientation in the 3D space provided by the optical system. The files were stored in CSV format to ensure its usability. The aim of this dataset is the availability of data for two main purposes: the analysis of techniques for the identification and evaluation of exercises using inertial sensors and the validation of inertial sensor-based algorithms for human motion monitoring. Furthermore, the database stores enough data to apply Machine Learning-based algorithms. The participants' age range is large enough to establish age-based metrics for the exercises evaluation or the study of differences in motions between different groups.

In-vitro validation of inertial-sensor-to-bone alignment

A major shortcoming in kinematic estimation using skin-attached inertial sensors is the alignment of sensor-embedded and segment-embedded coordinate systems. Only a correct alignment results in clinically relevant kinematics. Model-based inertial-sensor-to-bone alignment methods relate inertial sensor measurements with a model of the joint. Therefore, they do not rely on properly executed calibration movements or a correct sensor placement. However, it is unknown how accurate such model-based methods align the sensor axes and the underlying segment-embedded axes, as defined by clinical definitions. Also, validation of the alignment models is challenging, since an optical motion capture ground truth can be prone to disturbances from soft tissue movement, orientation estimation and manual palpation errors. We present an anatomical tibiofemoral ground truth on an unloaded cadaveric measurement set-up that intrinsically overcomes these disturbances. Additionally, we validate existing model-based alignment strategies. Modeling the degrees of freedom leads to the identification of rotation axes. However, there is no reason why these axes would align with the segment-embedded axes. Relative inertial-sensor orientation information and rich arbitrary movements showed to aid in identifying the underlying joint axes. The first dominant sagittal rotation axis aligned sufficiently well with the underlying segment-embedded reference. The estimated axes that relate to secondary kinematics tend to deviate from the underlying segment-embedded axes as much as their expected range of motion around the axes. In order to interpret the secondary kinematics, the alignment model should more closely match the biomechanics of the joint.

Custom IMU-Based Wearable System for Robust 2.4 GHz Wireless Human Body Parts Orientation Tracking and 3D Movement Visualization on an Avatar

Recent studies confirm the applicability of Inertial Measurement Unit (IMU)-based systems for human motion analysis. Notwithstanding, high-end IMU-based commercial solutions are yet too expensive and complex to democratize their use among a wide range of potential users. Less featured entry-level commercial solutions are being introduced in the market, trying to fill this gap, but still present some limitations that need to be overcome. At the same time, there is a growing number of scientific papers using not commercial, but custom do-it-yourself IMU-based systems in medical and sports applications. Even though these solutions can help to popularize the use of this technology, they have more limited features and the description on how to design and build them from scratch is yet too scarce in the literature. The aim of this work is two-fold: (1) Proving the feasibility of building an affordable custom solution aimed at simultaneous multiple body parts orientation tracking; while providing a detailed bottom-up description of the required hardware, tools, and mathematical operations to estimate and represent 3D movement in real-time. (2) Showing how the introduction of a custom 2.4 GHz communication protocol including a channel hopping strategy can address some of the current communication limitations of entry-level commercial solutions. The proposed system can be used for wireless real-time human body parts orientation tracking with up to 10 custom sensors, at least at 50 Hz. In addition, it provides a more reliable motion data acquisition in Bluetooth and Wi-Fi crowded environments, where the use of entry-level commercial solutions might be unfeasible. This system can be used as a groundwork for developing affordable human motion analysis solutions that do not require an accurate kinematic analysis.

Reference in-vitro dataset for inertial-sensor-to-bone alignment applied to the tibiofemoral joint

Skin-attached inertial sensors are increasingly used for kinematic analysis. However, their ability to measure outside-lab can only be exploited after correctly aligning the sensor axes with the underlying anatomical axes. Emerging model-based inertial-sensor-to-bone alignment methods relate inertial measurements with a model of the joint to overcome calibration movements and sensor placement assumptions. It is unclear how good such alignment methods can identify the anatomical axes. Any misalignment results in kinematic cross-talk errors, which makes model validation and the interpretation of the resulting kinematics measurements challenging. This study provides an anatomically correct ground-truth reference dataset from dynamic motions on a cadaver. In contrast with existing references, this enables a true model evaluation that overcomes influences from soft-tissue artifacts, orientation and manual palpation errors. This dataset comprises extensive dynamic movements that are recorded with multimodal measurements including trajectories of optical and virtual (via computed tomography) anatomical markers, reference kinematics, inertial measurements, transformation matrices and visualization tools. The dataset can be used either as a ground-truth reference or to advance research in inertial-sensor-to-bone-alignment.

Adherence Tracking With Smart Watches for Shoulder Physiotherapy in Rotator Cuff Pathology: Protocol for a Longitudinal Cohort Study

Background

Physiotherapy is essential for the successful rehabilitation of common shoulder injuries and following shoulder surgery. Patients may receive some training and supervision for shoulder physiotherapy through private pay or private insurance, but they are typically responsible for performing most of their physiotherapy independently at home. It is unknown how often patients perform their home exercises and if these exercises are performed correctly without supervision. There are no established tools for measuring this. It is, therefore, unclear if the full benefit of shoulder physiotherapy treatments is being realized.

Objective

The proposed research will (1) validate a smartwatch and machine learning (ML) approach for evaluating adherence to shoulder exercise participation and technique in a clinical patient population with rotator cuff pathology; (2) quantify the rate of home physiotherapy adherence, determine the effects of adherence on recovery, and identify barriers to successful adherence; and (3) develop and pilot test an ethically conscious adherence-driven rehabilitation program that individualizes patient care based on their capacity to effectively participate in their home physiotherapy.

Methods

This research will be conducted in 2 phases. The first phase is a prospective longitudinal cohort study, involving 120 patients undergoing physiotherapy for rotator cuff pathology. Patients will be issued a smartwatch that will record 9-axis inertial sensor data while they perform physiotherapy exercises both in the clinic and in the home setting. The data collected in the clinic under supervision will be used to train and validate our ML algorithms that classify shoulder physiotherapy exercise. The validated algorithms will then be used to assess home physiotherapy adherence from the inertial data collected at home. Validated outcome measures, including the Disabilities of the Arm, Shoulder, and Hand questionnaire; Numeric Pain Rating Scale; range of motion; shoulder strength; and work status, will be collected pretreatment, monthly through treatment, and at a final follow-up of 12 months. We will then relate improvement in patient outcomes to measured physiotherapy adherence and patient baseline variables in univariate and multivariate analyses. The second phase of this research will involve the evaluation of a novel rehabilitation program in a cohort of 20 patients. The program will promote patient physiotherapy engagement via the developed technology and support adherence-driven care decisions.

Results

As of December 2019, 71 patients were screened for enrollment in the noninterventional validation phase of this study; 65 patients met the inclusion and exclusion criteria. Of these, 46 patients consented and 19 declined to participate in the study. Only 2 patients de-enrolled from the study and data collection is ongoing for the remaining 44.

Conclusions

This study will provide new and important insights into shoulder physiotherapy adherence, the relationship between adherence and recovery, barriers to better adherence, and methods for addressing them.

International Registered Report Identifier (IRRID)

DERR1-10.2196/17841

Effect of rotator cuff muscle activation on glenohumeral kinematics: A cadaveric study

Healthy shoulder function requires the coordination of the rotator cuff muscles to maintain the humeral head's position in the glenoid. While glenohumeral stability has been studied in various settings, few studies have characterized the effect of dynamic rotator cuff muscle loading on glenohumeral translation during shoulder motion. We hypothesize that dynamic rotator cuff muscle activation decreases joint translation during continuous passive abduction of the humerus in a cadaveric model of scapular plane glenohumeral abduction. The effect of different rotator cuff muscle activity on glenohumeral translation was assessed using a validated shoulder testing system. The Dynamic Load profile is a novel approach, based on musculoskeletal modeling of human subject motion. Passive humeral elevation in the scapular plane was applied via the testing system arm, while the rotator cuff muscles were activated according to the specified force profiles using stepper motors and a proportional control feedback loop. Glenohumeral translation was defined according to the International Society of Biomechanics. The Dynamic load profile minimized superior translation of the humeral head relative to the conventional loading profiles. The total magnitude of translation was not significantly different (0.805) among the loading profiles suggesting that the compressive forces from the rotator cuff primarily alter the direction of humeral head translation, not the magnitude. Rotator cuff muscle loading is an important element of cadaveric shoulder studies that must be considered to accurately simulate glenohumeral motion. A rotator cuff muscle activity profile based on human subject muscle activity reduces superior glenohumeral translation when compared to previous RC loading profiles.

A Novel Exoskeleton with Fractional Sliding Mode Control for Upper Limb Rehabilitation

The robotic intervention has great potential in the rehabilitation of post-stroke patients to regain their lost mobility. In this paper, firstly, we present a design of a novel, 7 degree-of-freedom (DOF) upper limb robotic exoskeleton (u-Rob) that features shoulder scapulohumeral rhythm with a wide range of motions (ROM) compared to other existing exoskeletons. An ergonomic shoulder mechanism with two passive DOF was included in the proposed exoskeleton to provide scapulohumeral motion with corresponding full ROM. Also, the joints of u-Rob have more range of motions compared to its existing counterparts. Secondly, we propose a fractional sliding mode control (FSMC) to control u-Rob. Applying the Lyapunov theory to the proposed control algorithm, we showed the stability of it. To control u-Rob, FSMC has shown effectiveness to handle unmodeled dynamics (e.g. friction, disturbance, etc.) in terms of better tracking and chatter compared to traditional SMC.

Validation of Novel Relative Orientation and Inertial Sensor-to-Segment Alignment Algorithms for Estimating 3D Hip Joint Angles

Wearable sensor-based algorithms for estimating joint angles have seen great improvements in recent years. While the knee joint has garnered most of the attention in this area, algorithms for estimating hip joint angles are less available. Herein, we propose and validate a novel algorithm for this purpose with innovations in sensor-to-sensor orientation and sensor-to-segment alignment. The proposed approach is robust to sensor placement and does not require specific calibration motions. The accuracy of the proposed approach is established relative to optical motion capture and compared to existing methods for estimating relative orientation, hip joint angles, and range of motion (ROM) during a task designed to exercise the full hip range of motion (ROM) and fast walking using root mean square error (RMSE) and regression analysis. The RMSE of the proposed approach was less than that for existing methods when estimating sensor orientation ( 12 . 32 ∘ and 11 . 82 ∘ vs. 24 . 61 ∘ and 23 . 76 ∘ ) and flexion/extension joint angles ( 7 . 88 ∘ and 8 . 62 ∘ vs. 14 . 14 ∘ and 15 . 64 ∘ ). Also, ROM estimation error was less than 2 . 2 ∘ during the walking trial using the proposed method. These results suggest the proposed approach presents an improvement to existing methods and provides a promising technique for remote monitoring of hip joint angles.

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.

Shoulder kinematics and kinetics of team handball throwing: A scoping review

In recent years a number of studies have investigated shoulder biomechanics in handball throwing. The purpose of this scoping review is to summarize the current handball research in terms of shoulder joint kinematics and kinetics and identify gaps in the current research. Nineteen articles relevant to this topic were identified and included. The handball throw is characterized by large external shoulder rotation followed by a rapid internal rotation with minor changes in shoulder flexion and abduction. Generally timing sequence, joint angles and joint velocities were not affected by different conditions such as throwing type, arm position, ball weight and gender. However, large differences in shoulder angles and angular velocities were found between studies, which most likely are explained by methodological differences. Unfortunately, the information provided in the articles did not make it possible to transform measurements from one study to another and thereby eliminate the methodological differences. Only one study reported shoulder kinetics and found that kinetics were not affected by fatigue. This scoping review highlights the need for better descriptions of the methods used to obtain shoulder kinematics and for studies investigating shoulder kinetics in handball throwing.

Inertial sensors versus standard systems in gait analysis: a systematic review and meta-analysis

INTRODUCTION

The increasing popularity of inertial sensors in clinical practice is not supported by precise information on their reliability or guidelines for their use in rehabilitation. The authors investigated the state of the literature concerning the use of inertial sensors for gait analysis in both healthy and pathological adults comparing traditional systems. Furthermore, trying to define directions for clinicians.

EVIDENCE ACQUISITION

In accordance with the PRISMA statement, authors searched in PubMed, Web of Science and Scopus all paper published from January 1st, 2005 until December 31st, 2017. They included both healthy and pathological adults' subjects as population, wearable or inertial sensors used for gait analysis and compared with classical gait analysis performed in a Motion Lab as intervention and comparison, gait parameters as outcomes. Considering the methodological quality, authors focused on: sample; description of the study; type of gait analysis used for comparison; type of sensor; sensor placement on the body; gait task requested.

EVIDENCE SYNTHESIS

From a total of 888 articles, 16 manuscripts were selected and 7 of them were considered for meta-analysis for different gait parameters. Demographic data, tested devices, reference systems, test procedures and outcomes were analyzed.

CONCLUSIONS

Our results show a good agreement between inertial sensors and classical gait analysis for some gait parameters, supporting their use as a solution for capturing kinematic information over an extended space and time and even outside a laboratory in real-life conditions. Authors can support the use of portable inertial sensors for a practical gait analysis in clinical setting with good reliability. It will then be the experience of the clinician to direct the decision-making process.

Functional estimation of bony segment lengths using magneto-inertial sensing: Application to the humerus

Inertial sensor technology has assumed an increasingly important role in the field of human motion analysis. However, the reliability of the kinematic estimates could still be critical for specific applications in the field of functional evaluation and motor rehabilitation. Within this context, the definition of subject-specific multi-body kinematic models is crucial since it affects the accuracy and repeatability of movement reconstruction. A key step for kinematic model calibration is the determination of bony segment lengths. This study proposes a functional approach for the in vivo estimation of the humerus length using a single magneto-inertial measurement unit (MIMU) positioned on the right distal posterior forearm. The humerus length was estimated as the distance between the shoulder elevation axis and the elbow flexion-extension axis. The calibration exercise involved five shoulder elevations in the sagittal plane with the elbow completely extended and five elbow flexion-extensions with the upper arm rigidly aligned to the trunk. Validation of the method was conducted on five healthy subjects using the humerus length computed from magnetic resonance imaging as the gold standard. The method showed mean absolute errors of 12 ± 9 mm, which were in the estimate of the humerus length. When using magneto-inertial technology, the proposed functional method represents a promising alternative to the regressive methods or manual measurements for performing kinematic model calibrations. Although the proposed methodology was validated for the estimation of the humerus length, the same approach can be potentially extended to other body segments.

Elbow joint kinematics during cricket bowling using magneto-inertial sensors: A feasibility study

Magnetic and inertial measurement units (MIMUs) may provide an accessible, three-dimensional, in-field alternative to laboratory-restricted marker-based motion capture. Existing upper limb MIMU models have predominantly been validated with low-velocity motion and their suitability for use with sport-based movements remains relatively untested. We propose a MIMU system approach to enable the estimation of anatomically meaningful and participant-specific elbow kinematics with considerations for use with cricket bowling. A novel standardised elbow reference posture of 90 degrees flexion and 0 deg pronation, and functional definition of elbow joint axes of rotation calibrated the MIMU method model before it was validated across three experiments: (1) simple elbow rotations with a mechanical linkage; (2) low-velocity elbow rotations in human participants; and (3) low-medium velocity sport-based movements in human participants. The proposed MIMU method demonstrated high elbow kinematic measurement agreement when compared with a criterion measure across all three conditions. However, during experiment 3, sensor components neared their measurement capacity and the MIMU method elbow flexion measurement variability increased. We conclude that the proposed MIMU method can estimate anatomically referenced, participant-specific joint angles, however, the hardware specifications of currently available systems may limit application in high-velocity/acceleration situations, preventing the measurement of cricket bowling in-field for now.

Shoulder physiotherapy exercise recognition: machine learning the inertial signals from a smartwatch

OBJECTIVE

Participation in a physical therapy program is considered one of the greatest predictors of successful conservative management of common shoulder disorders. However, adherence to these protocols is often poor and typically worse for unsupervised home exercise programs. Currently, there are limited tools available for objective measurement of adherence in the home setting. The goal of this study was to develop and evaluate the potential for performing home shoulder physiotherapy monitoring using a commercial smartwatch.

APPROACH

Twenty healthy adult subjects with no prior shoulder disorders performed seven exercises from an evidence-based rotator cuff physiotherapy protocol, while 6-axis inertial sensor data was collected from the active extremity. Within an activity recognition chain (ARC) framework, four supervised learning algorithms were trained and optimized to classify the exercises: k-nearest neighbor (k-NN), random forest (RF), support vector machine classifier (SVC), and a convolutional recurrent neural network (CRNN). Algorithm performance was evaluated using 5-fold cross-validation stratified first temporally and then by subject.

MAIN RESULTS

Categorical classification accuracy was above 94% for all algorithms on the temporally stratified cross validation, with the best performance achieved by the CRNN algorithm (99.4%). The subject stratified cross validation, which evaluated classifier performance on unseen subjects, yielded lower accuracies scores again with CRNN performing best (88.9%).

SIGNIFICANCE

This proof of concept study demonstrates the technical feasibility of a smartwatch device and supervised machine learning approach to more easily monitor and assess the at-home adherence of shoulder physiotherapy exercise protocols.

Shoulder assessment according to the international classification of functioning by means of inertial sensor technologies: A systematic review

This review investigates current protocols using Inertial Measurement Units (IMUs) in shoulder research, and outlines future paths regarding IMU use for shoulder research. Different databases were searched for relevant articles. Criteria for study selection were (1) research in healthy persons or persons with shoulder problems, (2) IMUs applied as assessment tool for the shoulder (in healthy subjects and shoulder patients) or upper limb (in shoulder patients), (3) peer-reviewed, full-text papers in English or Dutch. Studies with less than five participants and without ethical approval were excluded. Data extraction included (1) study design, (2) participant characteristics, (3) type/brand of IMU, (4) tasks included in the assessment protocol, and (5) outcomes. Risk of bias was assessed using the Downs and Black checklist. Scapulothoracic/glenohumeral and humerothoracic kinematics were reported in respectively 10 and 27 of the 37 included papers. Only one paper in healthy persons assessed, next to scapulothoracic/glenohumeral kinematics, other upper limb joints. IMUs' validity and reliability to capture shoulder function was limited. Considering applied protocols, 39% of the protocols was located on the International-Classification-of-Functioning (ICF) function level, while 38% and 23% were on the 'capacity' and 'actual performance'-sublevel, of the ICF-activity level. Most available IMU-research regarding the shoulder is clinically less relevant, given the widely reported humerothoracic kinematics which do not add to clinical-decision-making, and the absence of protocols assessing the complete upper limb chain. Apart from knowledge on methodological pitfalls and opportunities regarding the use of IMUs, this review provides future research paths.