Monitoring Physical Behavior in Rehabilitation Using a Machine Learning-Based Algorithm for Thigh-Mounted Accelerometers: Development and Validation Study.

BACKGROUND: Physical activity is emerging as an outcome measure. Accelerometers have become an important tool in monitoring physical behavior, and newer analytical approaches of recognition methods increase the degree of details. Many studies have achieved high performance in the classification of physical behaviors through the use of multiple wearable sensors; however, multiple wearables can be impractical and lower compliance. OBJECTIVE: The aim of this study was to develop and validate an algorithm for classifying several daily physical behaviors using a single thigh-mounted accelerometer and a supervised machine-learning scheme. METHODS: We collected training data by adding the behavior classes-running, cycling, stair climbing, wheelchair ambulation, and vehicle driving-to an existing algorithm with the classes of sitting, lying, standing, walking, and transitioning. After combining the training data, we used a random forest learning scheme for model development. We validated the algorithm through a simulated free-living procedure using chest-mounted cameras for establishing the ground truth. Furthermore, we adjusted our algorithm and compared the performance with an existing algorithm based on vector thresholds. RESULTS: We developed an algorithm to classify 11 physical behaviors relevant for rehabilitation. In the simulated free-living validation, the performance of the algorithm decreased to 57% as an average for the 11 classes (F-measure). After merging classes into sedentary behavior, standing, walking, running, and cycling, the result revealed high performance in comparison to both the ground truth and the existing algorithm. CONCLUSIONS: Using a single thigh-mounted accelerometer, we obtained high classification levels within specific behaviors. The behaviors classified with high levels of performance mostly occur in populations with higher levels of functioning. Further development should aim at describing behaviors within populations with lower levels of functioning.

Despite patient-reported outcomes improve, patients with femoroacetabular impingement syndrome do not increase their objectively measured sport and physical activity level 1 year after hip arthroscopic surgery. Results from the HAFAI cohort.

PURPOSE: Patients with femoroacetabular impingement syndrome (FAIS) are young and middle-aged persons living physically active lives including sports activities. However, measurements of the physical activity level before and after hip arthroscopic surgery in patients with FAIS using both self-reported and objective accelerometer-based measures are lacking. Furthermore, comparing patients with a reference group of persons reporting no hip problems and conducting subgroup analyses investigating changes in physical activity level and self-reported outcomes according to pre-surgery activity level may further highlight the activity pattern for patients. METHODS: Sixty patients with FAIS eligible for hip arthroscopic surgery were consecutively included in a prospective cohort study (HAFAI cohort) together with 30 reference persons reporting no hip problems. Participants completed the Copenhagen Hip and Groin Outcome Score (HAGOS) together with questions regarding their sports activities. Furthermore, participants wore a three-axial accelerometer for five consecutive days during waking hours. The accelerometer-based data were analysed and presented as total activity and type, frequency and duration of activities. RESULTS: Patients experienced significant and clinically relevant changes in all HAGOS scores. 88% of patients participated in some kind of sports activity 1 year after surgery. Overall, objectively measured physical activity did not change from before to 1 year after surgery. However, subgroup analyses of the most sedentary patients preoperatively revealed significant changes towards a more active pattern. Compared to reference persons, patients performed less bicycling and running. CONCLUSION: Despite clinically relevant changes in self-reported outcomes, patients did not increase their overall physical activity level 1 year after surgery. Physical activity levels were lower in patients than in the reference group and patients continued bicycling and running less compared with the reference group. LEVEL OF EVIDENCE: II.

Use of accelerometer-based activity monitoring in orthopaedics: benefits, impact and practical considerations.

Studies of the effectiveness of orthopaedic interventions do not generally measure physical activity (PA). Applying accelerometer-based activity monitoring in orthopaedic studies will add relevant information to the generally examined physical function and pain assessment.Accelerometer-based activity monitoring is practically feasible in orthopaedic patient populations, since current day activity sensors have battery time and memory to measure continuously for several weeks without requiring technical expertise.The ongoing development in sensor technology has made it possible to combine functional tests with activity monitoring.For clinicians, the application of accelerometer-based activity monitoring can provide a measure of PA and can be used for clinical comparisons before and after interventions.In orthopaedic rehabilitation, accelerometer-based activity monitoring may be used to help patients reach their targets for PA and to coach patients towards a more active lifestyle through direct feedback. Cite this article: EFORT Open Rev 2019;4:678-685. DOI: 10.1302/2058-5241.4.180041.

Are patients with knee osteoarthritis and patients with knee joint replacement as physically active as healthy persons?

BACKGROUND: It is well documented that patients with knee osteoarthritis (KOA) suffer from reduced physical function and that function of the affected knee is improved after knee joint replacement (KJR). However, it remains uncertain whether patients with KOA are less physically active than healthy people and whether patients increase their level of physical activity after surgery to a level comparable with that of healthy people. The aim of this study was to examine whether patients with KOA are less physically active than healthy participants and whether patients who have undergone KJR show an increased activity and achieve the same level of physical activity as healthy participants 5 years postoperatively. METHODS: Fifty-four patients with KOA (29 women; mean age 62 ± 8.6; mean body mass index (BMI) 27 ± 5), 52 patients who had KJR due to KOA 5 years earlier (26 women; mean age 66 ± 7.2; mean BMI 30 ± 5) and 171 healthy participants (76 women, mean age 64 ± 9.7, mean BMI 26 ± 5) were included in this cross-sectional study. The level of physical activity was measured over a mean period of 5.5 days with a triaxial accelerometer mounted on the thigh. Number of daily steps, number of daily short walking bouts of <10 s duration and number of daily transfers from sitting to standing position were recorded. Data were analysed using two sample t tests and were adjusted for age, gender and BMI. RESULTS: Patients with KOA did not differ significantly from healthy participants regarding daily steps (+321, p = 0.50) or daily transfers from sitting to standing (+1.9, p = 0.52) but performed significantly less daily short walking bouts <10 s (-11.9, p = 0.02). Patients after KJR did not differ significantly from healthy participants regarding daily steps (-281, p = 0.60) of transfers from sitting to standing position (-3.2, p = 0.32) but performed significantly less daily short walking bouts <10 s (-21.7, p = 0.001). CONCLUSION: Patients with KOA and KJR showed no significant reduction in number of daily step counts and transfers from sitting to standing position when compared with matched healthy controls. However, the number of short walking bouts was reduced in patients with KOA and by twice as much in patients with KJR. This indicates that KOA and treatment with KJR hardly affect health-related general activity but do affect specific physical activity behaviour potentially indicative of KOA or post-KJR functional limitations. THE TRANSLATIONAL POTENTIAL OF THIS ARTICLE: Activity monitoring with an accelerometer-based method gives insights into health-related general activity levels such as total daily steps and specific parameters such as short walking bouts, which may serve as an objective outcome measure in clinical practice.

Objective assessment of physical activity and sedentary behaviour in knee osteoarthritis patients - beyond daily steps and total sedentary time.

BACKGROUND: Knee osteoarthritis patients may become physically inactive due to pain and functional limitations. Whether physical activity exerts a protective or harmful effect depends on the frequency, intensity, time and type (F.I.T.T.). The F.I.T.T. dimensions should therefore be assessed during daily life, which so far has hardly been feasible. Furthermore, physical activity should be assessed within subgroups of patients, as they might experience different activity limitations. Therefore, this study aimed to objectively describe physical activity, by assessing the F.I.T.T. dimensions, and sedentary behaviour of knee osteoarthritis patients during daily life. An additional goal was to determine whether activity events, based on different types and durations of physical activity, were able to discriminate between subgroups of KOA patients based on risk factors. METHODS: Clinically diagnosed knee osteoarthritis patients (according to American College of Rheumatology criteria) were monitored for 1 week with a tri-axial accelerometer. Furthermore, they performed three functional tests and completed the Knee Osteoarthritis Outcome Score. Physical activity levels were described for knee osteoarthritis patients and compared between subgroups. RESULTS: Sixty-one patients performed 7303 mean level steps, 319 ascending and 312 descending steps and 601 bicycle crank revolutions per day. Most waking hours were spent sedentary (61%), with 4.6 bouts of long duration (> 30 min). Specific events, particularly ascending and descending stairs/slopes, brief walking and sedentary bouts and prolonged walking bouts, varied between subgroups. CONCLUSIONS: From this sample of KOA patients, the most common form of activity was level walking, although cycling and stair climbing activities occurred frequently, highlighting the relevance of distinguishing between these types of PA. The total active time encompassed a small portion of their waking hours, as they spent most of their time sedentary, which was exacerbated by frequently occurring prolonged bouts. In this study, event-based parameters, such as stair climbing or short bouts of walking or sedentary time, were found more capable of discriminating between subgroups of KOA patients compared to overall levels of PA and sedentary time. Thereby, subtle limitations in physical behaviour of KOA-subgroups were revealed, which might ultimately be targeted in rehabilitation programs. TRIAL REGISTRATION: German Clinical Trials Registry under ' DRKS00008735 ' at 02.12.2015.

Clinical validation of a body-fixed 3D accelerometer and algorithm for activity monitoring in orthopaedic patients.

BACKGROUND/OBJECTIVE: Activity is increasingly being recognized as a highly relevant parameter in all areas of healthcare for diagnosis, treatment, or outcome assessment, especially in orthopaedics where the movement apparatus is directly affected. Therefore, the aim of this study was to develop, describe, and clinically validate a generic activity-monitoring algorithm, satisfying a combination of three criteria. The algorithm must be able to identify, count, and time a large set of relevant daily activities. It must be validated for orthopaedic patients as well as healthy individuals, and the validation must be in a setting that mimics free-living conditions. METHODS: Using various technical solutions, such as a dual-axis approach, dynamic inclinometry (hip flexion), and semiautomatic calibration (gait speed), the algorithms were designed to count and time the following postures, transfers, and activities of daily living: resting/sitting, standing, walking, ascending and descending stairs, sit-stand transitions, and cycling. In addition, the number of steps per walking bout was determined. Validation was performed with healthy individuals and patients who had undergone unilateral total joint arthroplasty, representing a wide spectrum of functional capacity. Video observation was used as the gold standard to count and time activities in a validation protocol approaching free-living conditions. RESULTS: In total 992 and 390 events (activities or postures) were recorded in the healthy group and patient group, respectively. The mean error varied between 0% and 2.8% for the healthy group and between 0% and 7.5% for the patient group. The error expressed in percentage of time varied between 2.0% and 3.0% for both groups. CONCLUSION: Activity monitoring of orthopaedic patients by counting and timing a large set of relevant daily life events is feasible in a user- and patient-friendly way and at high clinical validity using a generic three-dimensional accelerometer and algorithms based on empirical and physical methods. The algorithms performed well for healthy individuals as well as patients recovering after total joint replacement in a challenging validation set-up. With such a simple and transparent method real-life activity parameters can be collected in orthopaedic practice for diagnostics, treatments, outcome assessment, or biofeedback.

Validation of a novel activity monitor in impaired, slow-walking, crutch-supported patients.

BACKGROUND: A growing need in clinical practice of rehabilitation and orthopaedic medicine is for objective outcome tools to estimate physical activity. Current techniques show limited validity or are too demanding for routine clinical use. Accelerometer-based activity monitors (AMs) have shown promise for measuring physical activity in healthy people but lack validity in impaired patients. OBJECTIVES: This study aimed to validate an accelerometer-based AM in impaired, slow-walking, crutch-supported patients after total joint arthroplasty (TJA). METHODS: Shortly after TJA, patients who were safely mobilized with 2 crutches and 8 healthy participants completed a trial of different activities while wearing the AM on the lateral upper leg and being videotaped. Outcome variables (e.g., time walking, number of gait cycles, sit-stand-sit transfers) were compared to video recordings, and sensitivity, predictive value and mean percentage difference (MPD) values were calculated. RESULTS: We included 40 patients (mean age: 65±9 years; mean BMI: 30±6kg/m2; male:female ratio: 18:22) and 8 healthy participants (mean age: 49±20 years; mean BMI: 23±0.7kg/m2; male:female ratio: 5:3). The AM showed excellent sensitivity (>95%) and predictive value (>95%) in identifying activities (e.g., walking, sitting, resting) and detecting the number of gait cycles and sit-stand-sit transfers (mean percentage difference: ±2%). Detection of number of steps ascending and descending stairs and cadence was more difficult but still showed good results (mean percentage difference: ±7%). CONCLUSIONS: This is the first validation study to assess physical activity with an AM in impaired, slow-walking, crutch-supported patients. The AM was a valid tool for measuring physical activity in these patients. The tool may help in evaluating and optimizing rehabilitation programs for patients after TJA, those recovering from stroke or chronic impaired patients.

Quantifying Habitual Levels of Physical Activity According to Impact in Older People: Accelerometry Protocol for the VIBE Study.

Physical activity (PA) may need to produce high impacts to be osteogenic. The aim of this study was to identify threshold(s) for defining high impact PA for future analyses in the VIBE (Vertical Impact and Bone in the Elderly) study, based on home recordings with triaxial accelerometers. Recordings were obtained from 19 Master Athlete Cohort (MAC; mean 67.6 years) and 15 Hertfordshire Cohort Study (HCS; mean 77.7 years) participants. Data cleaning protocols were developed to exclude artifacts. Accelerations expressed in g units were categorized into three bands selected from the distribution of positive Y-axis peak accelerations. Data were available for 6.6 and 4.4 days from MAC and HCS participants respectively, with approximately 14 hr recording daily. Three-fold more 0.5-1.0g impacts were observed in MAC versus HCS, 20-fold more 1.0-1.5g impacts, and 140-fold more impacts ≥ 1.5g. Our analysis protocol successfully distinguishes PA levels in active and sedentary older individuals.

Frontal plane pelvic motion during gait captures hip osteoarthritis related disability.

Gait analysis has widely been accepted as an objective measure of function and clinical outcome. Ambulatory accelerometer-based gait analysis has emerged as a clinically more feasible alternative to optical motion capture systems but does not provide kinematic characterisation to identify disease dependent mechanisms causing walking disability. This study investigated the potential of a single inertial sensor to derive frontal plane motion of the pelvis (i.e. pelvic obliquity) and help identify hip osteoarthritis (OA) related gait alterations. Patients with advanced unilateral hip OA (n = 20) were compared to patients with advanced unilateral knee OA (n = 20) and to a healthy control group (n = 20). Kinematic characterisation of frontal plane pelvic motion during gait demonstrated decreased range of motion and increased asymmetry for hip OA patients specifically.

Physical Activity and Bone: May the Force be with You.

Physical activity (PA) is thought to play an important role in preventing bone loss and osteoporosis in older people. However, the type of activity that is most effective in this regard remains unclear. Objectively measured PA using accelerometers is an accurate method for studying relationships between PA and bone and other outcomes. We recently used this approach in the Avon Longitudinal Study of Parents and Children (ALSPAC) to examine relationships between levels of vertical impacts associated with PA and hip bone mineral density (BMD). Interestingly, vertical impacts >4g, though rare, largely accounted for the relationship between habitual levels of PA and BMD in adolescents. However, in a subsequent pilot study where we used the same method to record PA levels in older people, no >4g impacts were observed. Therefore, to the extent that vertical impacts need to exceed a certain threshold in order to be bone protective, such a threshold is likely to be considerably lower in older people as compared with adolescents. Further studies aimed at identifying such a threshold in older people are planned, to provide a basis for selecting exercise regimes in older people which are most likely to be bone protective.

Dynamic contrast-enhanced MR imaging of carotid atherosclerotic plaque: model selection, reproducibility, and validation.

PURPOSE: To compare four known pharmacokinetic models for their ability to describe dynamic contrast material-enhanced magnetic resonance (MR) imaging of carotid atherosclerotic plaques, to determine reproducibility, and to validate the results with histologic findings. MATERIALS AND METHODS: The study was approved by the institutional medical ethics committee. Written informed consent was obtained from all patients. Forty-five patients with 30%-99% carotid stenosis underwent dynamic contrast-enhanced MR imaging. Plaque enhancement was measured at 16 time points at approximately 25-second image intervals by using a gadolinium-based contrast material. Pharmacokinetic parameters (volume transfer constant, K(trans); extracellular extravascular volume fraction, v(e); and blood plasma fraction, v(p)) were determined by fitting a two-compartment model to plaque and blood gadolinium concentration curves. The relative fit errors and parameter uncertainties were determined to find the most suitable model. Sixteen patients underwent imaging twice to determine reproducibility. Carotid endarterectomy specimens from 16 patients who were scheduled for surgery were collected for histologic validation. Parameter uncertainties were compared with the Wilcoxon signed rank test. Reproducibility was assessed by using the coefficient of variation. Correlation with histologic findings was evaluated with the Pearson correlation coefficient. RESULTS: The mean relative fit uncertainty (±standard error) for K(trans) was 10% ± 1 with the Patlak model, which was significantly lower than that with the Tofts (20% ± 1), extended Tofts (33% ± 3), and extended graphical (29% ± 3) models (P < .001). The relative uncertainty for v(p) was 20% ± 2 with the Patlak model and was significantly higher with the extended Tofts (46% ± 9) and extended graphical (35% ± 5) models (P < .001). The reproducibility (coefficient of variation) for the Patlak model was 16% for K(trans) and 26% for v(p). Significant positive correlations were found between K(trans) and the endothelial microvessel content determined on histologic slices (Pearson ρ = 0.72, P = .005). CONCLUSION: The Patlak model is most suited for describing carotid plaque enhancement. Correlation with histologic findings validated K(trans) as an indicator of plaque microvasculature, and the reproducibility of K(trans) was good.