Hip osteoarthritis patients exhibit changes in kinematics and kinetics that affect joint loading. Monitoring this load can provide valuable information to clinicians. For example, a patient's joint loading measured across different activities can be used to determine the amount of exercise that the patient needs to complete each day. Unfortunately, current methods for measuring joint loading require a lab environment which most clinicians do not have access to. This study explores employing machine learning to construct a model that can estimate joint loading based on sensor data obtained solely from a mobile phone. In order to learn such a model, we collected a dataset from 10 patients with hip osteoarthritis who performed multiple repetitions of nine different exercises. During each repetition, we simultaneously recorded 3D motion capture data, ground reaction force data, and the inertial measurement unit data from a mobile phone attached to the patient's hip. The 3D motion and ground reaction force data were used to compute the ground truth joint loading using musculoskeletal modeling. Our goal is to estimate the ground truth loading value using only the data captured by the sensors of the mobile phone. We propose a machine learning pipeline for learning such a model based on the recordings of a phone's accelerometer and gyroscope. When evaluated for an unseen patient, the proposed pipeline achieves a mean absolute error of 29% for the left hip and 36% for the right hip. While our approach is a step in the direction of using a minimal number of sensors to estimate joint loading outside the lab, developing a tool that is accurate enough to be applicable in a clinical context still remains an open challenge. It may be necessary to use sensors at more than one location in order to obtain better estimates.
It has been hypothesized that internal oscillations can synchronize (i.e., entrain) to external environmental rhythms, thereby facilitating perception and behaviour. To date, evidence for the link between the phase of neural oscillations and behaviour has been scarce and contradictory; moreover, it remains an open question whether the brain can use this tentative mechanism for active temporal prediction. In our present study, we conducted a series of auditory pitch discrimination tasks with 181 healthy participants in an effort to shed light on the proposed behavioural benefits of rhythmic cueing and entrainment. In the three versions of our task, we observed no perceptual benefit of purported entrainment: targets occurring in‐phase with a rhythmic cue provided no perceptual benefits in terms of discrimination accuracy or reaction time when compared with targets occurring out‐of‐phase or targets occurring randomly, nor did we find performance differences for targets preceded by rhythmic versus random cues. However, we found a surprising effect of cueing frequency on reaction time, in which participants showed faster responses to cue rhythms presented at higher frequencies. We therefore provide no evidence of entrainment, but instead a tentative effect of covert active sensing in which a faster external rhythm leads to a faster communication rate between motor and sensory cortices, allowing for sensory inputs to be sampled earlier in time.
Inertial capture (InCap) systems combined with musculoskeletal (MSK) models are an attractive option for monitoring 3D joint kinematics in an ecological context. However, the primary limiting factor is the sensor-to-segment calibration, which is crucial to estimate the body segment orientations. Walking, running, and stair ascent and descent trials were measured in eleven healthy subjects with the Xsens InCap system and the Vicon 3D motion capture (MoCap) system at a self-selected speed. A novel integrated method that combines previous sensor-to-segment calibration approaches was developed for use in a MSK model with three degree of freedom (DOF) hip and knee joints. The following were compared: RMSE, range of motion (ROM), peaks, and R2 between InCap kinematics estimated with different calibration methods and gold standard MoCap kinematics. The integrated method reduced the RSME for both the hip and the knee joints below 5°, and no statistically significant differences were found between MoCap and InCap kinematics. This was consistent across all the different analyzed movements. The developed method was integrated on an MSK model workflow, and it increased the sensor-to-segment calibration accuracy for an accurate estimate of 3D joint kinematics compared to MoCap, guaranteeing a clinical easy-to-use approach.
This study’s aim is threefold: (I) Evaluate movement quality parameters of gait in people with hip or knee osteoarthritis (OA) compared to asymptomatic controls from a single trunk-worn 3D accelerometer. (II) Evaluate the sensitivity of these parameters to capture changes at 6-weeks, 3-, 6-, and 12-months following total knee arthroplasty (TKA). (III) Investigate whether observed changes in movement quality from 6-weeks and 12-months post-TKA relates to changes in patient-reported outcome measures (PROMs). We invited 20 asymptomatic controls, 20 people with hip OA, 18 people pre- and post-TKA to our movement lap. They wore a single trunk-worn accelerometer and walked at a self-selected speed. Movement quality parameters (symmetry, complexity, smoothness, and dynamic stability) were calculated from the 3D acceleration signal. Between groups and between timepoints comparisons were made, and changes in movement quality were correlated with PROMs. We found significant differences in symmetry and stability in both OA groups. Post-TKA, most parameters reflected an initial decrease in movement quality at 6-weeks post-TKA, which mostly normalised 6-months post-TKA. Finally, improved movement quality relates to improvements in PROMs. Thus, a single accelerometer can characterise movement quality in both OA groups and post-TKA. The correlation shows the potential to monitor movement quality in a clinical setting to inform objective, data-driven personalised rehabilitation.
(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.
18 January 2022 in Journal of Orthopaedic Research