Wearable sensors have the potential to provide clinicians with access to motor performance of people with movement disorder as they undergo intervention. However, sensor data often have to be manually classified and segmented before they can be processed into clinical metrics. This process can be time consuming. We recently proposed detection and segmentation algorithms based on peak detection using Inertial Measurement Units (IMUs) to automatically identify and isolate common activities during daily living such as standing up, walking, turning, and sitting down. These algorithms were developed using a homogenous population of healthy older adults. The aim of this study was to investigate the transferability of these algorithms in people with Parkinson’s disease (PD). A modified Timed Up And Go task was used since it is comprised of these activities, all performed in a continuous fashion. Twelve older adults diagnosed with early PD (Hoehn & Yahr ≤ 2) were recruited for the study and performed three trials of a 10 and 5-m TUG during OFF state. They were outfitted with 17 IMUs covering each body segment. Raw data from IMUs were detrended, normalized and filtered to reveal kinematics peaks that corresponded to different activities. Segmentation was accomplished by identifying the first minimum or maximum to the right and the left of these peaks. Segmentation times were compared to results from two examiners who visually segmented the activities. Specificity and sensitivity were used to evaluate the accuracy of the detection algorithms. Using the same IMUs and algorithms developed in the previous study, we were able to detect these activities with 97.6% sensitivity and 92.7% specificity (n = 432) in PD population. However, with modifications to the IMUs selection, we were able to detect these activities with 100% accuracy. Similarly, applying the same segmentation to PD population, we were able to isolate these activities within ~500 ms of the visual segmentation. Re-optimizing the filtering frequencies, we were able to reduce this difference to ~400 ms. This study demonstrates the agility and transferability of using a system of IMUs to accurately detect and segment activities in daily living in people with movement disorders.

Joints kinematics assessment based on inertial measurement systems, which include attitude and heading reference system (AHRS), are quickly gaining in popularity for research and clinical applications. The variety of the tasks and contexts they are used in require a deep understanding of the AHRS accuracy for optimal data interpretation. However, published accuracy studies on AHRS are mostly limited to a single task measured on a limited number of segments and participants. This study assessed AHRS sensors kinematics accuracy at multiple segments and joints through a variety of tasks not only to characterize the system’s accuracy in these specific conditions, but also to extrapolate the accuracy results to a broader range of conditions using the characteristics of the movements (i.e. velocity and type of motion). Twenty asymptomatic adults ( $$\overline{age}$$ = 49.9) performed multiple 5 m timed up and go. Participants’ head, upper trunk, pelvis, thigh, shank and foot were simultaneously tracked using AHRS and an optical motion capture system (gold standard). Each trial was segmented into basic tasks (sit-to-stand, walk, turn). At segment level, results revealed a mean root-mean-squared-difference $$\overline{(RMSD)}$$ varying between 1.1° and 5.5° according to the segment tracked and the task performed, with a good to excellent agreement between the systems. Relative sensor kinematics accuracy (i.e. joint) varied between 1.6° and 13.6° over the same tasks. On a global scheme, analysis of the effect of velocity on sensor kinematics accuracy showed that AHRS are better adapted to motions performed between 50°/s and 75°/s (roughly thigh and shank while walking). Results confirmed that pairing of modules to obtain joint kinematics affects the accuracy compared to segment kinematics. Overall, AHRS are a suitable solution for clinical evaluation of biomechanics under the multi-segment tasks performed although the variation in accuracy should be taken into consideration when judging the clinical meaningfulness of the observed changes.

Head pose assessment can reveal important clinical information on human motor control. Quantitative assessment have the potential to objectively evaluate head pose and movements’ specifics, in order to monitor the progression of a disease or the effectiveness of a treatment. Optoelectronic camera-based motion-capture systems, recognized as a gold standard in clinical biomechanics, have been proposed for head pose estimation. However, these systems require markers to be positioned on the person’s face which is impractical for everyday clinical practice. Furthermore, the limited access to this type of equipment and the emerging trend to assess mobility in natural environments support the development of algorithms capable of estimating head orientation using off-the-shelf sensors, such as RGB cameras. Although artificial vision is a popular field of research, limited validation of human pose estimation based on image recognition suitable for clinical applications has been performed. This paper first provides a brief review of available head pose estimation algorithms in the literature. Current state-of-the-art head pose algorithms designed to capture the facial geometry from videos, OpenFace 2.0, MediaPipe and 3DDFA_V2, are then further evaluated and compared. Accuracy is assessed by comparing both approaches to a baseline, measured with an optoelectronic camera-based motion-capture system. Results reveal a mean error lower or equal to $5.6^{\circ }$ for 3DDFA_V2 depending on the plane of movement, while the mean error reaches $14.1^{\circ }$ and $11.0^{\circ }$ for OpenFace 2.0 and MediaPipe, respectively. This demonstrates the superiority of the 3DDFA_V2 algorithm in estimating head pose, in different directions of motion, and suggests that this algorithm can be used in clinical scenarios.

Clinical mobility assessment is traditionally performed in laboratories using complex and expensive equipment. The low accessibility to such equipment, combined with the emerging trend to assess mobility in a free-living environment, creates a need for body-worn sensors (e.g., inertial measurement units—IMUs) that are capable of measuring the complexity in motor performance using meaningful measurements, such as joint orientation. However, accuracy of joint orientation estimates using IMUs may be affected by environment, the joint tracked, type of motion performed and velocity. This study investigates a quality control (QC) process to assess the quality of orientation data based on features extracted from the raw inertial sensors’ signals. Joint orientation (trunk, hip, knee, ankle) of twenty participants was acquired by an optical motion capture system and IMUs during a variety of tasks (sit, sit-to-stand transition, walking, turning) performed under varying conditions (speed, environment). An artificial neural network was used to classify good and bad sequences of joint orientation with a sensitivity and a specificity above 83%. This study confirms the possibility to perform QC on IMU joint orientation data based on raw signal features. This innovative QC approach may be of particular interest in a big data context, such as for remote-monitoring of patients’ mobility.

Gait assessment is of interest to clinicians and researchers because it provides information about patients’ functional mobility. Optoelectronic camera-based systems with gait event detection algorithms are considered the gold standard for gait assessment. Yet, the choice of the algorithm used to process data and extract the desired parameters from those detected gait events has an impact on the validity and reliability of the gait parameters computed. There are multiple techniques documented in the literature for computing gait events, including the analysis of the minimal position of the heel and toe markers, the computation of the relative distance between sacrum and foot markers, and the assessment of the smallest distance between the heel and toe markers. Validation studies conducted on these algorithms report variations in accuracy. Yet, these studies were conducted in different conditions, at varying gait velocities, and on different populations. The purpose of this study is to compare accuracy, precision, and robustness of three algorithms using motion capture data obtained from 25 healthy persons and 21 psoriatic arthritic patients walking at three distinct speeds on an instrumented treadmill. Errors in gait events recognition (heel strike—HS and toe-off—TO) and their impact on gait metrics (stance phase and stride length) are reported and compared to ground reaction force events measured with force plates. Over the 9114 collected steps across all walking speeds, more than 99% of gait events were recognized by all algorithms. On average, HS events were detected within 1.2 ms of the reference for two algorithms, while the third one detected HS late, with an average detection error of 40.7 ms. Yet, significant variations in accuracy were noted with gait speed; the performance decreased for all algorithms at slow speed. TO events were identified early by all algorithms, with an average error ranging from 16.0 to 100.0 ms. These gait events errors lead to 2–15% inaccuracies in stance phase assessment, while the impact on stride length remains below 0.3 cm. Overall, the algorithm based on the relative distance between the sacral and foot markers stood out for its accuracy, precision, and robustness at all walking speeds.

A significant amount of research has highlighted the importance of a motor component in the brain’s processing of emotional, motivational and social information. Posturography has emerged as an interesting way to assess motor correlates associated with this process. In this review, we highlight recent results within the functional context of painful stimulus perception and discuss the interest in broadening the use of posturography to other motivational and societal functional contexts. Although characterized by significant feasibility, the single measurement of the COP’s anteroposterior displacement presents limitations for attesting approach–avoidance behavior towards a visual target. Here, we discuss a number of methodological avenues that could go some way towards overcoming these limitations.