Beyond timing and step counting in 360° turning-in-place assessment: a scoping review

Abstract

Background Turning in place is a challenging motor task and is used as a brief assessment test of lower limb function and dynamic balance. This review aims to examine how research of instrumented analysis of turning in place is implemented. In addition to reporting the studied population, we covered acquisition systems, turn detection methods, quantitative parameters, and how these parameters are computed. Methods Following the development of a rigorous search strategy, the Web of Science and Scopus were systematically searched for studies involving the use of turning-in-place. From the selected articles, the study population, types of instruments used, turn detection method, and how the turning-in-place characteristics were calculated. Results Twenty-one papers met the inclusion criteria. The subject groups involved in the reviewed studies included young, middle-aged, and older adults, stroke, multiple sclerosis and Parkinson’s disease patients. Inertial measurement units (16 studies) and motion camera systems (5 studies) were employed for gathering measurement data, force platforms were rarely used (2 studies). Two studies used commercial software for turn detection, six studies referenced previously published algorithms, two studies developed a custom detector, and eight studies did not provide any details about the turn detection method. The most frequently used parameters were mean angular velocity (14 cases, 7 studies), turn duration (13 cases, 13 studies), peak angular velocity (8 cases, 8 studies), jerkiness (6 cases, 5 studies) and freezing-of-gait ratios (5 cases, 5 studies). Angular velocities were derived from sensors placed on the lower back (7 cases, 4 studies), trunk (4 cases, 2 studies), and shank (2 cases, 1 study). The rest (9 cases, 8 studies) did not report sensor placement. Calculation of the freezing-of-gait ratio was based on the acceleration of the lower limbs in all cases. Jerkiness computation employed acceleration in the medio-lateral (4 cases) and antero-posterior (1 case) direction. One study did not reported any details about jerkiness computation. Conclusion This review identified the capabilities of turning-in-place assessment in identifying movement differences between the various subject groups. The results, based on data acquired by inertial measurement units across studies, are comparable. A more in-depth analysis of tests developed for gait, which has been adopted in turning-in-place, is needed to examine their validity and accuracy.

Wearable sensor devices can automatically identify the ON-OFF status of patients with Parkinson's disease through an interpretable machine learning model

“…However, in our study, the Gait: Peak Horiz. Trunk Velocity (degrees/s) [Mean] and Turn: Peak Velocity (degrees/s) [Mean] were less important than the lower limb motor features, the turn-related motor features were considered the most important factors to distinguish between patients with PD and healthy elderly people ( 36 , 37 ). Similar to previous studies, patients with PD turned more slowly into the “OFF” status.…”

“…Similar to previous studies, patients with PD turned more slowly into the “OFF” status. Turning was a complex act that required everyone to change direction while keeping the trunk stable ( 37 ). The patients with PD seemed to have more difficulty turning because they could not precisely control the RoM of the lower limb joint, stride length, and peak velocity of the shank ( 37 ).…”

“…Turning was a complex act that required everyone to change direction while keeping the trunk stable ( 37 ). The patients with PD seemed to have more difficulty turning because they could not precisely control the RoM of the lower limb joint, stride length, and peak velocity of the shank ( 37 ). In addition, previous studies suggested that the efficacy of levodopa in improving trunk-related symptoms in patients with PD was limited ( 38 , 39 ).…”