Evaluation of Postural Sway in Post-stroke Patients by Dynamic Time Warping Clustering

Abstract

Post-stroke complications are the second most frequent cause of death and the third leading cause of disability worldwide. The motor function of post-stroke patients is often assessed by measuring the postural sway in the patients during quiet standing, based on sway measures, such as sway area and velocity, which are obtained from temporal variations of the center of pressure. However, such approaches to establish a relationship between the sway measures and patients' demographic factors have hardly been successful (e.g., days after onset). This study instead evaluates the postural sway features of post-stroke patients using the clustering method of machine learning. First, we collected the stroke patients' multi-variable motion-capture standing-posture data and processed them into t s long data slots. Then, we clustered the t-s data slots into K cluster groups using the dynamic-time-warping partition-around-medoid (DTW-PAM) method. The DTW measures the similarity between two temporal sequences that may vary in speed, whereas PAM identifies the centroids for the DTW clustering method. Finally, we used a post-hoc test and found that the sway amplitudes of markers in the shoulder, hip, knee, and center-of-mass are more important than their sway frequencies. We separately plotted the marker amplitudes and frequencies in the medial-lateral direction during a 5-s data slot and found that the post-stroke patients' postural sway frequency lay within the bandwidth of 0.5–1.5 Hz. Additionally, with an increase in the onset days, the cluster index of cerebral hemorrhage patients gradually transits in a four-cluster solution. However, the cerebral infarction patients did not exhibit such pronounced transitions over time. Moreover, we found that the postural-sway amplitude increased in clusters 1, 3, and 4. However, the amplitude of cluster 2 did not follow this pattern, owing to age effects related to the postural sway changes with age. A rehabilitation doctor can utilize these findings as guidelines to direct the post-stroke patient training.

Normalization and possibility of classification analysis using the optimal warping paths of dynamic time warping in gait analysis

“…Therefore, research related to this should be conducted promptly. If further research utilizing DTW is conducted across various age groups as well as for comparative studies between healthy individuals and patients such as Li et al (2021), Pulido-Valdeolivas et al (2018, Varatharajan et al (2018), it is probable that DTW will be utilized as a replacement for traditional indicators in the evaluation of gait pattern similarity.…”

Discrete cortical control during quiet stance revealed by desynchronization and rebound of beta oscillations

“…methodology [31,32], while we narrowed the selection-criteria of the micro fall events for the event-locked average only to those well-separated to the subsequent micro falls in order to avoid contaminations of EEG responses associated with adjacent overlapped cycles.…”

“…Note that length of the extracted data varied depending on the fall-recovery cycle. Then, each fallrecovery cycle data of CoM, EMG and ERSP was time-warped (i.e., either stretched or compressed along the time axis) [31,32], so that the length of the cycle became equal to the average length of the fall-recovery cycles across all cycles from all participants. Note that the time-warp was performed separately for the micro-fall-segment and the micro-recovery-segment of the cycle.…”

Deterioration of postural control due to the increase of similarity between center of pressure and smooth-pursuit eye movements during standing on one leg

“…The dynamic time warping (DTW) method is possible to quantify the similarity of two time-series data with non-linear extension and contraction allowed, even though the frequency and the number of datasets are different. Therefore, we used the DTW analysis to measure the similarity between two temporal sequences; that is, the displacement of the COP and eye movement [ 26 ]. We minimized the distance between the two temporal sequences using the DTW package in R software without band filters (Sakoe-Chiba and Itakura), because it was impossible to estimate the suitable window size for matching the COP and eye movements [ 27 ].…”