Complexity-Based Measures of Postural Sway during Walking at Different Speeds and Durations Using Multiscale Entropy

Ben Yi Liau, Fu Lien Wu, Chi Wen Lung, Xueyan Zhang, Xiaoling Wang, Yih Kuen Jan

Research output: Contribution to journalArticlepeer-review


Participation in various physical activities requires successful postural control in response to the changes in position of our body. It is important to assess postural control for early detection of falls and foot injuries. Walking at various speeds and for various durations is essential in daily physical activities. The purpose of this study was to evaluate the changes in complexity of the center of pressure (COP) during walking at different speeds and for different durations. In this study, a total of 12 participants were recruited for walking at two speeds (slow at 3 km/h and moderate at 6 km/h) for two durations (10 and 20 minutes). An insole-type plantar pressure measurement system was used to measure and calculate COP as participants walked on a treadmill. Multiscale entropy (MSE) was used to quantify the complexity of COP. Our results showed that the complexity of COP significantly decreased (p < 0.05) after 20 min of walking (complexity index, CI = -3.51) compared to 10 min of walking (CI = -3.20) while walking at 3 km/h, but not at 6 km/h. Our results also showed that the complexity index of COP indicated a significant difference (p < 0.05) between walking at speeds of 3 km/h (CI = -3.2) and 6 km/h (CI = -3.6) at the walking duration of 10 minutes, but not at 20 minutes. This study demonstrated an interaction between walking speeds and walking durations on the complexity of COP.

Original languageEnglish (US)
Article number1128
Issue number11
StatePublished - Nov 16 2019


  • Center of pressure
  • Complexity
  • Falls
  • Multiscale entropy
  • Postural control

ASJC Scopus subject areas

  • Physics and Astronomy(all)


Dive into the research topics of 'Complexity-Based Measures of Postural Sway during Walking at Different Speeds and Durations Using Multiscale Entropy'. Together they form a unique fingerprint.

Cite this