Back to results list
Please use this identifier to cite or link to this item:
|Title:||Fractal analysis on movement variability in spinal curvature||Authors:||Lau, Man Lung||Advisors:||Choy, Clifford (SD)||Keywords:||Human mechanics.
Spine -- Movements.
|Issue Date:||2016||Publisher:||The Hong Kong Polytechnic University||Abstract:||Human posture and movement sensing has become a crucial practice related to health monitoring, no matter for adults or children, especially with the introduction of various types of wearable technology in recent five years. The human movement acquisition process has become portable compared to traditional laboratory conditions. Although the business of wearable technology has an upward trend for the near future, research on human movement and the design requirements and criteria for using wearable technology remains limited. Although it looks easy to sit in the upright position, numerous researchers have tried to understand the complexity of human posture behind. In static conditions, the human upright posture exhibits an everlasting oscillatory behavior of complex nature, called the postural sway. Variability exists in the movement patterns. The exploration of the dynamics, physiology, and sensory motor control of human posture plays a crucial role in numerous areas of science. It is about the understanding of physiological signals generated by human movement. The understanding can then be applied to health, clinical, and medical applications. Issues related to pain are essential topics among those major applications. Through the understanding of human movement, preventive or monitoring measures are possible. Physiological signals have been extensively investigated to produce data that enable various types of analysis. These analysis techniques can then be designed for use by consumers to manage their health through monitoring applications by using wearable sensing devices. The conventional practice in analyzing human movement through the application of wearable sensing technology is to use descriptive statistical methods to describe the data. The methods summarize data samples, for example, central tendency and dispersion, rather than use the data to obtain further understanding. However, without the understanding of interdependency and the short- or long-range correlation of human movement, the conventional practice is inadequate to explain the complexity of physiological signals. The analytical approach has also been used to reveal the complexity of the physiological system. This approach identifies appropriate sub systems to separate the physiological system into smaller components to reveal parts of the structure to understand it. Some examples of this approach include inverted pendulum modeling and impulsive muscle control. However, a complete representation of the whole complex physiological system is hardly possible. By contrast, many of these studies rely on invasive instruments, for example, X-ray, to extract measurements or data from the inner body structure of participants. The obvious disadvantage is the harmful effects of radiation when human bodies are exposed to these instruments. The other limitation is the time-varying factor that these instruments are unable to capture. By realizing that the variability of human movement is an inevitable part of the research, the need for exploring the dynamics of complex human movement becomes essential. From the theoretical point of view, this study explores the dynamics of human movement and correspondingly associated to the motor control mechanism, which is affected by the human perception in sensation, processing, and the activation of movement. This study also contributes to the insight into the practical aspects of human movement in designing and engineering sensing devices. The process involves the design of a physiologically inspired information-processing model and then transforming it into design criteria for interactive technology and applications. The objective of this study is to explore the dynamics of human movement by using a fractal and multifractal approach. In addition, the study also aims to initiate a nonanalytical framework to understand human movement. The nonanalytical approach on human movement is based on computational techniques to obtain knowledge from the movement data extracted using noninvasive optical motion capture techniques. Some examples of the computational techniques include artificial neural networks, inductive learning, and skill-based expert systems. In this study, the focus on the analysis is based on the representation of physiological signals on spinal curvature movement. Design criteria can then be derived for the wearable sensing devices targeting monitoring applications.
The research is conducted starting with the method and procedure in the collection of experimental human movement on the basis of the static upright sitting posture, and then with the transformation of data for later analysis. The process then involves the in-depth analysis of the signals from structural investigation and evaluation regarding noise-like properties. The fractal analysis is adopted to reveal the existence of fractal structure and knowledge on the fluctuation of the underlying variations in movement performance. This is the first major achievement in this study. To our knowledge, this approach is the first attempt to be adopted in the analysis of static movement in spinal curvature. The research then involves a multifractal analysis on the exploration of variation space. Multifractality structure can be identified based on the experimental set of participants. The structure is reflected by the multifractal spectrum with parameters on the scaling exponent and singularity dimension, together with the width, height, and shape. These parameters are the major research findings that describe the variation of the dynamic structure according to the movement performance. These findings are analyzed, compared, and ranked across participants within the experimental set. It can be found that there exists consistency and variation among participants. The consistency suggests that the multifractality structure is common among captured movement. The variation is applicable in differentiating various characteristics of participants. To further investigate and validate the dynamic properties of cervical movement, correlation analysis is conducted in relation to neck performance and pain issues. The Neck Pain and Disability (NPAD) scale, which is a proven neck pain instrument for clinical use, is adopted. Results show that there exist groups of properties that describe the correlation across various dimensions between the dynamic properties and the NPAD scale. The analysis findings are then described according to the variation of the level of spinal curvature movement. The variation is further associated with motor strategies and neural activities. On the basis of the findings, the implication on the design criteria with regard to wearable sensing devices for monitoring purposes is explored. Core features are identified and further explained using a recent example, Fineck. Fineck is a wearable device on the neck to track the head movement. It identifies unfavorable habits and suggests exercises through gaming experience. On the basis of the 12 attributes defined in the fundamental guidelines for designing wearable systems, the applicability of the neck movement monitoring purpose is investigated. Criteria on the motion characteristics for the design of motion-sensing devices are considered by illustrating the analysis results qualitatively through neck movement sensing. Other design challenges from the aspects of interface and interaction experience in wearable technology are also investigated. In this study, the major and original contribution is the framework developed to investigate the small and large local scale fluctuations within the temporal dimension of physiological signals along the spine. The findings contribute to the understanding of human movement. This study presents the implication of the technical results on the design of wearable technology. It includes the illustration of the design aspects of key features for interactive applications, wearability attributes, the criteria for sensing, and the experience in wearable interfaces. This research provides insights into the guidelines on how computational techniques can be used for the development of wearable design applications, with considerations of both technology and interaction design aspects.
|Description:||PolyU Library Call No.: [THS] LG51 .H577P SD 2016 Lau
xIii, 373 pages :color illustrations
|URI:||http://hdl.handle.net/10397/55237||Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b28982745_link.htm||For PolyU Users||208 B||HTML||View/Open|
|b28982745_ira.pdf||For All Users (Non-printable)||6.22 MB||Adobe PDF||View/Open|
Citations as of Jun 18, 2018
Citations as of Jun 18, 2018
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.