Loading...

The Development of a Postural Control Model for People with Parkinson’s disease to Predict Rehabilitation Exercises Effects

Rahmati, Zahra | 2020

535 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 53537 (08)
  4. University: Sharif University of Technology
  5. Department: Mechanical Engineering
  6. Advisor(s): Behzadipour, Saeed; Firoozbakhsh, Keikhosrow; Taghizadeh, Ghorban
  7. Abstract:
  8. Background: Parkinson’s disease (PD) patients seriously suffer from instability and impaired postural control. Rehabilitation exercises help them to recover their ability through long-term practical sessions. For designing optimal tasks in each session, and in agreement with each patient’s state, it is essential to employ computational models and approaches.Objective: The goal of this study is to develop a computational postural control model of Parkinson’s disease in order to provide new understanding of the postural control in PD, and to gain insight on the effect of balance trainings on PD (the learning dynamics), from the view of this model.Methods: In the first phase of this study, the center-of-pressure (COP) data (static posturography) of 20 healthy control subjects (HC), and a group of 40 PD patients before and after a 12-session (4-week) balance training were recorded in 4 tasks; two tasks on rigid surface (R-tasks) and two on the foam (F-tasks). First, the COP data were investigated through a new framework, based on the frequency analysis. Next, a general postural control model of human stance (with an inverse pendulum as the human body, and a PID controller with time delay as the representative of the CNS) were fitted to the COP data, in order to identify the model parameters of each subject in each task. Using the model parameters, the gain and phase margin of each subject in each task were calculated. In the second phase of this study, and in order to find out the motor learning dynamics of postural control in PD, 20 new PD patients received 18-session (6-week) balance-training program. Patients were assessed experimentally by static posturography, for four time-points during the training program, i.e. before, at week 2, week 4, and after the program; and clinically for three times (pre, mid, post). Similar to the first phase, the COP sway measures and model parameters, as well as the gain and phase margin were calculated for each patient in each task and at each time point.Results: Analysis of COP data in a novel frequency analysis framework revealed that the differences in patients’ sway measures from HC stem from two main impairments of postural control in PD, i.e. “stability” and “flexibility”. Findings showed that PD patients had up to 30% lower control parameters (stability), to 34% lower flexibility (higher axial rigidity), to 28% higher time delay, and to 34% lower gain margin; as compared to HC. These features were improved after the 12-session balance-training program (to 23.5% in flexibility, 10% in stability, 24.6% in gain margin, and 15% in time delay). Moreover, the model provided a framework to quantitatively disentangle two main impaired features of postural control in PD, i.e. ‘stability’ and ‘flexibility’ degrees in each patient. In the second phase of this study, results showed that the poor ‘stability’, as well as low gain margin in patients improved faster (at week 4), and then plateaued out in the rest of the training. On the other hand, the ‘flexibility’ improved continuously during the training, which achieved significant improvement, late (at week 6).Conclusion: This thesis presents the first-ever (to our knowledge) clinically-applicable, computational postural control model of PD patients, with the potential to disentangle ‘stability’ and ‘flexibility’. Furthermore, studying the learning dynamics of postural control over a course of balance-training program provided a systematic approach for future evaluation of different training programs. Thus, the findings of this thesis substantially contribute to the future development of novel, patient-specific, computational ‘rehabilitative models’ of postural control for PD
  9. Keywords:
  10. Parkinson Disease ; Rehabilitation Exercises ; Posture Control ; Balance Training ; Stability Degree ; Flexibility Degree ; Center-of-Pressure (COP)Sway Measures ; Learning Dynamics

 Digital Object List

 Bookmark

No TOC