Wearable Real-Time Feedback System to Improve Gait and Posture in Parkinson's Disease

Wearable Real-Time Feedback System to Improve Gait and Posture in Parkinson's Disease

Progressive gait dysfunction is one of the main motor symptoms in people with Parkinson's disease (PD). It is generally expressed as reduced step length and gait speed, and as increased variability in step time and length. People with PD also exhibit stooped posture, which besides apparent disfigurement, also disrupts gait. The gait and posture impairments are usually resistant to the pharmacological treatment, worsen as the disease progresses, increase the likelihood of falls, and result in higher rates of hospitalization and mortality. These impairments may be caused by perceptual (spatial awareness) difficulties due to deficiency in processing information related to movement initiation and execution, which can result in misperceptions of the actual effort required to perform a desired movement and posture. Due to this, people with PD often depend on external cues during motor tasks.

Although numerous studies have shown that cues can improve gait in PD, they did not provide feedback of the performance in real-time which is crucial to perceive, modulate, and achieve the desired movements. There are a few studies that provided real-time feedback using treadmill-based systems and observed improvements in gait in PD, however, they are not suitable for practicing target movements conveniently during free-living conditions, which can strongly reinforce movement patterns and improve clinical outcomes. There has been very little investigations of wearable real-time feedback (WRTF) systems to improve gait and posture in PD. The investigators are aware of only one study that tried to improve gait using a wearable system with real-time feedback capabilities, but the study did not provide any feedback on posture. Also, some of the parameters used for feedback were not easy to perceive and modulate in real-time.

Based on the investigators' recent success with a treadmill-based real-time feedback system which improved gait and posture in people with PD, the proposed study will develop a WRTF system, validate its performance with gold standard measures from a motion capture system, and test its feasibility in a group of people with mild to moderate PD. The most novel aspects of the proposed system are that it will provide feedback on parameters such as step length, arm swing, step time, and upright posture which have been greatly affected in PD and shown to increase the risk factors for balance disorders and falls. In addition, the system will consists of two types of feedback: a Continuous Feedback (CF) mode and an On-Demand Feedback (ODF) mode. The CF mode will help users learn and practice desired gait and posture movements and the ODF mode will help to maintain them during activities of daily living. The gait and posture performances during feedback and non-feedback conditions will be compared and, if the expected benefits are observed, a follow-up randomized clinical trial will be performed to test the effectiveness of this novel technology during daily activities.

After validation of the proposed wearable real-time feedback (WRTF) system, the feasibility of utilizing this system to provide RTF of gait and posture parameters will be tested in a single group of twenty-six people with mild to moderate Parkinson's disease (PD). The sample size was chosen to observe an intra-class correlation coefficient (ICC (2,1)) value between 0.75 and 0.9 at alpha = 0.05 and power of 80%. Subjects of all races and both genders will be recruited. Despite higher representation of males in PD, the investigators will seek to achieve an equal distribution of males and females. As the incidence of PD increases with age, rising after the age of 55 years with a sharp increase after 60 years, the investigators will recruit participants between the ages of 50-80 years.

After enrollment, each subject will participate in one screening and clinical evaluation session at Muhammad Ali Parkinson Center (MAPC) at St. Joseph's Hospital and Medical Center (SJHMC) in Phoenix, AZ and all the experimental data collection sessions will be performed at the Arizona State University (ASU), Tempe, AZ. The 3 experimental data collection sessions will be (1) Validity Session, (2) Continuous feedback (CF) Mode Session, and (3) On-demand feedback (ODF) Mode session. All the screening, clinical evaluation, experimental data collection sessions will be carried out during subject's "medication-on" state (when the medication best controls symptoms of PD, usually 60-90 minutes after the subjects' usual dose of anti-parkinsonian medication). The order of CF and ODF mode sessions will be randomized across the subjects.

Screening and Clinical Evaluation Session (~ 2 hours): In this session, first the subjects will be screened by Dr. Shill for the eligibility criteria which includes UPDRS evaluation. If they are eligible, their balance and cognitive status will be evaluated using the Mini-Balance Evaluation Systems Test (Mini-BESTest) and the Scales for Outcomes of Parkinson' Disease-Cognition (SCOPA-Cog) respectively. The Mini-BESTest has been shown to detect even subtle balance deficits in people with PD. SCOPA-Cog is a PD-specific scale that tests non-verbal and verbal memory, learning, attention, and executive function, including complex motor planning, working memory, and verbal fluency and has undergone extensive clinimetric testing.

WRTF Concurrent Validity Session (SA1) (~ 2 hours): In this session, MIMU sensors will be placed on the lower legs, wrists, lower back, and upper back. Lightweight reflective markers will be placed according to the full-body marker setup (Vicon Motion Systems, Ltd, UK) and specialized software (Nexus 2.3, Vicon) will be used for kinematic assessment. Participants will be asked to complete twenty 15-meter overground walking trials at their self-selected comfortable speed. From the steady state segments of these trials, the gait events, spatio-temporal gait indices (including step length, step time, arm swing, and back angle) will be calculated in real-time for each step using the WRTF system (as described in earlier sections), and will be compared against the gold standard values obtained from Vicon system.

Continuous Feedback Mode Session (SA2) (~ 3 hours): In this session, RTF will be provided for the participant to practice walking with increased step length, arm swing, upright posture and regular step time. All the walking trials will be performed overground and feedback of only one parameter will be provided at a given time/trial. Various step-to-step spatio-temporal gait and posture measures will be obtained using the WRTF system.

First, the participants will be asked to walk at their self-selected speed for one 80-meter trial (with a 180 degree turn at 40-meter) in a long hallway to determine their baseline gait measures (including step length, arm swing, step time, and back angle). Then, the subjects will perform one 80-meter trial stepping on visual cues - strips placed on the floor at equal intervals, to obtain the step length target template (described in the 'Continuous Feedback Mode' section). Next, the subjects will be asked to walk two 80-meter trials during which they will be provided with RTF of their step length performance (their current movements in a sound form and specific instructions through conducting ear phones). Following this, participants will perform one 80-meter trial following a metronome to obtain the step time target template which will be followed by two 80-meter trials during which RTF of their step time will be provided to improve gait rhythmicity. Similarly, the trials for arm swing and upright posture will be performed as follows: one 80-meter trial (walking with bigger arm swings) to obtain the arm swing target template, two 80-meter trials providing RTF of arm swing performance, one 80-meter trial (walking upright) to obtain the uprightness target template, and two 80-meter trials providing RTF of upright posture.

On-Demand Feedback Mode Session (SA3) (~ 3 hours): This session will evaluate the use of on-demand feedback. After wearing the sensors of the WRTF system and ear phones, the subjects will perform one baseline 100-meter trial without feedback at their self-selected comfortable speed. Then, the subjects will participate in eight 100-meter trials with a 180 degree turn at 50-meters, of which, two trials will be performed with each type of feedback (step length, step time, arm swing, or upright posture). Spatio-temporal gait indices will be calculated from non-feedback and feedback trials. Sufficient rest periods will be provided between the trials.

Primary and Secondary Outcome Measures: The main primary outcomes will be intra-class correlation (ICC) and reliability coefficients (RC) to test the validity of the WRTF system. The other primary measures will be step length and time and their variability, arm swing, and upright posture. Also, correlations between cognitive status and balance control and ability to follow RTF will be explored.

Data and Statistical Analyses: For gait, mean values of step/stride length, asymmetry, variability in step length and time, gait speed, swing time, double support duration, and gait related items of clinical scores will be calculated. The posture control measures will include back angle during walking trials and time duration of maintaining different balance tasks during Mini-BESTest evaluation. Concurrent (criterion-related) validity between the two systems (WRTF vs. Vicon) will be analyzed with ICC (2,1) and RC. The effect of RTF (CF and ODF) on each primary measure will be assessed using paired t-tests. The difference in the degree of improvements between CF and ODF modes will also be compared by the application of paired t-test on the changes in gait and posture measures between respective feedback trials and baseline trials. For conditions violating normality, Wilcoxon signed-rank test will be used.

Inclusion Criteria:

1. Age range must be between 50-80 years for males and females
2. Diagnosis of idiopathic Parkinson's disease (PD) according to United Kingdom Brain Bank Criteria
3. Unified Parkinson's Disease Rating Scale (UPDRS) walking score >= 1 during on-medication state
4. Stable dose of PD medication for 4 weeks prior to screening
5. Ability to perform long (100-meter) walking trials

Exclusion Criteria:

1. Presence of dementia according to Emre criteria
2. Regular use of assistive gait device, such as a walker or cane
3. Presence of significant ON/OFF motor fluctuations (>25% throughout the day), frequent falls (UPDRS fall score > 1), or freezing leading to falls or balance impairment, that in the opinion of the medical monitor (Dr. Shill or a St. Joseph's Hospital and Medical Center sub-investigator) can affect the subject's safety or compliance with the study protocols
4. Recent history of unstable heart or lung disease, evidence of pregnancy (self-reported by patient), major neurological (e.g., stroke, except PD) or metabolic problems (e.g., diabetes).
5. Postural hypotension (symptomatic or any prior history of postural hypertension), cardiovascular disorders, musculoskeletal disorders, or vestibular dysfunction limiting locomotion or balance
6. Lack of approval from subject's cardiologist or primary care physician (PCP) to participation in the study
7. A medication schedule that results in having to take a dose of PD medication during the data collection sessions ( to avoid confounding factors of medication dosage due to necessity to take medication during data collection sessions
8. Untreated chemical addiction or abuse or uncontrolled psychiatric illness
9. Failure to sign informed consent