Quantitative Gait Analysis Using a Pose-Estimation Algorithm with a Single 2D-Video of Parkinson’s Disease Patients

2021 ◽  
pp. 1-13
Author(s):  
Jung Hwan Shin ◽  
Ri Yu ◽  
Jed Noel Ong ◽  
Chan Young Lee ◽  
Seung Ho Jeon ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Pose Estimation ◽  
Estimation Algorithm ◽  
Step Length ◽  
Video Tracking ◽  
Gait Velocity ◽  
Gait Parameters ◽  
Validation Experiment ◽  
Time Required

Background: Clinician-based rating scales or questionnaires for gait in Parkinson’s disease (PD) are subjective and sensor-based analysis is limited in accessibility. Objective: To develop an easily accessible and objective tool to evaluate gait in PD patients, we analyzed gait from a single 2-dimensional (2D) video. Methods: We prospectively recorded 2D videos of PD patients (n = 16) and healthy controls (n = 15) performing the timed up and go test (TUG). The gait was simultaneously evaluated with a pressure-sensor (GAITRite). We estimated the 3D position of toes and heels with a deep-learning based pose-estimation algorithm and calculated gait parameters including step length, step length variability, gait velocity and step cadence which was validated with the result from the GAITRite. We further calculated the time and steps required for turning. Then, we applied the algorithm to previously recorded and archived videos of PD patients (n = 32) performing the TUG. Results: From the validation experiment, gait parameters derived from video tracking were in excellent agreement with the parameters obtained with the GAITRite. (Intraclass correlation coefficient > 0.9). From the analysis with the archived videos, step length, gait velocity, number of steps, and the time required for turning were significantly correlated (Absolute R > 0.4, p < 0.005) with the Freezing of gait questionnaire, Unified PD Rating scale part III total score, HY stage and postural instability. Furthermore, the video-based tracking objectively measured significant improvement of step length, gait velocity, steps and the time required for turning with antiparkinsonian medication. Conclusion: 2D video-based tracking could objectively evaluate gait in PD patients.

Corelation of gait parameters, cognitive impairment and brain atrophy in patients with Parkinson’s disease and the «normally aging population»

2021 ◽  
Author(s):  
P. Y. Diachenko ◽  
I. I. Leta ◽  
G. S. Moskovko
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Cognitive Function ◽  
Brain Atrophy ◽  
Stride Length ◽  
Aging Population ◽  
Step Length ◽  
Normal Aging ◽  
Gait Velocity ◽  
Gait Parameters

Objective — to identify the most significant markers of gait that indicate a decrease in cognitive function based on investigation of the corelation of cognitive impairment, gait parameters and atrophy of brain structures in groups of patients with Parkinson’s disease and the «normally aging population». Methods and subjects. 66 subjects were examined: 30 patients with Parkinson’s disease (mean age 54.9 ± 5.9, 50 % men) and 33 without neurological pathology (mean age 52.7 ± 7.6, 66 % men). All of them underwent neurological examination, assessment of temporal and spatial gait parameters using the GaitRite system, grading of brain atrophy using a comprehensive visual rating scale of MRI scans and assessment of cognitive status using the Montreal Cognitive Assessment Scale. Results. Cognitive performance was significantly lower in the subgroup of patients with Parkinson’s disease compared to the subgroup of «normally aging population». The gait profile of patients with Parkinson’s disease significantly differed from the gait profile of individuals from the «normal aging» subgroup by slower gait velocity, shorter step length and stride length for both limbs. The gait parameters, which showed a strong correlation with cognitive tests, differed in the subgroups, but gait velocity, stride length and step length for both extremities were common among them. These common gait parameters showed a strong direct correlation with brain atrophy in the subgroup of patients with Parkinson’s disease, but only velocity correlated with atrophy in the subgroup of «normal aging» among all of them. It was determined by the method of multiple regression analysis that it was precisely the atrophy of the brain that turned out to be the most influential factor in the decrease in cognitive function in the general group and subgroups. Conclusions. The gait profile in Parkinson’s disease subgroup is characterized by lower velocity, shorter step length, stride length for both limbs and significantly differs from the subgroup of «normal aging». These changes are a consequence of the influence of the disease on the motor sphere. Velocity showed a strong correlation in both subgroups not only with cognitive abilities, but also with cerebral atrophy. This confirms the hypothesis about the possibility of using gait velocity as a universal sensitive marker for current and longitudinal assessment of cognitive function, especially in clinical practice.  

scholarly journals Gait and Balance Changes with Investigational Peripheral Nerve Cell Therapy during Deep Brain Stimulation in People with Parkinson’s Disease

2021 ◽  
Vol 11 (4) ◽  
pp. 500
Author(s):  
Geetanjali Gera ◽  
Zain Guduru ◽  
Tritia Yamasaki ◽  
Julie A. Gurwell ◽  
Monica J. Chau ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Deep Brain Stimulation ◽  
Peripheral Nerve ◽  
Brain Stimulation ◽  
Step Length ◽  
Gait Velocity ◽  
Balance Performance ◽  
Improved Performance ◽  
Deep Brain

Background: The efficacy of deep brain stimulation (DBS) and dopaminergic therapy is known to decrease over time. Hence, a new investigational approach combines implanting autologous injury-activated peripheral nerve grafts (APNG) at the time of bilateral DBS surgery to the globus pallidus interna. Objectives: In a study where APNG was unilaterally implanted into the substantia nigra, we explored the effects on clinical gait and balance assessments over two years in 14 individuals with Parkinson’s disease. Methods: Computerized gait and balance evaluations were performed without medication, and stimulation was in the off state for at least 12 h to best assess the role of APNG implantation alone. We hypothesized that APNG might improve gait and balance deficits associated with PD. Results: While people with a degenerative movement disorder typically worsen with time, none of the gait parameters significantly changed across visits in this 24 month study. The postural stability item in the UPDRS did not worsen from baseline to the 24-month follow-up. However, we measured gait and balance improvements in the two most affected individuals, who had moderate PD. In these two individuals, we observed an increase in gait velocity and step length that persisted over 6 and 24 months. Conclusions: Participants did not show worsening of gait and balance performance in the off therapy state two years after surgery, while the two most severely affected participants showed improved performance. Further studies may better address the long-term maintanenace of these results.

Gait velocity and step length at baseline predict outcome of Nordic walking training in patients with Parkinson's disease

2015 ◽  
Vol 21 (4) ◽  
pp. 413-416 ◽  
Author(s):  
Manon Herfurth ◽  
Jana Godau ◽  
Barbara Kattner ◽  
Silvia Rombach ◽  
Stefan Grau ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Step Length ◽  
Gait Velocity ◽  
Nordic Walking

scholarly journals Gait Training Methods and ChangesiIn Gait Parameters in Parkinson`s Disease (Literature Review)

2020 ◽  
Vol 2 (7) ◽  
Author(s):  
Aušra Stuopelytė ◽  
Rasa Šakalienė
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Step Length ◽  
Gait Training ◽  
Auditory Stimulation ◽  
Training Methods ◽  
Dual Tasking ◽  
Gait Parameters ◽  
Rhythmic Auditory Stimulation ◽  
Walking Velocity

Parkinson’s disease is a chronic progressive neurological disorder that can impact function to a variable degree. Changes in gait parameters are the most common signs of Parkinson’s disease. Patients with Parkinson’s disease walk with a reduced step length, step time, walking velocity and walking cadence, increased stride cycle time, coefficient of variation of the step amplitude and step time and increased risk to fall. So, various gait training methods are applied. The effect of rhythmic auditory stimulation on gait in Parkinson’s disease patients is analysed. We can use various kinds of music, metronome, scansion and clapping as a rhythmic auditory stimulation.One of gait training methods in Parkinson’s disease patients is treadmill training. There are attempts to combine treadmill training with transcranial magnetic stimulation and virtual reality. We can use Nordic walking method and because walking technique requires straight posture, trunk rotation, bigger step and heel stride. More often robot–assisted gait training is used in patients with Parkinson’s disease gait training. The effect of dual–tasking and walking with music methods for gait and balance training in patients with Parkinson’s disease is also analysed. This method requires participants to perform primary and secondary tasks at the same time. The secondary task can be cognitive or motor. Dual–tasking is widely analysed because opinions about applying this method are very controversial. Walking with music method is more often analysed in scientific literature. This method could not be compared to rhythmic auditory stimulation method because the latter requires precise walking to rhythm and walking with music method is oriented to emotional component (music is chosen according patients’ music taste). As these methods are applied, we can see an increase in chosen walking and maximal walking velocities, step length and time, distance covered, and decrease in the coefficient of variation of the step time and turning time.Keywords: Gait impairments, walking velocity, rhythmic auditory stimulation.

scholarly journals Influence of Autocorrelated Rhythmic Auditory Stimulations on Parkinson’s Disease Gait Variability: Comparison With Other Auditory Rhythm Variabilities and Perspectives

2020 ◽  
Vol 11 ◽  
Author(s):  
Alexis Lheureux ◽  
Thibault Warlop ◽  
Charline Cambier ◽  
Baptiste Chemin ◽  
Gaëtan Stoquart ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Long Range ◽  
Coefficient Of Variation ◽  
Gait Speed ◽  
Step Length ◽  
Temporal Organization ◽  
Fluctuation Analysis ◽  
Gait Parameters ◽  
Stride Duration

Parkinson’s Disease patients suffer from gait impairments such as reduced gait speed, shortened step length, and deterioration of the temporal organization of stride duration variability (i.e., breakdown in Long-Range Autocorrelations). The aim of this study was to compare the effects on Parkinson’s Disease patients’ gait of three Rhythmic Auditory Stimulations (RAS), each structured with a different rhythm variability (isochronous, random, and autocorrelated). Nine Parkinson’s Disease patients performed four walking conditions of 10–15 min each: Control Condition (CC), Isochronous RAS (IRAS), Random RAS (RRAS), and Autocorrelated RAS (ARAS). Accelerometers were used to assess gait speed, cadence, step length, temporal organization (i.e., Long-Range Autocorrelations computation), and magnitude (i.e., coefficient of variation) of stride duration variability on 512 gait cycles. Long-Range Autocorrelations were assessed using the evenly spaced averaged Detrended Fluctuation Analysis (α-DFA exponent). Spatiotemporal gait parameters and coefficient of variation were not modified by the RAS. Long-Range Autocorrelations were present in all patients during CC and ARAS although all RAS conditions altered them. The α-DFA exponents were significantly lower during IRAS and RRAS than during CC, exhibiting anti-correlations during IRAS in seven patients. α-DFA during ARAS was the closest to the α-DFA during CC and within normative data of healthy subjects. In conclusion, Isochronous RAS modify patients’ Long-Range Autocorrelations and the use of Autocorrelated RAS allows to maintain an acceptable level of Long-Range Autocorrelations for Parkinson’s Disease patients’ gait.

Effects of mechanical stimulation of the feet on gait and cardiovascular autonomic control in Parkinson's disease

2014 ◽  
Vol 116 (5) ◽  
pp. 495-503 ◽  
Author(s):  
Franca Barbic ◽  
Manuela Galli ◽  
Laura Dalla Vecchia ◽  
Margherita Canesi ◽  
Veronica Cimolin ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Three Dimensional ◽  
Motor Impairment ◽  
Rotation Velocity ◽  
Step Length ◽  
Autonomic Control ◽  
Gait Velocity ◽  
Central Processing ◽  
Cardiovascular Autonomic Control

Motor impairment in Parkinson's disease (PD) is partly due to defective central processing of lower limb afferents. Concomitant alterations in cardiovascular autonomic control leading to orthostatic hypotension may worsen motor ability. We evaluated whether mechanical activation of feet sensory afferents could improve gait and modify the response of cardiovascular autonomic control to stressors in 16 patients (age 66 ± 2 yr) with idiopathic PD (Hoehn & Yhar scale 2–3) on their usual therapy. Eight subjects ( group A) were randomized to undergo skin pressure (0.58 ± 0.04 kg/mm2) stimulation at the hallux tip and first metatarsal joint (effective stimulation; ES) of both feet. Eight remaining patients ( group B) underwent sham stimulation (SS) followed by ES. Three-dimensional movement analysis provided quantitative indexes of movement disability before (baseline) and 24 h after ES and SS. Spectral analysis of heart rate and blood pressure variability provided markers of cardiac sympatho-vagal (LF/HF) and vascular sympathetic (LFSAP) modulations. Markers were measured at rest and during 75° head-up tilt, before and 24 h after ES and SS. After ES, step length and gait velocity increased, upright rotation velocity was enhanced, and step number was decreased. After ES, LFSAP declined. The increase in LF/HF and LFSAP induced by tilt was greater than before feet stimulation. No changes in gait and autonomic parameters were observed after SS. Twenty-four hours after ES, patients with PD showed improved gait and increased cardiac and vascular sympathetic modulation during upright position compared with baseline. Conversely, SS was ineffective on both movement and autonomic parameters, indicating a site specificity effect of the stimulation.

scholarly journals Step Length Is a Promising Progression Marker in Parkinson’s Disease

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2292
Author(s):  
Julius Welzel ◽  
David Wendtland ◽  
Elke Warmerdam ◽  
Robbin Romijnders ◽  
Morad Elshehabi ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Disease Progression ◽  
Step Length ◽  
Minor Effect ◽  
Progression Marker ◽  
Gait Parameters ◽  
Ancova Model ◽  
Length Changes ◽  
A Minor

Current research on Parkinson’s disease (PD) is increasingly concerned with the identification of objective and specific markers to make reliable statements about the effect of therapy and disease progression. Parameters from inertial measurement units (IMUs) are objective and accurate, and thus an interesting option to be included in the regular assessment of these patients. In this study, 68 patients with PD (PwP) in Hoehn and Yahr (H&Y) stages 1–4 were assessed with two gait tasks—20 m straight walk and circular walk—using IMUs. In an ANCOVA model, we found a significant and large effect of the H&Y scores on step length in both tasks, and only a minor effect on step time. This study provides evidence that from the two potentially most important gait parameters currently accessible with wearable technology under supervised assessment strategies, step length changes substantially over the course of PD, while step time shows surprisingly little change in the progression of PD. These results show the importance of carefully evaluating quantitative gait parameters to make assumptions about disease progression, and the potential of the granular evaluation of symptoms such as gait deficits when monitoring chronic progressive diseases such as PD.

scholarly journals Effects of dance on gait and dual-task gait in Parkinson’s disease

2019 ◽  
Author(s):  
Nadeesha Kalyani Hewa Haputhanthirige ◽  
Karen Sullivan ◽  
Gene Moyle ◽  
Sandy Brauer ◽  
Erica Rose Jeffrey ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Dual Task ◽  
Verbal Fluency ◽  
Large Scale ◽  
Mixed Model ◽  
Linear Mixed Model ◽  
Step Length ◽  
Gait Velocity ◽  
Control Group

Abstract Background Gait impairments in Parkinson’s disease (PD) limit independence and quality of life. While dance based interventions could improve gait, further studies are needed to determine if the benefits generalise to different terrains and when dual-tasking. The aim was to perform a feasibility study of the effects of a dance intervention, based on the Dance for PD®(DfPD®) program, on gait under different dual-tasks (verbal fluency, serial subtraction) and surfaces (even, uneven), and to determine if a larger scale follow-up RCT is warranted.Methods A dance group (DG; n = 17; age = 65.8 ± 11.7 years) and a control group (CG: n = 16; age = 67.0 ± 7.7 years) comprised of non-cognitively impaired (Addenbrooke’s score: DG = 93.2 ± 3.6, CG = 92.6 ± 4.3) independently locomoting people with PD (Hoehn & Yahr I-III) participated in the study. The DG undertook a one-hour DfPD®based class, twice weekly for 12 weeks. The CG had treatment as usual. Gait analysis was performed at baseline and post-intervention while walking on two surfaces (even, uneven) under three conditions (regular walking; dual-task: verbal-fluency, serial-subtraction). The data was analysed by means of a linear mixed model. ResultsThe DG improved significantly compared to the CG in gait velocity, cadence, step-length, and stride-length when even surface walking, with and without a dual-task. On the uneven surface the DG walked more cautiously during regular walking but had improved gait velocity, cadence and step-length when performing serial-subtractions. Conclusions DfPD®-based classes produced clinically significant improvement on spatiotemporal gait parameters under dual-task conditions and on uneven surfaces. This could arise from improved movement confidence and coordination; emotional expression; cognitive skills (planning, multitasking), and; utilisation of external movement cues. A large-scale RCT of this program is warranted.Trial registration A protocol for this study has been registered retrospectively at Australian New Zealand Clinical Trials Registry on 12.11.2018. Identifier: ACTRN12618001834246.

scholarly journals Clinical Relevance of Standardized Mobile Gait Tests. Reliability Analysis Between Gait Recordings at Hospital and Home in Parkinson’s Disease: A Pilot Study

2020 ◽  
Vol 10 (4) ◽  
pp. 1763-1773
Author(s):  
Heiko Gaßner ◽  
Philipp Sanders ◽  
Alisa Dietrich ◽  
Franz Marxreiter ◽  
Bjoern M. Eskofier ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Pilot Study ◽  
Reliability Analysis ◽  
Clinical Relevance ◽  
Stride Length ◽  
Standardized Test ◽  
Gait Velocity ◽  
Gait Parameters ◽  
At Home

Background: Gait impairments in Parkinson’s disease (PD) are quantified using inertial sensors under standardized test settings in the hospital. Recent studies focused on the assessment of free-living gait in PD. However, the clinical relevance of standardized gait tests recorded at the patient’s home is unclear. Objective: To evaluate the reliability of supervised, standardized sensor-based gait outcomes at home compared to the hospital. Methods: Patients with PD (n = 20) were rated by a trained investigator using the Unified Parkinson Disease Rating Scale (UPDRS-III). Gait tests included a standardized 4×10 m walk test and the Timed Up and Go Test (TUG). Tests were performed in the hospital (HOSPITAL) and at patients’ home (HOME), and controlled for investigator, time of the day, and medication. Statistics included reliability analysis using Intra-Class correlations and Bland-Altman plots. Results: UPDRS-III and TUG were comparable between HOSPITAL and HOME. PD patients’ gait at HOME was slower (gait velocity Δ= –0.07±0.11 m/s, –6.1%), strides were shorter (stride length Δ= –9.2±9.4 cm; –7.3%), and shuffling of gait was more present (maximum toe-clearance Δ= –0.7±2.5 cm; –8.8%). Particularly, narrow walkways (<85 cm) resulted in a significant reduction of gait velocity at home. Reliability analysis (HOSPITAL vs. HOME) revealed excellent ICC coefficients for UPDRS-III (0.950, p < 0.000) and gait parameters (e.g., stride length: 0.898, p < 0.000; gait velocity: 0.914, p < 0.000; stance time: 0.922, p < 0.000; stride time: 0.907, p < 0.000). Conclusion: This pilot study underlined the clinical relevance of gait parameters by showing excellent reliability for supervised, standardized gait tests at HOSPITAL and HOME, even though gait parameters were different between test conditions.

scholarly journals Gait biofeedback training in people with Parkinson’s disease: a pilot study

2021 ◽  
Author(s):  
Kate McMaster ◽  
Michael H Cole ◽  
Daniel Chalkley ◽  
Mark W. Creaby
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Real Time ◽  
Step Length ◽  
Trial Registration ◽  
Biofeedback Training ◽  
Gait Velocity ◽  
Post Intervention ◽  
Trunk Motion ◽  
Risk Of Falls

Abstract Background People with Parkinson’s disease (PD) are at a high risk of falls, with ~ 60% experiencing a fall each year. Greater mediolateral head and pelvis motion during gait are known to increase the risk of falling in PD, however the ability to modify these aspects of gait has not been examined. Thus, this study aimed to examine whether mediolateral trunk, head and pelvis motion during walking could be successfully decreased in people with PD using real-time biofeedback. Methods Participants were provided with real-time biofeedback regarding their mediolateral trunk lean via a visual projection whilst walking along an 8-metre indoor walkway. Using the feedback provided, they were asked to reduce the magnitude of their mediolateral trunk lean. Gait was recorded for four conditions (i) Baseline, (ii) Intervention, (iii) immediately Post-Intervention, and (iv) one-week Follow-Up. Biomechanical variables associated with falls risk were compared between conditions, including normalised mediolateral motion, gait velocity and stride length. Results A reduction in mediolateral trunk lean, step length and gait velocity from Baseline to the Intervention and Post-intervention conditions was observed. Contrary to this, increased normalised ML pelvis and trunk motion was observed between the Baseline and Intervention conditions, but returned to Baseline levels in the Post-Intervention condition. Conclusions Results from the current study suggest that real-time visual biofeedback may be effective at modifying specific gait characteristics that are associated with falls in PD. Further research is required to better understand the influence of this intervention approach has upon falls incidence. Trial Registration: Australian New Zealand Clinical Trials Registry ACTRN12620000994987. Registered 10 June 2020 - Retrospectively registered, https://anzctr.org.au/Trial/Registration/TrialReview.aspx?id=380324.

Sign in / Sign up

Export Citation Format

Share Document