Selection of gait parameters for differential diagnostics of patients with <i>de novo</i> Parkinson’s disease

Djurić-Jovičić, Milica; Belić, Minja; Stanković, Iva; Radovanović, Saša; Kostić, Vladimir

doi:10.1080/01616412.2017.1348690

Cited by 27 publications

(27 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, previous literature has shown that, when using wearables to quantify gait, studies using 2 min CW protocols [18,19] achieved better results compared to those using 10m IW protocols [13,45]. In addition, studies showed that ML models derived from wearable inertial and force feet sensors [14,19,45,46] performed relatively better when compared to studies based on GAITRite data [17].…”

Section: Performance: An Overviewmentioning

confidence: 96%

“…The support vector machine with radial basis function (SVM-RBF) and random forest were used because these are the most widely used ML models for PD classification [13][14][15][16][17][18][19]40]. The models were trained on the same conceptual features from both sensing systems to compare the impact of walking protocols and gait assessment systems.…”

Section: Statistical Analysis Gait Normalization and Classification mentioning

confidence: 99%

“…This study had some limitations. Only two widely used ML models (SVM & RF) [13][14][15][16][17][18][19] were used in this study to compare the effect of walking protocols and gait assessment systems. However, future work should explore other classification models such as logistic regression and neural networks.…”

Section: Limitationsmentioning

confidence: 99%

“…In order to maximize information for disease classification, analysis of multiple characteristics can be enhanced using machine learning (ML) [6]. The most widely used ML models for PD classification are the support vector machine (SVM) and random forest (RF) [13][14][15][16][17][18][19]. Classification accuracy however is inconsistent across studies which may be largely due to methodological differences (e.g., testing protocols, gait assessment systems and normalization of participants' data) [13,14,17].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparison of Walking Protocols and Gait Assessment Systems for Machine Learning-Based Classification of Parkinson’s Disease

Rehman

Din

Shi

et al. 2019

Sensors

View full text Add to dashboard Cite

Early diagnosis of Parkinson’s diseases (PD) is challenging; applying machine learning (ML) models to gait characteristics may support the classification process. Comparing performance of ML models used in various studies can be problematic due to different walking protocols and gait assessment systems. The objective of this study was to compare the impact of walking protocols and gait assessment systems on the performance of a support vector machine (SVM) and random forest (RF) for classification of PD. 93 PD and 103 controls performed two walking protocols at their normal pace: (i) four times along a 10 m walkway (intermittent walk-IW), (ii) walking for 2 minutes on a 25 m oval circuit (continuous walk-CW). 14 gait characteristics were extracted from two different systems (an instrumented walkway—GAITRite; and an accelerometer attached at the lower back—Axivity). SVM and RF were trained on normalized data (accounting for step velocity, gender, age and BMI) and evaluated using 10-fold cross validation with area under the curve (AUC). Overall performance was higher for both systems during CW compared to IW. SVM performed better than RF. With SVM, during CW Axivity significantly outperformed GAITRite (AUC: 87.83 ± 7.81% vs. 80.49 ± 9.85%); during IW systems performed similarly. These findings suggest that choice of testing protocol and sensing system may have a direct impact on ML PD classification results and highlight the need for standardization for wide scale implementation.

show abstract

Section: Performance: An Overviewmentioning

confidence: 96%

Section: Statistical Analysis Gait Normalization and Classification mentioning

confidence: 99%

Section: Limitationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Comparison of Walking Protocols and Gait Assessment Systems for Machine Learning-Based Classification of Parkinson’s Disease

Rehman

Din

Shi

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…Furthermore, the disease duration is 7.3 ± 5.4 years, and, at this stage of the disease, the people with PD have a similar likelihood of unilateral and bilateral motor impairments [37]. One crucial question raised by these findings is the importance of gait analysis for detecting the motor asymmetry as a screening evaluation [38,39]. In our study, we consider gait asymmetry based on the study [23].…”

Section: Discussionmentioning

confidence: 99%

Effects of Nordic Walking on Gait Symmetry in Mild Parkinson’s Disease

et al. 2019

View full text Add to dashboard Cite

Individuals with Parkinson’s disease (PD) have gait asymmetries, and exercise therapy may reduce the differences between more and less affected limbs. The Nordic walking (NW) training may contribute to reducing the asymmetry in upper and lower limb movements in people with PD. We compared the effects of 11 weeks of NW aerobic training on asymmetrical variables of gait in subjects with mild PD. Fourteen subjects with idiopathic PD, age: 66.8 ± 9.6 years, and Hoehn and Yard stage of 1.5 points were enrolled. The kinematic analysis was performed pre and post-intervention. Data were collected at two randomized walking speeds (0.28 m·s−1 and 0.83 m·s−1) during five minutes on the treadmill without poles. The more affected and less affected body side symmetries (threshold at 5% between sides) of angular kinematics and spatiotemporal gait parameters were calculated. We used Generalized Estimating Equations with Bonferroni post hoc (α = 0.05). Maximal flexion of the knee (p = 0.007) and maximal abduction of the hip (p = 0.041) were asymmetrical pre and became symmetrical post NW intervention. The differences occurred in the knee was less affected and the hip was more affected. We concluded that 11 weeks of NW training promoted similarities in gait parameters and improved knee and hip angular parameters for PD subjects.

show abstract

The motor prodromes of parkinson’s disease: from bedside observation to large-scale application

et al. 2019

View full text Add to dashboard Cite

There is sufficient evidence that the pathological process that causes Parkinson's disease begins years before the clinical diagnosis is made. Over the last 15 years, there has been much interest in the existence of a prodrome in some patients, with a particular focus on non-motor symptoms such as reduced sense of smell, REM-sleep disorder, depression, and constipation. Given that the diagnostic criteria for Parkinson's disease depends on the presence of bradykinesia, it is somewhat surprising that there has been much less research into the possibility of subtle motor dysfunction as a pre-diagnostic pointer. This review will focus on early motor features and provide some advice on how to detect and measure them.

show abstract

Selection of gait parameters for differential diagnostics of patients with de novo Parkinson’s disease

Abstract: In early stages of PD, double support time and its CV appear to be weak indicators of the disease. We instead found step time asymmetry and support base CV to significantly contribute to classification accuracy.

Cited by 27 publications

References 28 publications

Comparison of Walking Protocols and Gait Assessment Systems for Machine Learning-Based Classification of Parkinson’s Disease

Comparison of Walking Protocols and Gait Assessment Systems for Machine Learning-Based Classification of Parkinson’s Disease

Effects of Nordic Walking on Gait Symmetry in Mild Parkinson’s Disease

The motor prodromes of parkinson’s disease: from bedside observation to large-scale application

Contact Info

Product

Resources

About