Turning Detection During Gait: Algorithm Validation and Influence of Sensor Location and Turning Characteristics in the Classification of Parkinson’s Disease

Rehman, Rana Zia Ur; Klocke, Philipp; Hryniv, Sofia; Galna, Brook; Rochester, Lynn; Din, Silvia Del; Alcock, Lisa

doi:10.3390/s20185377

Cited by 26 publications

(41 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Turns were identified by the Mobility Lab software, where the angular velocity about the vertical axis exceeded a threshold of 15 • / s. The start and end of each turn were set to the point where the angular velocity dropped below 5 • / s 16 . For each detected turn, the following features were extracted using the Mobility Lab software: turn angle, duration, turn velocity, steps within turn, and raw accelerometer data of all three sensors 16,29 . For determining the lateral acceleration, the sensor data was reoriented from the sensor body frame into a global reference frame using the orientation estimates provided by the Mobility Lab software 16 .…”

Section: Methods For Measuring Turning Movementsmentioning

confidence: 99%

Turning movements in real life capture subtle longitudinal and preataxic changes in cerebellar ataxia

Thierfelder

Seemann

John

et al. 2021

Preprint

View full text Add to dashboard Cite

OBJECTIVES: Clinical and regulatory acceptance of upcoming molecular treatments in degenerative ataxias might greatly benefit from ecologically valid endpoints which capture change in ataxia severity in patients real life. This longitudinal study aimed to unravel quantitative motor biomarkers in degenerative ataxias in real life turning movements which are sensitive for changes both longitudinally and at the preataxic stage. METHODS: Combined cross-sectional (n=30) and longitudinal (n=14, 1 year interval) observational study in degenerative cerebellar disease (including 8 pre-ataxic mutation carriers) compared to 23 healthy controls. Turning movements were assessed by three body-worn inertial sensors in three conditions: (1) instructed laboratory assessment, (2) supervised free walking, and (3) unsupervised real-life movements. RESULTS: Measures which quantified dynamic balance during turning, lateral velocity change (LVC) and outward acceleration, but not general turning measures such as speed, allowed differentiating ataxic against healthy subjects in real life with high effect size (δ=0.68), with LVC also differentiating preataxic against healthy subjects (δ=0.53). LVC was highly correlated with clinical ataxia severity (SARA score, effect size ρ=0.79) and subjective balance confidence (ABC score, ρ=0.66). Moreover, LVC in real life but not general turning measures, gait measures, or the SARA score allowed detecting significant longitudinal change in one-year follow-up with high effect size (rprb=0.66). CONCLUSIONS: Measures of turning allow to capture specific changes of dynamic balance in degenerative ataxia in real life, with high sensitivity to longitudinal differences in ataxia severity and to the preataxic stage. They thus present promising ecologically valid motor biomarkers for capturing change in real life, even in the highly treatment-relevant early stages of degenerative cerebellar disease.

show abstract

Section: Methods For Measuring Turning Movementsmentioning

confidence: 99%

Turning movements in real life capture subtle longitudinal and preataxic changes in cerebellar ataxia

Thierfelder

Seemann

John

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In addition, some studies were recently conducted to evaluate the optimal combination of turning characteristics for improving the classification and prediction performance of freezers among people with PD [ 12 , 17 , 18 ]. These studies have suggested that the occurrence of FOG while turning is associated with structures [ 12 ] such as the prefrontal areas, central pattern generators in the spinal cord, mesencephalic locomotor region, and executive frontal regions [ 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

“…Modeling using machine learning algorithms based on data combined with comprehensive turning characteristics was recently conducted [ 21 – 23 ]. Previous studies have identified the predictors for classification of people with PD as faller and non-faller [ 20 ]; these have determined the optimal combination of turning characteristics to distinguish between people with PD and controls [ 17 , 25 ] as well as categorized people with PD as freezers and non-freezers based on the classifiers [ 18 ]. The studies reported a classification accuracy of approximately 70–98% when logistic regression (LR), random forest (RF), support vector machine (SVM), extreme gradient boosting [ 24 ], probabilistic neural network [ 18 ], recursive feature elimination technique with SVM [ 25 ], and partial least square discriminant analysis [ 17 ] were used for training.…”

Section: Introductionmentioning

confidence: 99%

“…Previous studies have identified the predictors for classification of people with PD as faller and non-faller [ 20 ]; these have determined the optimal combination of turning characteristics to distinguish between people with PD and controls [ 17 , 25 ] as well as categorized people with PD as freezers and non-freezers based on the classifiers [ 18 ]. The studies reported a classification accuracy of approximately 70–98% when logistic regression (LR), random forest (RF), support vector machine (SVM), extreme gradient boosting [ 24 ], probabilistic neural network [ 18 ], recursive feature elimination technique with SVM [ 25 ], and partial least square discriminant analysis [ 17 ] were used for training. However, these machine learning studies were limited to small sample sizes, analyzed limited turning characteristics, and were vulnerable to the risk of overfitting the data owing to high correlation with multiple variables [ 26 ].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of Parkinson’s disease with freezing of gait based on 360° turning analysis using 36 kinematic features

Park

Shin

Youm

et al. 2021

J NeuroEngineering Rehabil

View full text Add to dashboard Cite

Background Freezing of gait (FOG) is a sensitive problem, which is caused by motor control deficits and requires greater attention during postural transitions such as turning in people with Parkinson’s disease (PD). However, the turning characteristics have not yet been extensively investigated to distinguish between people with PD with and without FOG (freezers and non-freezers) based on full-body kinematic analysis during the turning task. The objectives of this study were to identify the machine learning model that best classifies people with PD and freezers and reveal the associations between clinical characteristics and turning features based on feature selection through stepwise regression. Methods The study recruited 77 people with PD (31 freezers and 46 non-freezers) and 34 age-matched older adults. The 360° turning task was performed at the preferred speed for the inner step of the more affected limb. All experiments on the people with PD were performed in the “Off” state of medication. The full-body kinematic features during the turning task were extracted using the three-dimensional motion capture system. These features were selected via stepwise regression. Results In feature selection through stepwise regression, five and six features were identified to distinguish between people with PD and controls and between freezers and non-freezers (PD and FOG classification problem), respectively. The machine learning model accuracies revealed that the random forest (RF) model had 98.1% accuracy when using all turning features and 98.0% accuracy when using the five features selected for PD classification. In addition, RF and logistic regression showed accuracies of 79.4% when using all turning features and 72.9% when using the six selected features for FOG classification. Conclusion We suggest that our study leads to understanding of the turning characteristics of people with PD and freezers during the 360° turning task for the inner step of the more affected limb and may help improve the objective classification and clinical assessment by disease progression using turning features.

show abstract

“…Wearables have been used to assess gait in clinical and free-living conditions [23][24][25][26][27][28][29]. Gait characteristics obtained through signal-processing methods can be used to characterise fallers and non-fallers and these outcomes may be used to inform tailored intervention rehabilitation plans [30].…”

Section: Introductionmentioning

confidence: 99%

Gait Analysis with Wearables Can Accurately Classify Fallers from Non-Fallers: A Step toward Better Management of Neurological Disorders

Rehman

Zhou

Din

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Falls are the leading cause of mortality, morbidity and poor quality of life in older adults with or without neurological conditions. Applying machine learning (ML) models to gait analysis outcomes offers the opportunity to identify individuals at risk of future falls. The aim of this study was to determine the effect of different data pre-processing methods on the performance of ML models to classify neurological patients who have fallen from those who have not for future fall risk assessment. Gait was assessed using wearables in clinic while walking 20 m at a self-selected comfortable pace in 349 (159 fallers, 190 non-fallers) neurological patients. Six different ML models were trained on data pre-processed with three techniques such as standardisation, principal component analysis (PCA) and path signature method. Fallers walked more slowly, with shorter strides and longer stride duration compared to non-fallers. Overall, model accuracy ranged between 48% and 98% with 43–99% sensitivity and 48–98% specificity. A random forest (RF) classifier trained on data pre-processed with the path signature method gave optimal classification accuracy of 98% with 99% sensitivity and 98% specificity. Data pre-processing directly influences the accuracy of ML models for the accurate classification of fallers. Using gait analysis with trained ML models can act as a tool for the proactive assessment of fall risk and support clinical decision-making.

show abstract

Turning Detection During Gait: Algorithm Validation and Influence of Sensor Location and Turning Characteristics in the Classification of Parkinson’s Disease

Cited by 26 publications

References 48 publications

Turning movements in real life capture subtle longitudinal and preataxic changes in cerebellar ataxia

Turning movements in real life capture subtle longitudinal and preataxic changes in cerebellar ataxia

Classification of Parkinson’s disease with freezing of gait based on 360° turning analysis using 36 kinematic features

Gait Analysis with Wearables Can Accurately Classify Fallers from Non-Fallers: A Step toward Better Management of Neurological Disorders

Contact Info

Product

Resources

About