A Novel Approach to Monitor Rehabilitation Outcomes in Stroke Survivors Using Wearable Technology

Patel, Shyamal; Hughes, Richard L.; Hester, Todd; Stein, Joel; Akay, Metin; Dy, Jennifer G.; Bonato, Paolo

doi:10.1109/jproc.2009.2038727

Cited by 154 publications

(130 citation statements)

References 32 publications

Supporting

Mentioning

129

Contrasting

Order By: Relevance

“…Each accelerometer and gyroscope data stream (x, y and z) exhibit signal patterns that are distinctive for each of the arm movements, which is characterized by a set of features extracted from the signals [15]. In this investigation, we consider 10 time-domain features, extracted from the data from each of the three accelerometer axes and from each of the three gyroscope axes as follows: 1) standard deviation -measure of the variability from the mean of the signal, 2) root mean square (rms) -measure of the signal energy normalized by the number of samples, 3) information entropy -measure of the randomness of a signal [35], 4) jerk metric -rms value of the second derivative of the data normalized with respect to the maximum value of the first derivative [36], 5) peak number -obtained from gradient analysis of the signal, 6) maximum peak amplitude -measure of the amplitude of the peaks obtained after gradient analysis, 7) absolute difference -absolute difference between the maximum and the minimum value of a signal, 8) index of dispersion -ratio of variance to the mean, 9) kurtosis -measure of the 'peakedness' of a signal assuming a non-Gaussian distribution in the data, 10) skewness -measure of the symmetry of the data assuming a non-Gaussian distribution in the data [37].…”

Section: Feature Extractionmentioning

confidence: 99%

“…The RELIEF algorithm [36], Clamping technique [18] and Principal Component Analysis (PCA) [12] are the most commonly used ranking/selection algorithms in the field of human activity recognition but are computationally intensive. We use the low-complexity class-separability measure based on scatter matrices to rank the 30 features for each sensor/movement combination.…”

Section: Feature Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Recognizing upper limb movements with wrist worn inertial sensors using k-means clustering classification

Biswas

Cranny

Gupta

et al. 2015

Human Movement Science

View full text Add to dashboard Cite

In this paper we present a methodology for recognizing three fundamental movements of the human forearm (extension, flexion and rotation) using pattern recognition applied to the data from a single wrist-worn, inertial sensor.We propose that this technique could be used as a clinical tool to assess rehabilitation progress in neurodegenerative pathologies such as stroke or cerebral palsy by tracking the number of times a patient performs specific arm movements (e.g. prescribed exercises) with their paretic arm throughout the day. We demonstrate this with healthy subjects and stroke patients in a simple proof of concept study in which these arm movements are detected during an archetypal activity of daily-living (ADL) -'making-a-cup-of-tea'. Data is collected from a tri-axial accelerometer and a tri-axial gyroscope located proximal to the wrist. In a training phase, movements are initially performed in a controlled environment which are represented by a ranked set of 30 time-domain features. Using a sequential forward selection technique, for each set of feature combinations three clusters are formed using k-means clustering followed by 10 runs of 10-fold cross validation on the training data to determine the best feature combinations. For the testing phase, movements performed during the ADL are associated with each cluster label using a minimum distance classifier in a multi-dimensional feature space, comprised of the best ranked features, using Euclidean or Mahalonobis distance as the metric. Experiments were performed with four healthy subjects and four stroke survivors and our results show that the proposed methodology can detect the three movements performed during the ADL with an overall average accuracy of 88% using the accelerometer data and 83% using the gyroscope data across all healthy subjects and arm movement types. The average accuracy across all stroke survivors was 70% using accelerometer data and 66% using gyroscope data. We also use a Linear Discriminant Analysis (LDA) classifier and a Support Vector Machine (SVM) classifier in association with the same set of features to detect the three arm movements and compare the results to demonstrate the effectiveness of our proposed methodology.

show abstract

Section: Feature Extractionmentioning

confidence: 99%

Section: Feature Selectionmentioning

confidence: 99%

Recognizing upper limb movements with wrist worn inertial sensors using k-means clustering classification

Biswas

Cranny

Gupta

et al. 2015

Human Movement Science

View full text Add to dashboard Cite

show abstract

“…Patel used accelerometers to record the motion of stroke survivors. Through these data, they extracted some features and quantitatively evaluated the performance of the motion [12,13]. This study perfectly overcomes the drawbacks of traditional assessments, such as the waste of human resources and diverse results due to subjective evaluation.…”

Section: Introductionmentioning

confidence: 99%

A single task assessment system of upper-limb motor function after stroke

Pan

Jin

et al. 2016

THC

View full text Add to dashboard Cite

Abstract. BACKGROUND: Nowadays, stroke is a leading cause of disability in adults. Assessment of motor performance has played an important role in rehabilitation for post stroke patients. Therefore, it is quite important to develop an automatic assessment system of motor function. OBJECTIVE: The purpose of this study is to assess the performance of the single task upper-limb movements quantitatively among stroke survivors. METHODS: Eleven normal subjects and thirty-five subjects with stroke were involved in this study. The subjects, who were wearing the micro-sensor motion capture system, performed shoulder flexion in a sitting position. The system recorded threedimensional kinematics data of limb movements in quaternions. By extracting the significant features from these data, we built a linear model to acquire the functional assessment score (FAS). RESULTS: All of the kinematics features have a significant statistical difference (P < 0.05) between patients and healthy people, while the feature values have a high correlation with Fugl-Meyer (FM) scores (r > 0.5, p < 0.05), indicating that these features are able to reflect the level of motion impairment. Furthermore, most samples of the linear model locate in the confidence interval after regression, with the residual approaching a normal distribution. These results show that the FAS is capable of motor function assessment for stroke survivors. CONCLUSION: These findings represent an important step towards a system that can be utilized for precise single task motor evaluation after stroke, applicable to clinical research and as a tool for rehabilitation.

show abstract

“…Patel et al [11] have proposed a method based on Random. Forests algorithm to monitor rehabilitation outcomes in stroke patients with sensors attached to the hand, arm and trunk.…”

Section: Introductionmentioning

confidence: 99%

Automated Brunnstrom Assessment for Home Rehabilitation Based on GRNN Model

et al. 2017

View full text Add to dashboard Cite

Abstract.To realize the upper extremity rehabilitation assessment for post-stroke intelligently, a new movement assessment model is established in this paper. A WBSN system consisted of two inertial sensors is employed to acquisit patients' rehabilitation data. The data is stored in both local and server database. An intelligent Brunnstrom assessment based on GRNN model is built on the server. In order to test the accuracy and reliability of the model, twenty patients and four physicians were chosen as volunteers to finish a standard rehabilitation action -touching shoulder with affected hand. The accuracy of the assessment model can be 93.6%. The purpose to build an intelligent assessment is achieved. It makes patients train in the home setting and community possible.

show abstract

A Novel Approach to Monitor Rehabilitation Outcomes in Stroke Survivors Using Wearable Technology

Abstract: Accelerometer data is being used to evaluate the success of rehabilitation efforts on movements, from shoulder to finger tips, for patients who have suffered strokes.

Cited by 154 publications

References 32 publications

Recognizing upper limb movements with wrist worn inertial sensors using k-means clustering classification

Recognizing upper limb movements with wrist worn inertial sensors using k-means clustering classification

A single task assessment system of upper-limb motor function after stroke

Automated Brunnstrom Assessment for Home Rehabilitation Based on GRNN Model

Contact Info

Product

Resources

About