**7. Feature extraction and classification**

Time and frequency domain features of the EMG signal may be used to diagnose gait disorders. For example, an image processing technique can be used to detect pathological gait affected by abnormal firing of MUs [65]. Machine learning algorithms are important tools in detecting the pattern of normal and abnormal gait [66, 67]. They do so by making minimum assumptions about the data generating system, as it does not need a carefully controlled experimental design [9]. Application of machine learning algorithms to detect and classify gait disorders is suited to big data. Machine Learning is further divided into: (1) Supervised learning and (2) unsupervised learning. We will now discuss techniques to detect gait disorders using supervised and unsupervised learning algorithms.

#### **7.1 Unsupervised learning**

Unsupervised learning can be used to find structures in the EMG data. For example, cluster analysis has been used to identify alteration in the gait patterns, which are undetected by statistical tests. Patients with Parkinson's disease can be distinguished from a healthy individual by using cluster analysis of dimensionally

**29**

**Figure 5.**

*A Review of EMG Techniques for Detection of Gait Disorders*

to minimize objective function, which is given by (5).

*V*(*j*) = ∑

*k* is the number of clusters and ‖*xi* <sup>−</sup> *cj*‖

displayed in **Figure 5**.

**7.2 Supervised learning**

reduced feature vector [68, 69]. K-means clustering is a very common clustering technique that initially estimates K centroids randomly or selectively. The algorithm iterates between two steps, data assignment steps and updating centroid. The aim is

> *j*=1 *k* ∑ *i*=1 *n*

2

where *V*(*j*) is the objective function, *n* is the number of data points in *jth* cluster,

The hypothesis of muscle synergies has been applied in several studies [44, 45, 70]. Unsupervised Learning helps in grouping identical synergies and can be helpful in diagnosing gait disorders. Kim et al. [70] identified synergies using iterative *K*-mean clustering and intraclass correlation. Hierarchical, model-based, fuzzy c means clustering has been employed to group gait patterns [69, 71–73]. Dolatabadi et al. [71] used mixture model clustering on spatiotemporal gait pattern to classify pathological gait. Pathological disorders such as cerebral palsy that show higher inter-stride variability can be analyzed with a hierarchical clustering method proposed by Rosati et al. [72]. Feature Fusion technique with Davies Bouldin Index (DBI) based on fuzzy C means algorithm was used in a trip/fall study [73]. The DBI can be used to evaluate the clustering algorithm. We have used K mean cluster analysis to cluster normal gait and gait with constraints, which are

In supervised learning, the predictive models are based on the input and output data. Some of the widely used learning algorithms are decision trees, Bayesian networks, support vector machine, artificial neural networks, and linear discriminant analysis (LDA). After feature extraction and classification, the EMG time series can be modeled to control prosthetic or rehabilitative device. The fundamental

The performance of different algorithms (SVM, LDA, MLP) in classifying gait disorders (Cerebral Palsy) was compared [74]. SVM classifier, compared to LDA and MLP, performed better when the analysis was done on kinematic data [74]. The normalization of the EMG data from different limb configurations increased

*A total of four clusters were chosen to group sEMG signal based on 93% variability in data within each cluster. The clusters were plotted for the first two principal components for walking with and without constraint.*

approach to classification of EMG signal is shown in **Figure 6** [66].

‖*xi* − *cj*‖

is the square of Euclidean distance.

<sup>2</sup> (5)

*DOI: http://dx.doi.org/10.5772/intechopen.84403*

*Artificial Intelligence - Applications in Medicine and Biology*

*IMNF*(*t*) <sup>=</sup> <sup>∑</sup>*j*=1

*<sup>W</sup>*(*x*, *<sup>y</sup>*) <sup>=</sup> \_\_1

**7. Feature extraction and classification**

Wavelet transform such as Multitaper is well suited for non-stationary signals.

*<sup>N</sup> fiW*(*fi*,*t*) \_\_\_\_\_\_\_\_\_\_\_\_ ∑*j*=1

*y*(*t*)ψ.

*<sup>N</sup> <sup>W</sup>*(*fi*,*t*) (3)

(*<sup>t</sup>* <sup>−</sup> *<sup>y</sup>*) \_\_\_\_\_ *<sup>x</sup> dt* (4)

Wavelet transform elicits good localization of energy when the MUAP shape matches that of the wavelet [8]. Continuous wavelet transform (CWT) of bandpass filtered EMG showed alteration in the motor unit among stroke patients when a foot drop stimulator device was used (FDS) [63]. Energy localization below 100 Hz that resulted from foot drop was caused by slow motor unit recruitment. The neuromuscular activation improved with FDS. The time-frequency plot for Gastrocnemius showed that peak energy localization shifted from 50 to 100 Hz as a neuromuscular strategy [63]. Instantaneous mean frequency (IMNF) is the average frequency of power density spectrum of a signal and is computed from time-frequency distribution, W(*f, t*) [63], where W is obtained from continuous wavelet transformation

> √ \_\_ *x* ∫

−∞ +∞

In the above, *x* is the scaling factor that controls the width of the wavelet, *y* controls its location in time, ψ is the mother wavelet function and *y*(*t*) is the signal. Instantaneous mean frequency can also be computed from the scalogram of CWT by its dimensional reduction. The scalogram has three dimensional space with time (x axis), frequency (y axis) and power (z axis) [63, 64]. In growing children, the higher IMNF level computed from scalogram revealed difference with respect to the children with cerebral palsy. The IMNF frequency component, unlike healthy children, decreased with age and maturation for children with cerebral palsy. IMNF also provided significant differences between the affected and unaffected site

Time and frequency domain features of the EMG signal may be used to diagnose gait disorders. For example, an image processing technique can be used to detect pathological gait affected by abnormal firing of MUs [65]. Machine learning algorithms are important tools in detecting the pattern of normal and abnormal gait [66, 67]. They do so by making minimum assumptions about the data generating system, as it does not need a carefully controlled experimental design [9]. Application of machine learning algorithms to detect and classify gait disorders is suited to big data. Machine Learning is further divided into: (1) Supervised learning and (2) unsupervised learning. We will now discuss techniques to detect gait

Unsupervised learning can be used to find structures in the EMG data. For example, cluster analysis has been used to identify alteration in the gait patterns, which are undetected by statistical tests. Patients with Parkinson's disease can be distinguished from a healthy individual by using cluster analysis of dimensionally

disorders using supervised and unsupervised learning algorithms.

**6.2 The wavelet transform**

defined by (3) and (4).

among stroke patients [63].

**7.1 Unsupervised learning**

**28**

reduced feature vector [68, 69]. K-means clustering is a very common clustering technique that initially estimates K centroids randomly or selectively. The algorithm iterates between two steps, data assignment steps and updating centroid. The aim is to minimize objective function, which is given by (5).

$$\left| V\{\hat{j} \} \right\rangle = \sum\_{j=1}^{k} \sum\_{i=1}^{n} \left\| \left| \mathbf{x}\_{i} - \mathbf{c}\_{j} \right\| \right\|^{2} \tag{5}$$

where *V*(*j*) is the objective function, *n* is the number of data points in *jth* cluster, *k* is the number of clusters and ‖*xi* <sup>−</sup> *cj*‖ 2 is the square of Euclidean distance.

The hypothesis of muscle synergies has been applied in several studies [44, 45, 70]. Unsupervised Learning helps in grouping identical synergies and can be helpful in diagnosing gait disorders. Kim et al. [70] identified synergies using iterative *K*-mean clustering and intraclass correlation. Hierarchical, model-based, fuzzy c means clustering has been employed to group gait patterns [69, 71–73]. Dolatabadi et al. [71] used mixture model clustering on spatiotemporal gait pattern to classify pathological gait. Pathological disorders such as cerebral palsy that show higher inter-stride variability can be analyzed with a hierarchical clustering method proposed by Rosati et al. [72]. Feature Fusion technique with Davies Bouldin Index (DBI) based on fuzzy C means algorithm was used in a trip/fall study [73]. The DBI can be used to evaluate the clustering algorithm. We have used K mean cluster analysis to cluster normal gait and gait with constraints, which are displayed in **Figure 5**.

### **7.2 Supervised learning**

In supervised learning, the predictive models are based on the input and output data. Some of the widely used learning algorithms are decision trees, Bayesian networks, support vector machine, artificial neural networks, and linear discriminant analysis (LDA). After feature extraction and classification, the EMG time series can be modeled to control prosthetic or rehabilitative device. The fundamental approach to classification of EMG signal is shown in **Figure 6** [66].

The performance of different algorithms (SVM, LDA, MLP) in classifying gait disorders (Cerebral Palsy) was compared [74]. SVM classifier, compared to LDA and MLP, performed better when the analysis was done on kinematic data [74]. The normalization of the EMG data from different limb configurations increased

#### **Figure 5.**

*A total of four clusters were chosen to group sEMG signal based on 93% variability in data within each cluster. The clusters were plotted for the first two principal components for walking with and without constraint.*

**Figure 6.**

*Block diagram of an EMG Signal classification system.*

classification accuracy [74, 75]. Feature level fusion is used to extract the feature space from daily life activities [73]. Patients with Parkinson's were classified with high accuracy using SVM with leave-one-out cross-validation [75]. Results from Nair et al. [76] suggest that least square kernel algorithm performed better than LDA, Neural Network, MLP and learning vector quantification (LVQ ) for patients with arthritis. Decision Tree (DT) classifier used to classify toe walking gait disorder revealed three major toe-walking patterns [77]: (1) muscle weakness of TA and quadriceps and spasticity of Tibialis Surae; (2) severe spasticity of Tibialis Surae with limited range of ankle motion; and, (3) hamstring spasticity. The MLP, on the other hand, exhibited higher accuracy while classifying gait disorders associated with myopathy and neuropathy. Based on the literature studied, normalization, feature extraction and selection are important steps for accurately classifying gait disorders [75, 76].

Artificial neural networks (ANNs) are considered better at discovering nonlinear relationships in data. Ozsert et al. [78] classified biceps, frontalis and abductor muscles using ANN. The authors used wavelet transform for pre-processing the sEMG signal and an AR model to train the ANN. Senanayake et al. [79] used EMG RMS value and soft tissue deformation parameter (STDP) extracted from the video recordings to train a feed-forward-backward propagation neural network (FFBPN) to identify gait patterns. The proposed evaluation scheme improved classification accuracy between healthy and injured subject's gait patterns as Vastus Medialis and Lateralis revealed higher positive correlation between EMG and STDP for healthy individuals [79].

An adaptive neuro-fuzzy inference system (ANFIS) successfully diagnosed neurological disorders [8, 80]. In a number of studies, ANN and SVM worked well in diagnosing the gait pathology [7, 8, 71, 81]. Naik et al. [82] decomposed needle EMG from brachial biceps with ensemble empirical mode decomposition (EMD). The authors used Fast ICA and LDA classifier with majority voting to diagnose healthy participants from ALS, and myopathic individuals [82]. The algorithm of Naik et al. [83] for walking, sitting and standing tasks, achieved 86% classification accuracy for participants with and 96% without knee pathology. ICA via entropy bound minimization, time domain feature extraction, and feature selection with fisher score were performed prior to LDA classification. Ai et al. [30] used fused accelerometer and EMG data to discriminate among four participants including an amputee; more amputees in the study could provide better insight of the suggested technique [30].

There is no perfect machine learning algorithm to detect gait disorders. Signal processing techniques for feature extraction and selection, and standardization of the time series play a crucial role in enhancing classification accuracy. We also see

**31**

Decision tree

*A Review of EMG Techniques for Detection of Gait Disorders*

consistent improvement in the existing models with increased classification accuracy [84]. ANN classifier has some deficiencies, such as high training process time and overfitting. Extreme Machine Learning algorithm (EML) improves on these anomalies at no cost to classification accuracy [8]. SVM accuracy was low for eight

**Classifier Authors Year Conditions Classification Performance**

deformation

Gait pattern identification between healthy and injured

and osteoarthritis

EMG of healthy and rheumatoid arthritis

Gait pattern identification using stride length and cadence

Movement classification for healthy and patients with knee pathology

EMG of healthy and rheumatoid arthritis

Movement-based classification for normal and amputee subject

Gait pattern identification using stride length and cadence

Gait pattern identification using stride length and cadence

between healthy and Parkinson patients by autostep segmentation

Movement-based classification for normal and amputee subject

for daily life activities including Fall

Identification of ankle kinematic patterns for toe walkers

Accuracy = 98%

Accuracy = 89.4 ± 11.8%

Accuracy = 57 ± 1 8%

Accuracy = 94.87%

Accuracy = 86% (Unhealthy) and 96%

Accuracy = 72 ± 20%

Accuracy = 95.6 ± 2.2%

Accuracy = 93.59%

Accuracy = 96.8%

Specificity = 90% and Sensitivity = 90%

Accuracy = 98.1 ± 1.6%

Accuracy = 100%

Accuracy = 81%

(Healthy)

2014 Soft tissue

Nair et al. 2010 Osteoarthritis EMG of healthy

arthritis

pathology

arthritis

amputated

2006 Cerebral palsy

2006 Cerebral palsy

Kugler et al. 2013 Parkinson Differentiate

amputated

Xi et al. 2018 Fall Gait recognition

disorders

2006 Cerebral palsy

Nair et al. 2010 Rheumatoid

Nair et al. 2010 Rheumatoid

Ai et al. 2017 Normal and

Ai et al. 2017 Normal and

Armand et al. 2006 Toe Walking

*DOI: http://dx.doi.org/10.5772/intechopen.84403*

Senanayake et al.

Kamruzzaman and Begg.

LDA Naik et al. 2018 Knee

Kamruzzaman and Begg.

SVM Kamruzzaman and Begg.

Neural networks *Artificial Intelligence - Applications in Medicine and Biology*

*Block diagram of an EMG Signal classification system.*

classification accuracy [74, 75]. Feature level fusion is used to extract the feature space from daily life activities [73]. Patients with Parkinson's were classified with high accuracy using SVM with leave-one-out cross-validation [75]. Results from Nair et al. [76] suggest that least square kernel algorithm performed better than LDA, Neural Network, MLP and learning vector quantification (LVQ ) for patients with arthritis. Decision Tree (DT) classifier used to classify toe walking gait disorder revealed three major toe-walking patterns [77]: (1) muscle weakness of TA and quadriceps and spasticity of Tibialis Surae; (2) severe spasticity of Tibialis Surae with limited range of ankle motion; and, (3) hamstring spasticity. The MLP, on the other hand, exhibited higher accuracy while classifying gait disorders associated with myopathy and neuropathy. Based on the literature studied, normalization, feature extraction and selection are important steps for accurately classifying gait disorders [75, 76].

Artificial neural networks (ANNs) are considered better at discovering nonlinear relationships in data. Ozsert et al. [78] classified biceps, frontalis and abductor muscles using ANN. The authors used wavelet transform for pre-processing the sEMG signal and an AR model to train the ANN. Senanayake et al. [79] used EMG RMS value and soft tissue deformation parameter (STDP) extracted from the video recordings to train a feed-forward-backward propagation neural network (FFBPN) to identify gait patterns. The proposed evaluation scheme improved classification accuracy between healthy and injured subject's gait patterns as Vastus Medialis and Lateralis revealed higher positive correlation between EMG and STDP for healthy

An adaptive neuro-fuzzy inference system (ANFIS) successfully diagnosed neurological disorders [8, 80]. In a number of studies, ANN and SVM worked well in diagnosing the gait pathology [7, 8, 71, 81]. Naik et al. [82] decomposed needle EMG from brachial biceps with ensemble empirical mode decomposition (EMD). The authors used Fast ICA and LDA classifier with majority voting to diagnose healthy participants from ALS, and myopathic individuals [82]. The algorithm of Naik et al. [83] for walking, sitting and standing tasks, achieved 86% classification accuracy for participants with and 96% without knee pathology. ICA via entropy bound minimization, time domain feature extraction, and feature selection with fisher score were performed prior to LDA classification. Ai et al. [30] used fused accelerometer and EMG data to discriminate among four participants including an amputee; more amputees in the study could provide better insight of the suggested

There is no perfect machine learning algorithm to detect gait disorders. Signal processing techniques for feature extraction and selection, and standardization of the time series play a crucial role in enhancing classification accuracy. We also see

**30**

individuals [79].

**Figure 6.**

technique [30].

consistent improvement in the existing models with increased classification accuracy [84]. ANN classifier has some deficiencies, such as high training process time and overfitting. Extreme Machine Learning algorithm (EML) improves on these anomalies at no cost to classification accuracy [8]. SVM accuracy was low for eight


