scholarly journals Upper-Limb Electromyogram Classification of Reaching-to-Grasping Tasks Based on Convolutional Neural Networks for Control of a Prosthetic Hand

2021 ◽  
Vol 15 ◽  
Author(s):  
Keun-Tae Kim ◽  
Sangsoo Park ◽  
Tae-Hyun Lim ◽  
Song Joo Lee
Keyword(s):  
Upper Limb ◽  
Healthy Subjects ◽  
Principal Component ◽  
Upper Body ◽  
Support Vector ◽  
Assessment Procedure ◽  
Robotic Hands ◽  
Upper Arm ◽  
Auto Regressive ◽  
People With Physical Disabilities

In recent years, myoelectric interfaces using surface electromyogram (EMG) signals have been developed for assisting people with physical disabilities. Especially, in the myoelectric interfaces for robotic hands or arms, decoding the user’s upper-limb movement intentions is cardinal to properly control the prosthesis. However, because previous experiments were implemented with only healthy subjects, the possibility of classifying reaching-to-grasping based on the EMG signals from the residual limb without the below-elbow muscles was not investigated yet. Therefore, we aimed to investigate the possibility of classifying reaching-to-grasping tasks using the EMG from the upper arm and upper body without considering wrist muscles for prosthetic users. In our study, seven healthy subjects, one trans-radial amputee, and one wrist amputee were participated and performed 10 repeatable 12 reaching-to-grasping tasks based on the Southampton Hand Assessment Procedure (SHAP) with 12 different weighted (light and heavy) objects. The acquired EMG was processed using the principal component analysis (PCA) and convolutional neural network (CNN) to decode the tasks. The PCA–CNN method showed that the average accuracies of the healthy subjects were 69.4 ± 11.4%, using only the EMG signals by the upper arm and upper body. The result with the PCA–CNN method showed 8% significantly higher accuracies than the result with the widely used time domain and auto-regressive-support vector machine (TDAR–SVM) method as 61.6 ± 13.7%. However, in the cases of the amputees, the PCA–CNN showed slightly lower performance. In addition, in the aspects of assistant daily living, because grip force is also important when grasping an object after reaching, the possibility of classifying the two light and heavy objects in each reaching-to-grasping task was also investigated. Consequently, the PCA–CNN method showed higher accuracy at 70.1 ± 9.8%. Based on our results, the PCA–CNN method can help to improve the performance of classifying reaching-to-grasping tasks without wrist EMG signals. Our findings and decoding method can be implemented to further develop a practical human–machine interface using EMG signals.

Variability in Muscle Recruitment Strategy Between Operators During Assisted Assembly Tasks

2018 ◽  
Author(s):  
Yingxin Qiu ◽  
Keerthana Murali ◽  
Jun Ueda ◽  
Atsushi Okabe ◽  
Dalong Gao
Keyword(s):  
Upper Limb ◽  
Lower Limb ◽  
Wrist Joint ◽  
Principal Component ◽  
Classification Performance ◽  
Body Motion ◽  
Recruitment Strategy ◽  
Support Vector ◽  
Recruitment Strategies ◽  
Muscle Recruitment

This paper reports the variability in muscle recruitment strategies among individuals who operate a non-powered lifting device for general assembly (GA) tasks. Support vector machine (SVM) was applied to the classification of motion states of operators using electromyography (EMG) signals collected from a total of 15 upper limb, lower limb, shoulder, and torso muscles. By comparing the classification performance and muscle activity features, variability in muscle recruitment strategy was observed from lower limb and torso muscles, while the recruitment strategies of upper limb and shoulder muscles were relatively consistent across subjects. Principal component analysis (PCA) was applied to identify key muscles that are highly correlated with body movements. Selected muscles at the wrist joint, ankle joint and scapula are considered to have greater significance in characterizing the muscle recruitment strategies than other investigated muscles. PCA loading factors also indicate the existence of body motion redundancy during typical pick-and-place tasks.

Identification and classification of upper limb motions using PCA

2018 ◽  
Vol 63 (2) ◽  
pp. 191-196 ◽  
Author(s):  
Karan Veer ◽  
Renu Vig
Keyword(s):  
Upper Limb ◽  
Principal Component ◽  
Measured Data ◽  
The Other ◽  
Signal Acquisition ◽  
Myoelectric Signal ◽  
Triceps Brachii ◽  
Upper Arm ◽  
Semg Signals

Abstract:This paper describes the utility of principal component analysis (PCA) in classifying upper limb signals. PCA is a powerful tool for analyzing data of high dimension. Here, two different input strategies were explored. The first method uses upper arm dual-position-based myoelectric signal acquisition and the other solely uses PCA for classifying surface electromyogram (SEMG) signals. SEMG data from the biceps and the triceps brachii muscles and four independent muscle activities of the upper arm were measured in seven subjects (total dataset=56). The datasets used for the analysis are rotated by class-specific principal component matrices to decorrelate the measured data prior to feature extraction.

scholarly journals Effect of vibration characteristics and vibror arrangement on the tactile perception of the upper arm in healthy subjects and upper limb amputees

2019 ◽  
Vol 16 (1) ◽  
Author(s):  
Matthieu Guemann ◽  
Sandra Bouvier ◽  
Christophe Halgand ◽  
Florent Paclet ◽  
Leo Borrini ◽  
...  
Keyword(s):  
Upper Limb ◽  
Healthy Subjects ◽  
Longitudinal Axis ◽  
Vibration Characteristics ◽  
Spatial Discrimination ◽  
Sensory Substitution ◽  
Limb Amputation ◽  
Vibrotactile Stimulation ◽  
Upper Arm ◽  
Intensity Perception

Abstract Background Vibrotactile stimulation is a promising venue in the field of prosthetics to retrain sensory feedback deficits following amputation. Discrimination is well established at the forearm level but not at the upper arm level. Moreover, the effects of combining vibration characteristics such as duration and intensity has never been investigated. Method We conducted experiments on spatial discrimination (experiment 1) and tactile intensity perception (experiment 2), using 9 combinations of 3 intensities and 3 durations of vibror stimulations device. Those combinations were tested under 4 arrangements with an array of 6 vibrors. In both experiments, linear orientation aligned with the upper arm longitudinal axis were compared to circular orientation on the upper arm circumference. For both orientations, vibrors were placed either with 3cm space between the center of 2 vibrors or proportionally to the length or the circumference of the subject upper arm. Eleven heathy subjects underwent the 2 experiments and 7 amputees (humeral level) participated in the spatial discrimination task with the best arrangement found. Results Experiment 1 revealed that circular arrangements elicited better scores than the linear ones. Arrangements with vibrors spaced proportionally elicited better scores (up to 75% correct) than those with 3 cm spacing. Experiment 2, showed that the perceived intensity of the vibration increases with the intensity of the vibrors’ activation, but also with their duration of activation. The 7 patients obtained high scores (up to 91.67% correct) with the circular proportional (CP) arrangement. Discussion These results highlight that discrete and short vibrations can be well discriminated by healthy subjects and people with an upper limb amputation. These new characteristics of vibrations have great potential for future sensory substitution application in closed-loop prosthetic control.

scholarly journals Morphological factors extraction of upper limb for young female

2021 ◽  
Vol 237 ◽  
pp. 01032
Author(s):  
Xin He ◽  
Hongshu Jin
Keyword(s):  
Principal Component Analysis ◽  
Upper Limb ◽  
Measurement Method ◽  
Principal Component ◽  
Young Female ◽  
Upper Body ◽  
Critical Parameters ◽  
Independent Factor ◽  
Oblique Angle ◽  
Morphological Variables

This paper is proposed to extract the morphological factors of upper limb shapes related to the personalized sleeve design. The 36 items of lateral upper limb morphology were measured for 50 young female aged 19 to 22 years old by the photo measurement method. Based on the correlation relationships of morphological variables of upper limb, there are 4 morphological factors of upper limb shapes with eigenvalues above 1 were extracted by the principal component analysis, and the cumulative variance reached 83.504%. Among them, the factors of girth and height of upper limb were relevance to the girths of torso and stature respectively, and the factor of upper limb oblique angle is influenced by the upper body axis inclination, and the arm root height is interpreted as an independent factor describing the arm root shape. These morphological factors provided the references in key feature indicator sifting for upper limb shape subdivisions and the critical parameters in personalized sleeve structure designing.

Reflex Responses in Upper Limb Muscles to Cutaneous Stimuli

1993 ◽  
Vol 20 (04) ◽  
pp. 271-278 ◽  
Author(s):  
R. Chen
Keyword(s):  
Stimulus Intensity ◽  
Upper Limb ◽  
Healthy Subjects ◽  
Voluntary Contraction ◽  
Cutaneous Reflexes ◽  
Focal Lesions ◽  
Normal Ranges ◽  
Conduction Velocities ◽  
Limb Muscles ◽  
Biphasic Responses

ABSTRACT:Cutaneous reflexes in the upper limb were elicited by stimulating digital nerves and recorded by averaging rectified EMG from proximal and distal upper limb muscles during voluntary contraction. Distal muscles often showed a triphasic response: an inhibition with onset about 50 ms (Il) followed by a facilitation with onset about 60 ms (E2) followed by another inhibition with onset about 80 ms (12). Proximal muscles generally showed biphasic responses beginning with facilitation or inhibition with onset at about 40 ms. Normal ranges for the amplitude of these components were established from recordings on 22 arms of 11 healthy subjects. An attempt was made to determine the alterent fibers responsible for the various components by varying the stimulus intensity, by causing ischemic block of larger fibers and by estimating the afferent conduction velocities. The central pathways mediating these reflexes were examined by estimating central delays and by studying patients with focal lesions

QSAR Study of PARP Inhibitors by GA-MLR, GA-SVM and GA-ANN Approaches

2020 ◽  
Vol 16 (8) ◽  
pp. 1088-1105
Author(s):  
Nafiseh Vahedi ◽  
Majid Mohammadhosseini ◽  
Mehdi Nekoei
Keyword(s):  
Present Report ◽  
Principal Component ◽  
Parp Inhibitors ◽  
Support Vector ◽  
Ann Model ◽  
Statistical Parameters ◽  
Qsar Study ◽  
Data Set ◽  
Test Set ◽  
Non Linear

Background: The poly(ADP-ribose) polymerases (PARP) is a nuclear enzyme superfamily present in eukaryotes. Methods: In the present report, some efficient linear and non-linear methods including multiple linear regression (MLR), support vector machine (SVM) and artificial neural networks (ANN) were successfully used to develop and establish quantitative structure-activity relationship (QSAR) models capable of predicting pEC50 values of tetrahydropyridopyridazinone derivatives as effective PARP inhibitors. Principal component analysis (PCA) was used to a rational division of the whole data set and selection of the training and test sets. A genetic algorithm (GA) variable selection method was employed to select the optimal subset of descriptors that have the most significant contributions to the overall inhibitory activity from the large pool of calculated descriptors. Results: The accuracy and predictability of the proposed models were further confirmed using crossvalidation, validation through an external test set and Y-randomization (chance correlations) approaches. Moreover, an exhaustive statistical comparison was performed on the outputs of the proposed models. The results revealed that non-linear modeling approaches, including SVM and ANN could provide much more prediction capabilities. Conclusion: Among the constructed models and in terms of root mean square error of predictions (RMSEP), cross-validation coefficients (Q2 LOO and Q2 LGO), as well as R2 and F-statistical value for the training set, the predictive power of the GA-SVM approach was better. However, compared with MLR and SVM, the statistical parameters for the test set were more proper using the GA-ANN model.

Application of Machine Learning in Animal Disease Analysis and Prediction

2020 ◽  
Vol 15 ◽  
Author(s):  
Shuwen Zhang ◽  
Qiang Su ◽  
Qin Chen
Keyword(s):  
Machine Learning ◽  
Unsupervised Learning ◽  
Supervised Learning ◽  
Clustering Algorithm ◽  
Principal Component ◽  
Support Vector ◽  
Animal Disease ◽  
Human Beings ◽  
Animal Diseases ◽  
Disease Analysis

Abstract: Major animal diseases pose a great threat to animal husbandry and human beings. With the deepening of globalization and the abundance of data resources, the prediction and analysis of animal diseases by using big data are becoming more and more important. The focus of machine learning is to make computers learn how to learn from data and use the learned experience to analyze and predict. Firstly, this paper introduces the animal epidemic situation and machine learning. Then it briefly introduces the application of machine learning in animal disease analysis and prediction. Machine learning is mainly divided into supervised learning and unsupervised learning. Supervised learning includes support vector machines, naive bayes, decision trees, random forests, logistic regression, artificial neural networks, deep learning, and AdaBoost. Unsupervised learning has maximum expectation algorithm, principal component analysis hierarchical clustering algorithm and maxent. Through the discussion of this paper, people have a clearer concept of machine learning and understand its application prospect in animal diseases.

scholarly journals The Dynamic Thumb-in-Palm Pattern in Children with Spastic Cerebral Palsy and Its Effects on Upper Limb Function

Children ◽  
2020 ◽  
Vol 8 (1) ◽  
pp. 17
Author(s):  
Ja Young Choi ◽  
Dong-Wook Rha ◽  
Seon Ah Kim ◽  
Eun Sook Park
Keyword(s):  
Cerebral Palsy ◽  
Upper Limb ◽  
Multivariable Analysis ◽  
Spastic Cerebral Palsy ◽  
Type I ◽  
Upper Arm ◽  
Limb Function ◽  
Upper Limb Function ◽  
Limb Deformities ◽  
Arm Function

The thumb-in-palm (TIP) pattern is one of the most common upper limb deformities in cerebral palsy (CP). This study was designed to investigate the effect of the dynamic TIP pattern on upper limb function in children with spastic CP. This prospective observational study included a total of 106 children with CP with dynamic TIP. The House TIP classification while grasping small or large objects, Melbourne Assessment of Unilateral Upper Limb Function (MUUL), Shriners Hospital Upper Extremity Evaluation (SHUEE), Zancolli classification for wrist–finger flexor deformity, and degree of swan neck deformity were assessed. Type I was the most common and highest functioning House TIP classification type. However, there were no significant differences in upper arm function between types II, III, and IV. The three components of the SHUEE showed stronger association with MUUL than House TIP and Zancolli classifications. After multivariable analysis, functional use of the wrist–finger and the thumb played a more significant role than the dynamic alignment of the thumb. In conclusion, the House TIP classification is useful to describe the TIP pattern. The SHUEE thumb assessment is a useful tool for reflecting upper arm function. The upper arm function was related more with the associated wrist flexor deformity than dynamic TIP.

Sign in / Sign up

Export Citation Format

Share Document