Evaluating subsampling strategies for sEMG-based prediction of voluntary muscle contractions

Kõiva, Risto; Hilsenbeck, Barbara; Castellini, Claudio

doi:10.1109/icorr.2013.6650492

Cited by 9 publications

(5 citation statements)

References 16 publications

(29 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Patterns involving the individuated activation of the four fingers (patterns 1-4) are all characterized by high performance, while patterns involving the thumb or simultaneous activation of multiple digits show considerably worse performance. This difficulty of predicting thumb activations was observed as well by Kõiva et al [29]. A likely explanation for this phenomenon is that in our acquisition no sEMG activity is recorded from the majority of thumb muscles.…”

Section: A Force Regressionsupporting

confidence: 72%

Movement Error Rate for Evaluation of Machine Learning Methods for sEMG-Based Hand Movement Classification

Gijsberts

Atzori

Castellini

et al. 2014

IEEE Trans. Neural Syst. Rehabil. Eng.

Self Cite

169

148

View full text Add to dashboard Cite

There has been increasing interest in applying learning algorithms to improve the dexterity of myoelectric prostheses. In this work, we present a large-scale benchmark evaluation on the second iteration of the publicly released NinaPro database, which contains surface electromyography data for 6 DOF force activations as well as for 40 discrete hand movements. The evaluation involves a modern kernel method and compares performance of three feature representations and three kernel functions. Both the force regression and movement classification problems can be learned successfully when using a nonlinear kernel function, while the exp- χ(2) kernel outperforms the more popular radial basis function kernel in all cases. Furthermore, combining surface electromyography and accelerometry in a multimodal classifier results in significant increases in accuracy as compared to when either modality is used individually. Since window-based classification accuracy should not be considered in isolation to estimate prosthetic controllability, we also provide results in terms of classification mistakes and prediction delay. To this extent, we propose the movement error rate as an alternative to the standard window-based accuracy. This error rate is insensitive to prediction delays and it allows us therefore to quantify mistakes and delays as independent performance characteristics. This type of analysis confirms that the inclusion of accelerometry is superior, as it results in fewer mistakes while at the same time reducing prediction delay.

show abstract

Section: A Force Regressionsupporting

confidence: 72%

Movement Error Rate for Evaluation of Machine Learning Methods for sEMG-Based Hand Movement Classification

Gijsberts

Atzori

Castellini

et al. 2014

IEEE Trans. Neural Syst. Rehabil. Eng.

Self Cite

169

148

View full text Add to dashboard Cite

show abstract

“…Surprisingly, our results show that this device can be used to obtain a precision of up to 1.5 Newtons in the prediction of the fingertip forces. Such a precision is comparable to that obtained with sEMG [13], [7]. The device is wearable, weighing about 65 grams, and costs as a prototype less than 50 EUR.…”

Section: Introductionsupporting

confidence: 61%

A wearable low-cost device based upon Force-Sensing Resistors to detect single-finger forces

Castellini

Ravindra

2014

5th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics

Self Cite

View full text Add to dashboard Cite

In rehabilitation robotics it is highly desirable to find novel human-machine interfaces for the disabled, in particular to substitute or augment surface electromyography (sEMG), trying to keep at the same time its easiness of use, precision and non-invasiveness. In this paper we design and demonstrate one such device, based upon Force-Sensing Resistors (FSRs). An array of 10 FSRs was wrapped around the proximal section of the forearm of ten intact subjects engaged in pressing on an accurate force sensor with their fingers (this includes the rotation of the thumb). The FSRs would detect the forearm surface deformations induced by muscle activity; the signals provided by the FSRs were then matched to the recorded forces. The experimental results show that finger forces can be predicted using this device with the same accuracy obtained in literature using sEMG. The device, even as an academic prototype, weighs about 65 grams and costs around 50 EUR. Thus, it is remarkably light and cheap in comparison to standard sEMG electrode arrays.

show abstract

“…The classification was performed on all movements (rest included) and is balanced according to movement number repetitions. Before performing the classification, the data from database 1 were preprocessed using a 1st order Butterworth low-pass filter with a 1 Hz cutoff frequency 44 .…”

Section: Technical Validationmentioning

confidence: 99%

Electromyography data for non-invasive naturally-controlled robotic hand prostheses

et al. 2014

Self Cite

View full text Add to dashboard Cite

Recent advances in rehabilitation robotics suggest that it may be possible for hand-amputated subjects to recover at least a significant part of the lost hand functionality. The control of robotic prosthetic hands using non-invasive techniques is still a challenge in real life: myoelectric prostheses give limited control capabilities, the control is often unnatural and must be learned through long training times. Meanwhile, scientific literature results are promising but they are still far from fulfilling real-life needs. This work aims to close this gap by allowing worldwide research groups to develop and test movement recognition and force control algorithms on a benchmark scientific database. The database is targeted at studying the relationship between surface electromyography, hand kinematics and hand forces, with the final goal of developing non-invasive, naturally controlled, robotic hand prostheses. The validation section verifies that the data are similar to data acquired in real-life conditions, and that recognition of different hand tasks by applying state-of-the-art signal features and machine-learning algorithms is possible.

show abstract

Evaluating subsampling strategies for sEMG-based prediction of voluntary muscle contractions

Cited by 9 publications

References 16 publications

Movement Error Rate for Evaluation of Machine Learning Methods for sEMG-Based Hand Movement Classification

Movement Error Rate for Evaluation of Machine Learning Methods for sEMG-Based Hand Movement Classification

A wearable low-cost device based upon Force-Sensing Resistors to detect single-finger forces

Electromyography data for non-invasive naturally-controlled robotic hand prostheses

Contact Info

Product

Resources

About