A survey on the feasibility of surface EMG in facial pacing

Rantanen, Ville; Ilves, Mirja; Vehkaoja, Antti; Kontunen, Anton; Lylykangas, Jani; Mäkelä, E. Antero; Rautiainen, Markus; Surakka, Veikko; Lekkala, Jukka

doi:10.1109/embc.2016.7591040

Cited by 15 publications

(11 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In closed-loop methods, the system detects eye blink -related signals from the healthy side of the face in order to trigger concurrent blink on the paralyzed side eye. The measurement methods include, for example, electromyography (EMG), electrooculography (EOG), and the use of infrared (IR) emitter-sensor pair or gyroscopic sensor in detecting eye closure [3,4,5,6]. Previous studies have shown that stimulation of the orbicularis oculi muscle and/or facial nerve branches using transcutaneous electrodes can be useful in restoring eye blink in about 50% of the cases [7].…”

Section: A Facial Pacing Methodsmentioning

confidence: 99%

Artificial Eye Blink Pacemaker - A First Investigation into the Blink Production Using Constant-Interval Electrical Stimulation

et al. 2017

Self Cite

View full text Add to dashboard Cite

Facial paralysis due to damage of the facial nerve affects the function of facial muscles, including the muscles responsible for eye blinking. The absence of the eye blink can lead to severe and permanent corneal damage as the protection and lubrication of the eye is decreased. Thus, it would be highly important to provide an aid to sustain the eye health. The present aim was to study the effects of long-term electrical eye blink stimulation using a facial stimulation prototype. Five healthy participants watched a movie for 78 minutes, while the eye blinks were produced to their left eye by pre-programmed, timer-triggered blink stimulation at fixed intervals. We analyzed the functionality of the stimulation prototype, potential changes in the quality of the produced blinks, and the ratings of experiences in terms of pain, discomfort, and naturalness. We also analyzed the acuity of vision before and after the stimulation. The results showed that the stimulated eye blink was rated as not painful, somewhat uncomfortable, and slightly unnatural. With three participants the stimulation evoked a full eye closure throughout the study, and with two participants, the stimulation evoked partial blink after some time. Further, on four of the cases, the vision of the stimulated eye was better after the movie than before it. The participants told that the stimulation did not disturb the movie watching. As the findings were promising, the next steps include more comprehensive tests both with intact participants and with persons having an acute facial paralysis.

show abstract

Section: A Facial Pacing Methodsmentioning

confidence: 99%

Artificial Eye Blink Pacemaker - A First Investigation into the Blink Production Using Constant-Interval Electrical Stimulation

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our main goal is to provide a dataset that is suited for deep learning network-based EEG denoising study. In this respect, we acquired EEG, EOG and EMG data from several publicly available data repositories [53,54,55,56,57,58,59,60]. Then, we preprocessed the data, and segmented them into epochs of 2 seconds; afterwards, we scaled the epochs to the same variance and visually checked them to ensure they were clean and ready for use; lastly, we uploaded the epochs of each category to a publicly available repository named EEGdenoiseNet .…”

Section: Data Acquisition and Preprocessingmentioning

confidence: 99%

“…For the myogenic artifact epochs (Figure 1c), a facial surface electromyography (EMG) dataset was used [60]. We choose facial EMG because they are the main sources of myogenic artefacts.…”

Section: Data Acquisition and Preprocessingmentioning

confidence: 99%

EEGdenoiseNet: A benchmark dataset for end-to-end deep learning solutions of EEG denoising

Zhang¹,

Zhao²,

Chen³

et al. 2020

Preprint

View full text Add to dashboard Cite

Deep learning networks are increasingly attracting attention in several fields. Among other applications, deep learning models have been used for denoising of electroencephalography (EEG) data. These models provided comparable performance with that of traditional techniques. At present, however, the lack of well-structured, standardized datasets with specific benchmark limits the development of deep learning solutions for EEG denoising. Here, we present EEGdenoiseNet, a benchmark EEG dataset that is suited for training and testing deep learning-based denoising models. This permits making valid comparisons across different models. EEGdenoiseNet contains 4514 clean EEG epochs, 3400 ocular artifact epochs and 5598 muscular artifact epochs. This allow the user to produce a large number of noisy EEG epochs with ground truth for model training and testing. We used EEGdenoiseNet to evaluate the performance of four classical deep learning networks (a fully-connected network, a simple convolution network, a complex convolution network and a recurrent neural network). Our analysis suggested that deep learning methods have great potential for EEG denoising even under high noise contamination. Through EEGdenoiseNet, we hope to accelerate the development of the emerging field of deep learning-based EEG denoising. The dataset and the code for benchmarking deep learning networks are publicly available on github (https://github.com/ncclabsustech/EEGdenoiseNet).

show abstract

“…It produces a stochastic signal whose amplitude reflects the intensity of muscle activations. 5 It is a noninvasive technique because it requires the use of pre-gelled silver chloride electrodes to make the electrical contact between the skin and the sensor. It is usually well tolerated by patients, even youngest ones.…”

Section: Introductionmentioning

confidence: 99%

Activity Evaluation of Facial Muscles by Surface Electromyography

Bertozzi

Bianchi

Salvagni

et al. 2020

Plastic and Reconstructive Surgery - Global Open

View full text Add to dashboard Cite

Background: Surface electromyography (sEMG) is an easy, noninvasive, and reproducible way to assess spontaneous electrical activity of muscles in real time. In this study, we report data on the correlation between sEMG and mimetic muscle activity during specific tasks so as to create a case–control reference for future studies on acute, chronic, and congenital facial palsy. Methods: Twelve healthy participants were enrolled (6 women and 6 men) with a mean age of 42.75 (range, 26–58 years) years. sEMG signals were recorded at rest and while performing voluntary and specific tasks that elicited selective contraction of frontalis, orbicularis oculi, zygomaticus major, orbicularis oris, and platysma muscles simultaneously and bilaterally for each muscle group. Statistical analysis was performed to evaluate if there was a statistically significant difference of the average contraction values between left- and right-side data sets and between male and female participants. Results: No statistically significant difference was found between male and female participants or between the right and left sides at rest and while performing the requested tasks, even though they were not identical. No participant complained about the procedure. Conclusions: Interindividual and intraindividual variability of the sEMG signal as well as crosstalk between muscles groups were reported downsides that we did not encounter. The absolute noninvasiveness of our procedure makes it feasible to be applied even to young children. This dataset obtained in healthy participants might also be useful in the observation of patients undergoing regeneration/reinnervation procedure following recently acquired facial palsy or mimetic muscle reconstruction for congenital/inveterate one.

show abstract

A survey on the feasibility of surface EMG in facial pacing

Cited by 15 publications

References 15 publications

Artificial Eye Blink Pacemaker - A First Investigation into the Blink Production Using Constant-Interval Electrical Stimulation

Artificial Eye Blink Pacemaker - A First Investigation into the Blink Production Using Constant-Interval Electrical Stimulation

EEGdenoiseNet: A benchmark dataset for end-to-end deep learning solutions of EEG denoising

Activity Evaluation of Facial Muscles by Surface Electromyography

Contact Info

Product

Resources

About