The influence of confounding factors on the relationship between muscle contraction level and MF and MPF values of EMG signal: a review

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

Discrete myoelectric control-based gesture recognition has recently gained interest as a possible input modality for many emerging ubiquitous computing applications. Unlike the continuous control commonly employed in powered prostheses, discrete systems seek to recognize the dynamic sequences associated with gestures to generate event-based inputs. More akin to those used in general-purpose human-computer interaction, these could include, for example, a flick of the wrist to dismiss a phone call or a double tap of the index finger and thumb to silence an alarm. Moelectric control systems have been shown to achieve near-perfect classification accuracy, but in highly constrained offline settings. Real-world, online systems are subject to 'confounding factors' (i.e. factors that hinder the real-world robustness of myoelectric control that are not accounted for during typical offline analyses), which inevitably degrade system performance, limiting their practical use. Although these factors have been widely studied in continuous prosthesis control, there has been little exploration of their impacts on discrete myoelectric control systems for emerging applications and use cases. Correspondingly, this work examines, for the first time, three confounding factors and their effect on the robustness of discrete myoelectric control: (1)limb position variability, (2)cross-day use, and a newly identified confound faced by discrete systems (3)gesture elicitation speed. Results from four different discrete myoelectric control architectures: (1) Majority Vote LDA, (2) Dynamic Time Warping, (3) an LSTM network trained with Cross Entropy, and (4) an LSTM network trained with Contrastive Learning, show that classification accuracy is significantly degraded (p<0.05) as a result of each of these confounds. This work establishes that confounding factors are a critical barrier that must be addressed to enable the real-world adoption of discrete myoelectric control for robust and reliable gesture recognition.

Are proximal and distal neuromuscular parameters able to predict hip and knee frontal plane kinematics during single-leg landing?

OBJECTIVE

To determine if proximal and distal neuromuscular parameters (EMG amplitude and median frequency - MDF) can predict frontal plane kinematics during single-leg landing.

STUDY DESIGN

Cross sectional study.

SETTING

Laboratory.

PARTICIPANTS

Fifteen participants (7 female) performed six single-leg landings with measures of frontal plane kinematics and EMG obtained 230 ms after first foot contact, totalizing 90 landings.

MAIN OUTCOME MEASURES

(i) 2D hip adduction [hip ADD] and knee frontal plane projection angle [knee FPPA]; (ii) EMG amplitude and MDF of gluteus medius [GMed], tensor fascia latae [TFL], peroneus longus [PL] and tibialis anterior [TA].

RESULTS

We observed that MDF of TA was a significant predictor of hip ADD (p = 0.037; β = -0.049 Hz; R²c = 0.30). Also, MDF of PL was significant predictor of knee FPPA (p = 0.043; β = 0.042 Hz; R²c = 0.37). Hip muscles and EMG amplitude parameters were not considered predictors of frontal plane kinematics.

CONCLUSION

The firing frequency of ankle muscles predicted the variance of hip and knee frontal plane kinematics during single-leg landing.

Surface EMG Statistical and Performance Analysis of Targeted-Muscle-Reinnervated (TMR) Transhumeral Prosthesis Users in Home and Laboratory Settings

A pattern-recognition (PR)-based myoelectric control system is the trend of future prostheses development. Compared with conventional prosthetic control systems, PR-based control systems provide high dexterity, with many studies achieving >95% accuracy in the last two decades. However, most research studies have been conducted in the laboratory. There is limited research investigating how EMG signals are acquired when users operate PR-based systems in their home and community environments. This study compares the statistical properties of surface electromyography (sEMG) signals used to calibrate prostheses and quantifies the quality of calibration sEMG data through separability indices, repeatability indices, and correlation coefficients in home and laboratory settings. The results demonstrate no significant differences in classification performance between home and laboratory environments in within-calibration classification error (home: 6.33 ± 2.13%, laboratory: 7.57 ± 3.44%). However, between-calibration classification errors (home: 40.61 ± 9.19%, laboratory: 44.98 ± 12.15%) were statistically different. Furthermore, the difference in all statistical properties of sEMG signals is significant (p < 0.05). Separability indices reveal that motion classes are more diverse in the home setting. In summary, differences in sEMG signals generated between home and laboratory only affect between-calibration performance.

Effect of fatigue on kinematics, kinetics and muscle activities of lower limbs during gait

Muscle fatigue, as a serious social problem, affects the performance of daily living activities, especially for workers. Decrease of movement control ability caused by muscle fatigue is one of the risk and intrinsic factors for occupational accidents, such as slips, trips, falls, etc. In order to reduce the accident rate and optimize the existing prevention measures, it is necessary to investigate the effect of fatigue on kinematics, kinetics, and muscle activities of human body. In this paper, 26 healthy participants were recruited. The kinematics and kinetics analysis of lower limb joints, and surface electromyograms (sEMG) time-domain and frequency-domain analysis of lower limb periarticular muscles were utilized to investigate the effects of muscle fatigue. The results showed that the fatigue reduced the range-of-motion (RoM) of the lower limb joints. Smaller plantarflexion, knee flexion and hip flexion angles, and greater dorsiflexion angles were observed after fatigue. For the joint moment, the fatigue did not alter the joint moments except for the smaller knee flexion moment. For the joint power, the fatigue decreased the generation power of ankle, knee and hip joint and the absorption power of ankle and knee joints, whereas increased the absorption power of hip joint. Besides, the fatigue increased the normalized integrated sEMG (iEMG) and root-mean-square (RMS) of sEMG, and shifted the median frequency (MF) and mean power frequency (MPF) of sEMG toward lower frequencies. The results from the present study concluded that the muscle fatigue changed the kinematics, kinetics and muscle activities of lower limbs during gait, and then could increase the risk rate of occupational accidents.

Effects of Three Needling Manipulations of Zusanli (ST 36) on Deqi Sensations and Surface Myoelectricity in Healthy Participants

OBJECTIVE

To investigate the effects of different acupuncture manipulations on Deqi sensations and surface myoelectricity, and explore the correlation between Deqi sensations and needling manipulations.

METHODS

Forty-five healthy participants accepted twirling, lifting-thrusting, and twirling plus lifting-thrusting manipulanions at right Zusanli (ST 36), respectively. The acupuncturist's and participants' Deqi sensations were collected by MGH Acupuncture Sensation Scale (MASS). The intensity and occurrence rate of soreness, dull pain, pressure, heaviness, fullness, numbness, sharp pain, warmth, coolness, and throbbing feelings of participants, and tightness, smooth, and tangle feelings of acupuncturist were measured. The correlation between the acupuncturist's and participant's Deqi sensations was analyzed. Surface electromyogram (EMG) were recorded before, during and after needling in 30 participants. The integrated EMG (iEMG), mean power frequency (MPF) and media frequency (MF) were analyzed.

RESULT

Both fullness and soreness of participants and tightness of acupuncturist were the most frequently occurred ones. A positive correlation between participants' fullness and acupuncturist's tightness was observed during the three aforementioned needling manipulations (P<0.05, OR>1). Almost all the needling sensations measured in the present study could be induced by the three needling manipulations. However, strength of Deqi sensations was exhibited as lifting-thrusting > twirling plus lifting-thrusting > twirling according to MASS index. The iEMG values were increased and MPF, MF values were decreased during needling compaired to those before needling, especially during lifting-thrusting (P<0.01).

CONCLUSIONS

The intensity and occurrence rate of the different Deqi sensations induced by different needling manipulations were basically similar. The fullness and soreness were both the most frequently induced Deqi sensations. The strongest Deqi sensation could be induced by lifting-thrusting manipulation. There is a positive correlation between participants' fullness and acupuncturist's tightness during the three needling manipulations. The myoelectricity around the acupoint is related to Deqi responses. (Registration No. AMCTR-IOR-20000314).

Effects of Different Sling Settings on Electromyographic Activities of Selected Trunk Muscles: A Preliminary Research

Introduction

The supine and prone sling exercise may facilitate activation of the local trunk muscles. Does the side-lying sling exercise activate trunk muscles more easily than the supine and prone training with sling settings? Clinical work has shown that the side-lying sling exercise could reduce pain in patients with unilateral low back pain (LBP), but the mechanism behind it is unclear. The fundamental purpose of this preliminary study was to examine the electromyography (EMG) characteristics of trunk muscles during different sling lumbar settings on sixteen healthy adults.

Methods

Amplitude and mean power frequency (MPF) of EMG signals were recorded from the transversus abdominis (TA), rectus abdominis (RA), multifidus (MF), erector spinae (ES), gluteus maximus (Gmax), and gluteus medius (Gmed) muscles while the subjects performed the supine lumbar setting (SLS), prone lumbar setting (PLS), left side-lying lumbar setting (LSLS), and right side-lying lumbar setting (RSLS).

Results

During SLS and PLS, TA and MF showed significantly higher activity than RA and ES on the same side, respectively. The EMG activities of ES, TA, MF, Gmax, and Gmed had significant differences between the different sides during LSLS and RSLS, and the dominant-side muscles showed higher activity than the other side. There was no significant difference in core trunk muscles between different sling lumbar settings-only that the SLS of the MF/ES ratio was significantly higher than LSLS and RSLS.

Conclusions

Sling exercises can be an effective measure to enhance MF and TA EMG activity, and the side-lying position can increase dominant-side Gmax and Gmed activity. Side-lying sling training does not activate more core muscles than the supine and prone training. Supine and prone exercise should be preferred over SLT to stabilize the lumbar region because of its high local/global muscle ratio.

Single session exercises and concurrent functional electrical stimulation are more effective on muscles' force generation than only exercises in spinal cord injured persons: a feasibility study

OBJECTIVES

To evaluate impact of first therapy session, containing functional electrical stimulation (FES) and therapeutic exercises (TE) on erector spinae (ES) and rectus abdominis (RA) force generation in persons with spinal cord injury (SCI).

METHODS

Five men with SCI were divided in two groups - FES+TE received concurrent FES on ES and RA and TE, TE only TE. Participants performed exercises for improving sitting balance and posture. Muscles' electrical activity was evaluated by electromyography; amplitude (AEMG) and median frequency (MF) were used for analysis.

RESULTS

AEMG of ES left (L) increased 292.9% (g=-0.92), right (R) 175% (g=-1.01), RA L 314.3% (g=-0,81, P<0.05), R 266.7% (g=-0.08) in FES+TE. AEMG of ES L increased 47.6% (g=-0.46), R 96.4% (g=-0.95); RA L 7.1% (g=-0.97), but R decreased 6.7% (g=0.12) in TE. MF of ES L increased 108.5% (g=-0.74), R 184% (g=-1.25); RA L 886.7% (g=3-05, P<0.05), R 817.6% (g=-2.55, P<0.05) in FES+TE. MF of ES L increased 95.2% (g=-1.02), R 161.4% (g=-1.64); RA L 3,2% (g=-0.06), R 30.8% (g=-0.46) in TE.

CONCLUSIONS

In SCI persons, single session exercises and concurrent functional electrical stimulation may be more effective on muscles` force generation than only exercises. However, replication of the results is needed before clinical implementation.

Comparative Study of Kinematics and Muscle Activity Between Elite and Amateur Table Tennis Players During Topspin Loop Against Backspin Movements

This study investigated differences of lower limb kinematics and muscle activity during table tennis topspin loop against backspin movements between elite players (EPs) and amateur players (APs). Ten EPs and ten APs performed crosscourt backhand loop movements against the backspin ball with maximal power. Vicon motion analysis and a MEGA ME6000 system was used to capture kinematics and surface EMG data. The motion was divided into two phases, including the backswing and swing. The joints' flexion and extension angle tendency between EPs and APs differed significantly. The coefficient of multiple correlation (CMC) values for EPs were all beyond 0.9, indicating high similarity of joint angles change. APs presented moderate similarity with CMC values from 0.5 to 0.75. Compared to APs, EPs presented larger ankle eversion, knee and hip flexion at the beginning moment of the backswing. In the sEMG test, EPs presented smaller standardized AEMG (average electromyography) of the lower limb muscles in the rectus femoris and tibia anterior on both sides. Additionally, the maximum activation of each muscle for EPs was smaller and MPF (mean power frequency) of the lower limb was greater during the whole movement. The present study revealed that EPs could complete this technical motion more economically than APs, meanwhile, EPs were more efficient in muscle usage and showed better balance ability.

Core and skin temperature influences on the surface electromyographic responses to an isometric force and position task

The large body of work demonstrating hyperthermic impairment of neuromuscular function has utilized maximal isometric contractions, but extrapolating these findings to whole-body exercise and submaximal, dynamic contractions may be problematic. We isolated and compared core and skin temperature influences on an isometric force task versus a position task requiring dynamic maintenance of joint angle. Surface electromyography (sEMG) was measured on the flexor carpi radialis at 60% of baseline maximal voluntary contraction while either pushing against a rigid restraint (force task) or while maintaining a constant wrist angle and supporting an equivalent inertial load (position task). Twenty participants performed each task at 0.5°C rectal temperature (T_re) intervals while being passively heated from 37.1±0.3°C to ≥1.5°C T_re and then cooled to 37.8±0.3°C, permitting separate analyses of core versus skin temperature influences. During a 3-s contraction, trend analysis revealed a quadratic trend that peaked during hyperthermia for root-mean-square (RMS) amplitude during the force task. In contrast, RMS amplitude during the position task remained stable with passive heating, then rapidly increased with the initial decrease in skin temperature at the onset of passive cooling (p = 0.010). Combined hot core and hot skin elicited shifts toward higher frequencies in the sEMG signal during the force task (p = 0.003), whereas inconsistent changes in the frequency spectra occurred for the position task. Based on the patterns of RMS amplitude in response to thermal stress, we conclude that core temperature was the primary thermal afferent influencing neuromuscular response during a submaximal force task, with minimal input from skin temperature. However, skin temperature was the primary thermal afferent during a position task with minimal core temperature influence. Therefore, temperature has a task-dependent impact on neuromuscular responses.