A wearable electromyography-controlled functional electrical stimulation system improves balance,gait function,and symmetry in older adults

Abstract

BACKGROUND:

Wearable technologies have been developed for healthy aging. The technology for electromyography (EMG)-controlled functional electrical stimulation (FES) systems has been developed, but research on how helpful it is in daily life has been insufficient.

OBJECTIVE:

The purpose of this study was to investigate the effect of the EMG-controlled FES system on muscle morphology, balance, and gait in older adults.

METHODS:

Twenty-nine older adults were evaluated under two randomly assigned conditions (non-FES and FES assists). Muscle morphology, balance, gait function, and muscle effort during gait were measured using ultrasonography, a physical test, a gait analysis system, and EMG.

RESULTS:

The EMG-controlled FES system improved gait speed by 11.1% and cadence by 15.6% ( $P<$ 0.01). The symmetry ratio of the bilateral gastrocnemius was improved by 9.9% in the stance phase and 11.8% in the swing phase ( $P<$ 0.05). The degrees of coactivation of the knee and ankle muscles were reduced by 45.1% and 50.5%, respectively ( $P<$ 0.05). Balance improved by 6–10.7% ( $P<$ 0.01).

CONCLUSION:

The EMG-controlled FES system is useful for balance and gait function by increasing muscle symmetry and decreasing muscle coactivation during walking in older adults.

Keywords

Aging wearable technology muscle symmetry gait function

1. Introduction

As the population of older adults grows, seniors are increasingly interested in pursuing higher physical functions and improving their quality of life, that is, healthy aging, rather than simply extending their lives [1]. Healthy aging is defined as the process by which older adults improve and maintain their functional ability for well-being [2]. Therefore, older adults have complex care demands for healthy aging through independent life [3]. To meet these demands, wearable technology has been developed for simply detecting an emergency situation and preventing accidents, and recently, for the purpose of helping movement using real-time feedback [4, 5, 6]. However, most wearable devices made to help human movement have used mechanical support such as exoskeleton and orthoses type, which required an electric power supply [7, 8]. This type of device has the disadvantages of requiring customization for weight, battery capacity, noise, and cost [9]. Therefore, to overcome the shortcomings of exoskeleton-like equipment, a wearable technology using functional electrical stimulation (FES) has been developed [9, 10].

FES has been used to recover motor function in neurological conditions such as stroke or spinal cord injury, and has been widely used in the field of rehabilitation to improve the decreased muscle strength and performance of the elderly [11, 12]. The repetitive and continuous pattern of FES can lead to changes in the spinal cord synapse between the upper motor neurons and $\alpha$ motor neurons by the principle of neuroplasticity [13, 14]. In general, in clinical practice, FES has been applied passively with an initially set intensity or as a triggered FES, in which the power is turned on/off according to the movement of the participants [15, 16]. However, in both types of FES devices, the effect of neuroplasticity may be reduced because the participants’ spontaneous movements and the timing of stimulation of the FES are not consistent [14]. Therefore, much research has been conducted on the technology that can apply FES at the appropriate moment by identifying participants’ intention to move [10, 17]. Electromyography (EMG) sensors and inertial measurement units are the representative equipment used to determine the intention of movement, but research has shown that better muscle contraction control is achieved in an FES device using EMG [17]. In addition, a system that controlled the intensity of the electrical stimulation by identifying the intention of movement through a brain-computer interface controlled-FES was developed, but compared with EMG-FES, it showed no remarkable difference in the change in motor-evoked potentials and muscle force [14]. Therefore, in terms of device complexity and cost, it is desirable to select a method for controlling FES by identifying the movement intention using EMG. However, electrical signals from the FES can interfere with the EMG signals from muscles [14, 18]. With the advancement of technology, research has been conducted to use FES intensity control to detect the voluntary EMG signal of pure muscle origin, despite the interfering signal of FES [19, 20]. Clinical studies have been conducted using these technologies, but most of them control the movement of the upper limbs such as the forearm or wrist, or performed evaluations only in simple tasks [10, 20]. Finally, there is insufficient research to investigate changes in muscle parameters in older adults who require assistance in movement or to demonstrate their effect on daily functional abilities such as balance and walking.

Therefore, the purpose of this study was to investigate the immediate effects of wearable EMG-controlled FES on the lower limb muscle morphology, balance, and gait in older adults.

2. Methods

2.1 Participants

Twenty-nine individuals (10 men and 19 women), aged 65–84 years, with no history of central nervous system disease or abnormality were recruited for the study. The sample size was calculated using the G-Power program (IBM Inc., USA). On the basis of previous studies, when the effect size was set to 0.56, the power was set to 0.8, and the alpha error was set to 0.05, the minimum sample size was 28 people [16]. The exclusion criteria were as follows: (1) difficulty walking independently owing to problems such as visual field defect, fracture, severe muscle paralysis; (2) poorly controlled hypertension and diabetes; (3) severe dizziness; (4) serious cognitive problems; and (5) other reasons deemed inappropriate by researcher’s judgement. The study was approved by the institutional review board of the University of Sahmyook (2-7001793-AB-N-0120180861HR). All participants provided written informed consent.

2.2 Procedures

This study was designed as a crossover study to measure the test conditions. The measurement conditions were EMG-controlled FES with and without assist. All the participants went through the assessment for muscle morphology measurement, balance test, and gait with FES assist (FA) and non-FES assist (NFA) in random order. The participants had sufficient rest between the two tests conditions. EMG-controlled FES electrodes were attached bilaterally to the following lower extremity muscles: the rectus femoris (RF), biceps femoris (BF), tibialis anterior (TA), and medial part of gastrocnemius (GCM). To attach the FES electrodes, we used STIMPLUS (DP-200, CyberMedic, Iksan, Korea) to find the optimal motor point for each muscle. By using a pen-electrode, the skin of the target muscle was pressed gently for around 5 s to find the area where the twitching reaction appeared; the frequency of stimulation was set to 2 Hz using a monophasic wave with a pulse width of 100 $\mu$ s [21]. With the FES electrodes attached, the EMG electrodes were attached to the center of the belly of the target muscle in accordance with the SENIAM guidelines, without overlapping the FES electrodes. Before attaching the electrode, the skin was shaved and abraded with isopropyl alcohol to minimize skin impedance and signal noise [22]. After electrode attachment, maximal voluntary isometric contraction (MVC) measurements were performed for EMG-controlled FES activation. For MVC measurement of RF and BF, the participants sat in an upright position on a chair with a backrest, with a towel rolled under the thigh, and waited in a leg dangling state; manual resistance was provided above the malleolus [23]. The MVC measurement of TA was made at the maximum contraction with manual resistance provided to the metatarsals in the sitting position. For the GCM measurement, the participants were asked to make the heel rise from the ground as strongly as possible in the standing position [23, 24]. The EMG signal of each muscle was measured 3–5 times for 5 s in MVC with sufficient rest between measurements. Subsequently, on the basis of on the measured MVC, to check whether the EMG-controlled FES provided adequate stimulation, the participants walked for 5–10 min. At the end of all preparations, we measured the subjects’ muscle architecture, balance, gait function, and muscle effort during gait in two measurement conditions, with and without the EMG-controlled FES equipment in operation.

Figure 1.

EMG and FES electronode attachment locations (A), EMG-controlled FES system (B), and diagram of EMG-controlled FES system (C). vEMG indicates volitional electromyography.

2.3 EMG-Controlled FES

The EMG-controlled FES system (Electronics and Telecommunications Research Institute, ETRI, Deajeon, Republic of Korea) was designed as a gait assistance strategy for older adults (Fig. 1B) and is controlled using a MATLAB-based custom graphical user interface program installed in a personal computer. It is a biofeedback device that extracts volitional EMG (vEMG) signals through dual-channel EMG signal processing algorithms and adjusts the FES intensity based on the signal of pure muscle origin regardless of FES signal interference (Fig. 1C) [18]. The system consists of one FES channel and two EMG channels for each muscle and is attached to a total of eight muscles. The signal received by each EMG channel is expressed as follows:

$\displaystyle\textit{EMG}_{j,1}(i)=V_{\textit{FES},j,1}(i)+V_{V,j,1}(i)$ $\displaystyle\textit{EMG}_{j,2}(i)=V_{\textit{FES},j,2}(i)+V_{V,j,2}(i)$

$\textit{EMG}_{j,1}(i)$ is the EMG signal received by channel 1, $V_{\textit{FES},j,1}(i)$ is the signal generated from FES and received by EMG channel 1, and $V_{V,j,1}(i)$ is the vEMG signal; $j$ is the stimulation index; and $i$ is the sampling index. The symbols have the same meaning in the equation for $\textit{EMG}_{j,2}(i)$ . The vEMG algorithm, which detects stimulus artifacts and M waves generated by FES from two EMG channels, removes generated signals of the same amplitudes, and detects synchronous physiological activity from the brain, is defined as follows [25]:

$\displaystyle V_{V,j}(i)=(\textit{EMG}_{j,1}(i)-\textit{EMG}_{j-1,1}(i))-(% \textit{EMG}_{j,2}(i)-\textit{EMG}_{j-1,2}(i))$

EMG signals were recorded using custom-developed hardware from the ECG/EEG analog front-end system (ADS-1299, Texas Instruments, Texas, USA). The common mode rejection rate was 110 dB, and the resolution of the signal was 24 bits. The sampling data were collected at 1000 Hz. Raw data were band-pass (10–200 Hz) filtered, and a 300-ms root mean square sliding window function was used for smoothing.

The vEMG raw data measured by the algorithm were automatically filtered and displayed as %MVC and delivered to the connected FES unit to adjust the stimulation pulse in proportion to %MVC in real time. FES stimulation was provided from the point when %MVC exceeded 0, and as %MVC increased, it approached the maximum intensity. The maximum intensity of FES was set to a maximum tolerable level within 10–40 mA of the subject’s painless range. The frequency of FES was fixed at 40 Hz and at a pulse width of 250 $\mu$ s, and the current was adjusted in real time.

The FES unit consists of Rehastim I (Hasomed GmbH, Magdeburg, Germany) and has eight stimulation channels. Rehastim I is a biphasic rectangle pulse with a frequency range of 1–140 Hz, a pulse width range of 20–500 $\mu$ s, and a current output range of 0–130 mA [26].

2.4 Measurements

2.4.1 Muscle effort

Muscle effort was defined as the mean normalized EMG activity of the lower extremities during the stance and swing phases during gait [8]. It was measured with an EMG-controlled FES system during 5-m walking by dividing the stance and swing phases. The measurements were averaged three times each in accordance with the EMG-controlled FES condition. The stance phase and swing phase events were specified using foot switches located on both heels and the first metatarsal heads. A foot switch is a simple mechanical sensor that can only be turned on/off and detects from heel contact to toe-off.

2.4.2 Muscle morphology

The quadriceps muscle thickness (QMT) is the sum of the thicknesses of the rectus femoris and vastus intermedius [27]. The dual-probe personal-computer-based muscle viewer (DPC-BMW) was used to measure bilateral QMTs simultaneously. The DPC-BMW (MicrUS-duo, TELEMED, Vilnius, Lithuania) linear probe frequency range is 6–14 MHz, with a depth of 4–10 cm, and was preset for optimal muscle image measurement in B-mode [28, 29]. The probe was positioned perpendicular to the longitudinal quadriceps femoris axis midway between the anterior superior iliac spine and the proximal end of the patella [27, 30].

All measurements were performed at rest and maximal voluntary contraction (MVC), and three measurements were taken to calculate the average value. The ultrasonographic measurement protocol for rest and MVC was used based on previous studies [31]. To measure the resting image, the participant rested with the knee slightly flexed with a towel under the knee in a long sitting position on the bed. For MVC images, the isometric contraction was achieved by pressing as much as possible against the towel under the knee. To improve acoustic coupling and image quality, a water-soluble transmission gel was placed on the probe, and the probe’s angle was adjusted slightly [27, 30]. To reduce bias, all measurements were performed by a trained operator [32]. Muscle thickness was calculated using Echo Wave II software, ver. 3.5 (version). DPC-BMW had very good to excellent reliability (ICC $=$ 0.916–0.994) and high correlation ( $r=$ 0.855–0.996) when compared with the Medison Mysono system (U5, Samsung Medison, Seoul, Korea), which is commonly used to measure muscle thickness [27]. The minimal detectable change (MDC) for QMT in the older participants was 5.9 mm [33].

2.4.3 Balance

A short physical performance battery (SPPB) was used to measure the physical function and strength of older adults [34]. The SPPB included the following three tasks: standing balance, 4-m walking, and sit-to-stand. For the standing balance test, the participants held their feet side-by-side for 10 s. We evaluated whether this position could be maintained at the semi-tandem and tandem positions. The 4-m walking test started in the standing position, and we measured the time it took the participants to walk 4 m at a normal pace. The sit-to-stand test measured the time from sitting on a chair to standing five times. Each of the three tasks was rated as 0 (unable to complete) to 4. The total scores ranged from 0 to 12, with higher scores indicating better function and leg strength [35]. The SPPB indicated excellent test-retest reliability (ICC $=$ 0.91), and the MDC total score was 0.8 [36].

The timed up-and-go (TUG) test was used to measure the participants’ functional mobility and dynamic balance [37]. The TUG test has good inter-rater reliability (ICC $=$ 0.99) and excellent correlation ( $r=-$ 0.81) with the Berg balance test [38]. The participants were asked to rest on a chair. At the operator’s “start” instruction, the participants got up from their chair, walked 3 m in a straight line at a normal walking pace, turned 180 ${}^{\circ}$ , and walked back to their chairs. This test was performed once, and the time was measured [35]. According to the test results, if a test was performed in $<$ 10 s, between 20 and 29 s, and $>$ 30 s, the participant was categorized as having a conditioned input, having limited functional capacity, and “totally dependent” in activities of daily living [38]. The MDC value for the TUG was 4 s [39].

The four-square step test (FSST) is an effective tool for quickly measuring dynamic standing balance and mobility, and has moderate to strong correlations ( $r=$ 0.595–0.88) and good to excellent test-retest reliability (ICC $=$ 0.73–0.98) with other commonly used balance and mobility measures [40]. The FSST required four canes and a stopwatch [41]. The four canes were placed crosswise and divided into four sections. Participants stood in the first square, facing the second square, and moved clockwise to the next square on to the operator’s verbal command. On reaching the first square, they returned to the first square as quickly as possible in the counterclockwise direction. The FSST score was calculated from the measured time, and if it took longer than 15 s, it was determined that there was a risk of falling. The FSST cutoff score that determines faller and non-faller for patients with Parkinson’s disease is 9.68 s, but the value for MDC is not available [42].

2.4.4 Gait function

The GAITRite system (CIR System Inc., New Jersey, USA) was used to measure the gait function. GAITRite measures temporal and spatial gait parameters on an electronic walkway that is 5 m long, 0.6 m wide, and 0.6 cm high, with 16,128 pressure activated sensors 1 cm in diameter arranged vertically every 1.27 cm. This instrument has an excellent agreement (ICC $=$ 0.92–0.99; repeatability coefficients $=$ 1.0–5.9%) [43]. We measured gait speed, cadence, step length, stride length, and step width. The MDC for gait speed was 0.11 m/s [44].

The measurement protocol was modified based on previous studies [45]. Older adults walked three times each with the EMG-controlled FES system on and off (six trials in total). Participants walked at a self-selected speed according to the examiner’ s verbal instructions. Each walk was approximately 9 m long: 5 m GAITRite mat and 2 m acceleration and deceleration periods.

2.5 Data analysis

2.5.1 EMG data calculation

To compare individual muscle activation patterns during gait, EMG data were normalized to the percentage of MVC (%MVC). As a result, %MVC means muscle effort. To determine the symmetry of bilateral leg muscles in the stance and swing phases during gait, %MVC data were calculated as weakest/strongest, which was defined as the symmetry ratio (SR) [46]. As SR approaches 1, both muscles are symmetrical. Finally, the co-activation index (CI) was calculated to compare the degree of co-activation between agonist and antagonist (RF: BF, TA: GCM) during the gait process [24, 47]:

$\displaystyle\text{CI}({\%})=2\left({\frac{\smallint\min({\textit{sEMG}_{% \textit{agonist}},\textit{sEMG}_{\textit{antagonist}}})}{\smallint\textit{sEMG% }_{\textit{agonist}}+\smallint\textit{sEMG}_{\textit{antagonist}}}}\right)% \times 100$

CI was calculated as the average of 20 strides in each condition.

2.5.2 Ultrasound-image-based calculation

QMT was measured as the distance between the superior border of the subcutaneous fascia and the superior border of the femur [48]. The SR equation was used to compare the symmetry of both QMTs and was defined as the quadriceps symmetry ratio (QSR) [46].

2.6 Statistical analysis

PASW Statistics 18 (SPSS Inc., Quarry Bay, Hong Kong) was used for all statistical analyses. The Shapiro-Wilk test was performed to confirm the normality of the subject characteristics. All data are expressed as the mean (SD). A paired $t$ -test was used to compare outcomes with and without EMG-controlled FES. The statistical significance threshold was set at $P<$ 0.05.

3. Results

3.1 Baseline participant characteristics

The baseline characteristics of the 29 older adults are summarized in Table 1.

Table 1
Participant characteristics ${}^{\text{a}}$

	Older adults ( $n=$ 29)
Sex, female/male	19/10
Age, years	75.00 (5.36)
Height, cm	156.16 (8.64)
Weight, kg	62.35 (8.55)
BMI, kg/m ${}^{2}$	25.47 (2.92)

Abbreviations: BMI, Body mass index. ${}^{\text{a}}$ Values are expressed as mean (SD).

3.2 Muscle effort

FA gait significantly lowered muscle effort in the right RF and both GCM in the stance phase compared to the NFA gait ( $P<$ 0.05). In the swing phase, the FA gait showed significantly lower muscle effort in the left GCM than in the NFA gait ( $P<$ 0.05; Fig. 2). There was a significant difference in the SR of bilateral GCM muscles in both stance and swing phases in FA gait ( $P<$ 0.05). In addition, the agonist-antagonist CI of the RF, BF, GCM, and TA muscles decreased significantly during FA gait ( $P<$ 0.05; Table 2).

Table 2
Muscle effort, muscle thickness, physical function and gait function ${}^{\text{a}}$

	NFA		FA
Muscle effort
Symmetry ratio of stance phase (GCM)	1.11	(0.19)	1.01	(0.08) ${}^{*}$
Symmetry ratio of swing phase (GCM)	0.85	(0.12)	0.95	(0.07) ${}^{*}$
Co-activation index (RF:BF)	53.05	(10.68)	36.57	(8.04) ${}^{*}$
Co-activation index (TA:GCM)	52.18	(18.57)	34.66	(5.48) ${}^{*}$
QMT
Right muscle thickness, mm	18.20	(3.21)	17.79	(2.49)
Left muscle thickness, mm	16.80	(2.57)	16.84	(2.52)
QSR	0.92	(0.06)	0.97	(0.07) ${}^{\text{c}}$
Physical function
SPPB, points	10.52	(1.67)	11.15	(1.63) ${}^{**}$
TUG, s	10.65	(3.33)	10.03	(2.84)
FSST, s	10.23	(1.74)	9.24	(1.66) ${}^{**}$
Gait function
Gait speed, m/s	0.98	(0.36)	1.10	(0.32) ${}^{**}$
Cadence, steps/min	103.56	(21.84)	119.76	(26.75) ${}^{**}$
Step length, cm	38.26	(11.42)	40.11	(9.35)
Stride length, cm	78.41	(15.74)	82.35	(20.34)
Step width, cm	17.53	(8.24)	16.34	(6.29)

Abbreviations: NFA, non-FES assist; FA, FES assist; RF, rectus femoris; BF, biceps femoris; TA, tibialis anterior; GCM, gastrocnemius; QMT, quadriceps muscle thickness; QSR, quadriceps symmetry ratio; SPPB, short physical performance battery; TUG, time up and go test; FSST, four square step test. ${}^{\text{a}}$ Values are expressed as mean (SD). ${}^{*}P<$ 0.05 significant difference comparing the NFA. ${}^{**}P<$ 0.01 significant difference comparing the NFA.

Figure 2.

Mean (standard deviation) muscle effort (%MVC) of swing phase (A) and stance phase (B), according to each muscle. ${}^{*}$ Significantly different in relation to the NFA ( $P<$ 0.05). NFA indicates non-FES assist gait and FA indicates FES assist gait.

3.3 Muscle morphology

Left and right QMT in FA or NFA gaits were not statistically significant ( $P>$ 0.05). However, the difference in QSR was statistically significant ( $P=$ 0.004; Table 2).

3.4 Balance

The SPPB showed a statistically significant improvement in the FA condition compared to the NFA condition ( $P<$ 0.01). There was also a statistically significant difference in FSST between the two measurement conditions ( $P<$ 0.01). However, there was no statistically significant difference in the TUG test results ( $P=$ 0.072; Table 2).

3.5 Gait function

Gait speed was significantly faster in the FA condition than in the NFA condition ( $P<$ 0.01). There was also a significant difference in cadence ( $P<$ 0.01). However, the step length did not show a statistically significant difference between the two measurement conditions ( $P=$ 0.16). Similarly, there was no statistically significant difference in stride length ( $P=$ 0.08). There was no statistically significant difference in step width ( $P=$ 0.15; Table 2).

4. Discussion

The purpose of our study was to investigate the effects of EMG-controlled FES on muscle effort, muscle morphology, balance, and gait in older adults.

The results showed that the application of EMG-controlled FES reduced muscle effort during gait in older adults. In particular, bilateral GCM muscle effort was significantly reduced in the stance phase (right 18.18%, left 10.17%). There was no significant difference in the other muscles assessed, but most of them showed a decreasing pattern during the FA gait. This is similar to previous research findings that it is possible to move efficiently by using less energy during trunk forward progression at terminal stance by reducing the muscle consumption rate of GCM through a gait assistance device [8]. Another study found that when performing functional movements such as arm movement, the effort of the muscles used increased in older adults compared to that in young adults [49]. Therefore, in order to use muscles efficiently for gait performance in older adults, muscle effort of the GCM should be set as an important factor. The SR of bilateral GCM muscle effort was significantly improved during FA gait (stance phase, 9%; swing phase, 10.53%). These results are similar to previous studies that single session gait training using FES improved ankle moment symmetry [50]. The degree of co-activation increases with the decrease of strength and neurological function with aging, which means that older adults make joints stiff for stability compared to young adults even if they perform the same task [47]. We confirmed a significant decrease in co-activation in both the thigh (31.06%) and shank (33.58%) during FA gait. Consequently, our results showed that the FA gait was more efficient in muscle effort, and the stiffness of the lower limb joints was reduced, resulting in smooth motion.

In this study, there was no difference in muscle thickness during the MVC of the quadriceps muscle, FA or NFA. The reason for this result is similar to that of a previous study, in which single sessions tended not to elicit morphologic changes such as muscle mass alterations in older adults [51]. However, QSR showed a significant improvement in the FA condition (5.15%). According to previous studies, the increase in knee extensor muscle asymmetry is related to gait asymmetry, and unilateral movement is important during training [52]. Similarly, in this study, the use of EMG-controlled FES led to symmetrical contraction of the quadriceps muscles in the elderly and to a symmetrical pattern of muscle activation during gait. Therefore, the EMG-controlled FES system assists in symmetrical muscle activity.

EMG-controlled FES affected the physical function of older adults. After wearing the EMG-controlled FES equipment, scores improved in the SPPB (5.65%) and FSST (9.68%), indicating that EMG-controlled FES was effective in improving physical performance and balance. This was consistent with the findings of previous studies that reported better gait and balance performance, as well as improved muscle physiology and function in older adults with FES-based training [11].

The gait parameter is an objective evaluation method and is used as an important biomarker in aging and pathological studies [3]. Our results show that FA gait has a positive effect on gait speed (10.9%) and cadence (13.55%) in the elderly. In previous studies, it was reported that when FES was applied, for voluntary muscle contraction, the gait performance of older adults improved higher than when only passive FES was applied [11]. Therefore, it is necessary to compare the difference in the level of balance or gait parameters after training in which the FES intensity is adjusted in real time through EMG-controlled FES and the existing voluntary muscle contraction FES training, in a follow-up study. In another previous study, it was reported that gait speed decreased with increasing co-activation of the agonist and antagonist muscles in the stance and swing phases [24]. Similarly, in our study, gait speed increased and co-activation of the lower limb muscles decreased according to FA in older adults.

Finally, our results demonstrate that the EMG-controlled FES system is a potentially useful gait assist system for improving gait function by increasing the muscle symmetry of the lower limbs and by decreasing muscle co-activation during walking in elderly adults.

The limitation of this study is the small sample size used to support the cross-sectional study design. In addition, it was difficult to accurately determine the effect of EMG-controlled FES because it was aimed at healthy elderly people without gait problems. Therefore, in future research, it will be necessary to proceed with a larger sample size and various groups.

5. Conclusion

The EMG-controlled FES not only improved balance, walking speed, and muscle morphological ability, but also positively altered the asymmetrical walking pattern in older adults. The proper timing of electrical stimulation also had an effect on the reduction of joint muscle contraction.

This study verified the supplementary effects of electrical stimulation in older adults, and future studies are needed to verify the effects in the long-term, in various groups, and training methods using EMG-controlled FES equipment.

Footnotes

Acknowledgments

This study was supported by the Institute for Information and Communications Technology Promotion (IITP) grant funded by the Korean government (MSIT) (2017-0-00050, Development of Human Enhancement Technology for auditory and muscle support) and by a grant from the NRF (NRF-2016R1A6A3A11930931 and NRF-2018R1D1A1B07042870), which is funded by the Korean government.

Conflict of interest

The authors declare no conflicts of interest.

References

Cosco

Howse

Brayne

. Healthy ageing, resilience and wellbeing. Epidemiol Psychiatr Sci. 2017; 26(6): 579-83.

World Health Organization. Global strategy and action plan on ageing and health. 2017.

Godfrey

. Wearables for independent living in older adults: gait and falls. Maturitas. 2017; 100: 16-26.

Radder

Prange-Lasonder

Kottink

Holmberg

Sletta

Van Dijk

, et al. The effect of a wearable soft-robotic glove on motor function and functional performance of older adults. Assist Technol. 2020; 32(1): 9-15.

Kekade

Hseieh

C-H

Islam

Atique

Khalfan

Y-C

, et al. The usefulness and actual use of wearable devices among the elderly population. Comput Methods Programs Biomed. 2018; 153: 137-59.

Del Din

Lewis

Gray

Collin

Kissima

Rochester

, et al. Monitoring walking activity with wearable technology in rural-dwelling older adults in Tanzania: a feasibility study nested within a frailty prevalence study. Exp Aging Res. 2020; 46(5): 367-81.

Moltedo

Baček

Verstraten

Rodriguez-Guerrero

Vanderborght

Lefeber

. Powered ankle-foot orthoses: the effects of the assistance on healthy and impaired users while walking. J Neuroeng Rehabil. 2018; 15(1): 86.

Lee

Chang

Seo

Shim

Choi

, et al. A Wearable hip assist robot can improve gait function and cardiopulmonary metabolic efficiency in elderly adults. IEEE Trans Neural Syst Rehabil Eng. 2017; 25(9): 1549-57.

Taghavi

Luecke

Jeffery

. A wearable body controlling device for application of functional electrical stimulation. Sensors (Basel). 2018; 18(4): 1251.

10.

Osuagwu

Whicher

Shirley

. Active proportional electromyogram controlled functional electrical stimulation system. Sci Rep. 2020; 10(1): 21242.

11.

Langeard

Bigot

Chastan

Gauthier

. Does neuromuscular electrical stimulation training of the lower limb have functional effects on the elderly: a systematic review. Exp Gerontol. 2017; 91: 88-98.

12.

Kapadia

Moineau

Popovic

. Functional electrical stimulation therapy for retraining reaching and grasping after spinal cord injury and stroke. Front Neurosci. 2020; 14: 718.

13.

Morris

. D.O. Hebb: The Organization of Behavior, Wiley: New York; 1949. Brain Res Bull. 1999; 50(5-6): 437.

14.

McGie

Zariffa

Popovic

Nagai

. Short-term neuroplastic effects of brain-controlled and muscle-controlled electrical stimulation. Neuromodulation. 2015; 18(3): 233-40.

15.

Pomeroy

King

Pollock

Baily-Hallam

Langhorne

. Electrostimulation for promoting recovery of movement or functional ability after stroke. Cochrane Database Syst Rev. 2006(2).

16.

Howlett

Lannin

Ada

McKinstry

. Functional electrical stimulation improves activity after stroke: a systematic review with meta-analysis. Arch Phys Med Rehabil. 2015; 96(5): 934-43.

17.

Tigra

Lucas

Benjamin

David

Antonio

Emerson

, et al. Towards FES-assisted grasping controlled by residual muscle contraction and movement on persons with tetraplegia. Ann Phys Rehabil Med. 2018; 61: e485-e6.

18.

Yeom

Chang

. Autogenic EMG-controlled functional electrical stimulation for ankle dorsiflexion control. J Neurosci Methods. 2010; 193(1): 118-25.

19.

Shalaby

RE-S

. Development of an electromyography detection system for the control of functional electrical stimulation in neurological rehabilitation. 2011. doi: 10.14279/depositonce-2904.

20.

Salchow

Dorn

Valtin

Schauer

. Intention recognition for fes in a grasp-and-release task using volitional emg and inertial sensors. Curr Dir Biomed Eng. 2017; 3(2): 161-5.

21.

Gobbo

Maffiuletti

Orizio

Minetto

. Muscle motor point identification is essential for optimizing neuromuscular electrical stimulation use. J Neuroeng Rehabil. 2014; 11: 17.

22.

Hermens

Freriks

Disselhorst-Klug

Rau

. Development of recommendations for SEMG sensors and sensor placement procedures. J Electromyogr Kinesiol. 2000; 10(5): 361-74.

23.

Colombo

Mazzini

Mora

Parenzan

Creola

Pirali

, et al. Measurement of isometric muscle strength: a reproducibility study of maximal voluntary contraction in normal subjects and amyotrophic lateral sclerosis patients. Med Eng Phys. 2000; 22(3): 167-74.

24.

Lee

Chang

Choi

Ryu

Kim

. Age-related differences in muscle co-activation during locomotion and their relationship with gait speed: a pilot study. BMC Geriatr. 2017; 17(1): 44.

25.

Lynch

Popovic

. Functional electrical stimulation. IEEE Control Syst. 2008; 28(2): 40-50.

26.

Lew

Alavi

Randhawa

Menon

. An Exploratory investigation on the use of closed-loop electrical stimulation to assist individuals with stroke to perform fine movements with their hemiparetic arm. Front Bioeng Biotechnol. 2016; 4: 20.

27.

Ikezoe

Asakawa

Fukumoto

Tsukagoshi

Ichihashi

. Associations of muscle stiffness and thickness with muscle strength and muscle power in elderly women. Geriatr Gerontol Int. 2012; 12(1): 86-92.

28.

Guerreiro

Tonelli

Orzechowski

Dalla Corte

Moriguchi

de Mello

. Bedside ultrasound of quadriceps to predict rehospitalization and functional decline in hospitalized elders. Front Med (Lausanne). 2017; 4: 122.

29.

Cho

J-E

Cho

Yoo

Lee

W-H

. Reliability and validity of a dual-probe personal computer-based muscle viewer for measuring the pennation angle of the medial gastrocnemius muscle in patients who have had a stroke. Top Stroke Rehabil. 2018; 25(1): 6-12.

30.

Nishihara

Kawai

Kera

Hirano

Watanabe

Fujiwara

, et al. Correlation of physical function with the thickness of multiple muscles of the quadriceps femoris in community-dwelling elderly individuals. Clin Interv Aging. 2018; 13: 1945-51.

31.

Kellis

. Antagonist muscle architecture and aponeurosis/tendon strain of biceps femoris long head during maximal isometric efforts. Eur J Appl Physiol. 2019; 119(1): 73-83.

32.

Nozoe

Kanai

Kubo

Kitamura

Shimada

Mase

. Changes in quadriceps muscle thickness in acute non-ambulatory stroke survivors. Top Stroke Rehabil. 2016; 23(1): 8-14.

33.

Agyapong-Badu

Warner

Samuel

Narici

Cooper

Stokes

. Anterior thigh composition measured using ultrasound imaging to quantify relative thickness of muscle and non-contractile tissue: a potential biomarker for musculoskeletal health. Physiol Meas. 2014; 35(10): 2165-76.

34.

Guralnik

Simonsick

Ferrucci

Glynn

Berkman

Blazer

, et al. A short physical performance battery assessing lower extremity function: association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol. 1994; 49(2): M85-94.

35.

Vikberg

Sorlen

Branden

Johansson

Nordstrom

Hult

, et al. Effects of resistance training on functional strength and muscle mass in 70-year-old individuals with pre-sarcopenia: a randomized controlled trial. J Am Med Dir Assoc. 2019; 20(1): 28-34.

36.

Olsen

Bergland

. “Reliability of the Norwegian version of the short physical performance battery in older people with and without dementia”. BMC Geriatr. 2017; 17(1): 124.

37.

Brandao

Oliveira

LVF

Brandao

Silva

Sampaio

AAC

Urbano

, et al. Effect of a home-based exercise program on functional mobility and quality of life in elderly people: protocol of a single-blind, randomized controlled trial. Trials. 2018; 19(1): 684.

38.

Podsiadlo

Richardson

. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991; 39(2): 142-8.

39.

Mangione

Craik

McCormick

Blevins

White

Sullivan-Marx

, et al. Detectable changes in physical performance measures in elderly African Americans. Phys Ther. 2010; 90(6): 921-7.

40.

Moore

Barker

. The validity and reliability of the four square step test in different adult populations: a systematic review. Syst Rev. 2017; 6(1): 187.

41.

Dite

Temple

. A clinical test of stepping and change of direction to identify multiple falling older adults. Arch Phys Med Rehabil. 2002; 83(11): 1566-71.

42.

Duncan

Earhart

. Four square step test performance in people with parkinson disease. J Neurol Phys Ther. 2013; 37(1): 2-8.

43.

Webster

Wittwer

Feller

. Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait. Gait Posture. 2005; 22(4): 317-21.

44.

Goldberg

Schepens

. Measurement error and minimum detectable change in 4-meter gait speed in older adults. Aging Clin Exp Res. 2011; 23(5): 406-12.

45.

Brown

Hanson

Firmani

Liu

McAllister

Merali

, et al. Gait speed and variability for usual pace and pedestrian crossing conditions in older adults using the GAITRite walkway. Gerontol Geriatr Med. 2015; 1: 2333721415618858.

46.

Skelton

Kennedy

Rutherford

. Explosive power and asymmetry in leg muscle function in frequent fallers and non-fallers aged over 65. Age Ageing. 2002; 31(2): 119-25.

47.

Peterson

Martin

. Effects of age and walking speed on coactivation and cost of walking in healthy adults. Gait Posture. 2010; 31(3): 355-9.

48.

Ando

Saito

Umemura

Akima

. Local architecture of the vastus intermedius is a better predictor of knee extension force than that of the other quadriceps femoris muscle heads. Clin Physiol Funct Imaging. 2015; 35(5): 376-82.

49.

Poirier

Papaxanthis

Mourey

Gaveau

. Motor planning of vertical arm movements in healthy older adults: does effort minimization persist with aging? Front Aging Neurosci. 2020; 12: 37.

50.

Palmer

Hsiao

Wright

Binder-Macleod

. Single session of functional electrical stimulation-assisted walking produces corticomotor symmetry changes related to changes in poststroke walking mechanics. Phys Ther. 2017; 97(5): 550-60.

51.

Dirks

Wall

Kramer

Zorenc

Goessens

Gijsen

, et al. A single session of neuromuscular electrical stimulation does not augment postprandial muscle protein accretion. Am J Physiol Endocrinol Metab. 2016; 311(1): E278-E85.

52.

Laroche

Cook

Mackala

. Strength asymmetry increases gait asymmetry and variability in older women. Med Sci Sports Exerc. 2012; 44(11): 2172-81.

A wearable electromyography-controlled functional electrical stimulation system improves balance,gait function,and symmetry in older adults

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Methods

2.1 Participants

2.2 Procedures

2.4 Measurements

2.4.1 Muscle effort

2.4.2 Muscle morphology

2.4.3 Balance

2.4.4 Gait function

2.5 Data analysis

2.5.1 EMG data calculation

2.5.2 Ultrasound-image-based calculation

2.6 Statistical analysis

3. Results

3.1 Baseline participant characteristics

Table 1 Participant characteristics a

Table 2 Muscle effort, muscle thickness, physical function and gait function a

3.4 Balance

3.5 Gait function

4. Discussion

5. Conclusion

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Participant characteristics ${}^{\text{a}}$

Table 2
Muscle effort, muscle thickness, physical function and gait function ${}^{\text{a}}$