. 2021 Nov 18;16(11):e0260166. doi: 10.1371/journal.pone.0260166

Table 9. A summary of evoked surface EMG data collection, recording and signal processing.

EMG recording and signal processing for studies carrying out neurophysiological assessments.

	PREPARATION/RECORDING		SIGNAL PROCESSING				RESULTS
Study	Muscles [Preparation Described]	Recording Device [Sampling Frequency]	Filter Passband [Stim artefact filtering]	Rectification	Cycle averaging	Amplitude Normalization	Output Presented
Dy et al., 2010 [14]	Sol, MG, TA, med hams, VL	Hard wired A/D board and customized labVIEW software	20-1000Hz for resting and standing	P2P amplitude for resting standing. Full wave rectified and peak for stepping.	12 MMR responses	Mean MMR for each muscle was normalized to sol responses as stim electrode placement was determined by optimization of sol response	Quantitative comparison between NI and SCI for MMR resting and standing, and phase-dependant MMR during stepping
	[✓]	[200Hz]	40-500Hz for stepping
			[✓]
Emeliannikov et al., 2016 [15]	RF, BF, TA, and LG	Viasys Viking Select	-	P2P amplitude	-	-	Comparison of MMR, H-Reflex and M-Wave at rest.
Emeliannikov et al., 2016 [15]	[✗]	[NS]	[✗]	P2P amplitude	-	-	Comparison of MMR, H-Reflex and M-Wave at rest.
Hofstoetter et al., 2018 [16]	RF, BF, TA and TS	DasyLab 11.0	1. 10-500Hz	P2P amplitude	-	-	Quantitative comparison of 1^st and 2^nd MMR amplitude for TSS and ESS. Onset offset and duration of 1^st and 2^nd MMR responses. Normalized response thresholds for TSS and ESS
	[✓]	Codas ADC system	2. 30-700Hz with add. 500Hz low-pass
		[2048 and 2002Hz]	[✓]
Hofstoetter et al., 2019 [17]	RF, BF, TA and Sol	Phoenix multichannel EMG system	10-1000Hz	P2P amplitude	10	Response amplitude of 2^nd stimulus in each pair was normalized to the respective 1^st for increasing inter-pulse interval (20-5000ms)	Quantitative comparison between NI and SCI for recovery of 2^nd PRM as inter-pulse interval increased.
Hofstoetter et al., 2019 [17]	[✓]	[2048Hz]	[✓]	P2P amplitude	10
Murray and Knikou, 2019 [20]	Sol, MG, PL, TA, med hams, lat ham, RF, and GRC	1401 Plus System	10-1000Hz	Full wave rectified AUC for each TEP response.	15	Responses at increasing intensities were normalized to the associated max response for recruitment curve.	Quantitative comparison of recruitment curve sigmoid parameters, PAD and HD, before and after 60min TSS.
	[✗]	[2000Hz]	[✗]			Responses at increasing frequency (0.1–1.0 Hz) normalized to response at 0.1Hz for HD.
	[✗]	[2000Hz]	[✗]			Response amplitude of 2^nd stimulus of a pair was normalized to the respective 1^st for PAD
Atkinson et al., 2020 [26]	RF, VL, med ham, TA, MG, Sol.	MA300 EMG System	-	P2P amplitude	10	Recruitment responses normalized to P2P amplitude at the maximum rate of recruitment (RRmax) within each muscle. CTI: 2^nd stimulus in each pair was normalized to the respective 1^st for increasing inter-pulse interval (40-160ms)	Quantitative comparison of interlimb conditioning between NI and SCI.
Atkinson et al., 2020 [26]	[✗]	[5000Hz]	[✗]	P2P amplitude	10
Militskova et al., 2020 [25]	RF, med ham, TA, sol	Neuro MEP- (Neurosoft, Ivanovo, Russia)	-	-	10	-	Quantitative comparison of SEP latency, threshold and amplitude across (A) 3 stim sites, (B) lying supine vs. standing and (C) pre- post-rehab
Militskova et al., 2020 [25]	[✗]	[5000Hz]	[✗]	-	10	-
Wu et al., 2020 [23]	APB, ADM, FCR, BB	Customized LabVIEW software (National Instruments USB-6363)	15-2000Hz	-	10	Response amplitude of 2^nd stimulus of a pair was normalized to the respective 1^st (PAD)	Quantitative comparison recruitment curves across stim configuration. Quantitative comparison of PAD across stim intensity between NI and SCI.
Wu et al., 2020 [23]	[✗]	[5000Hz]	[✗]	-	10
Islam et al., 2021 [72]	Sol, MG, TA, PL [✓]	Data acquisition card (National Instruments, Austin, TX, USA) [2000Hz]	20–500 Hz [✗]	Full wave rectification, linear enveloping via 20Hz low pass filter. Average RMS for entire step cycle.	-	Maximal stepping EMG without stimulation	Quantitative comparison between stim off, stim on (pulse train or single pulses) and post stim of RMS EMG amplitude, mean power frequency, inter/intra limb coordination and h-reflex modulation during stepping in individuals with and without SCI
Islam et al., 2021 [72]	Sol, MG, TA, PL [✓]		20–500 Hz [✗]	P2P amplitude for h-reflex	-	M-wave for h-reflex

Abbreviations: ADM; abductor digiti minimi, APB; abductor pollicis brevis, BB; biceps brachii, AUC; area under curve, BR; brachioradialis, CTI; conditioning-test interval, Delt; deltoid, DGO; ED; extensor digitorum, FCR; flexor carpi radialis, FD; flexor digitorum, GRC; gracilis, ham; hamstrings, HD; homosynaptic depression, lat ham; lateral hamstrings, LG; lateral gastrocnemius, med ham; medial hamstrings, MG; medial gastrocnemius, MMR; multisegmental monosynaptic response, P2P; peak-to-peak, PAD; post-activation depression, PL; peroneus longus, Q; quadriceps, RA; rectus abdominis, RF; rectus femoris, RMS; root mean square, Sol; soleus, SEP; spinally evoked potential, TA; Tibialis Anterior, TB; triceps brachii, TFL; tensor fascia lata, TP; tibialis posterior, TS; triceps surae/calf, VL; vastus lateralis.

*Preparation described refers to a clear description of preparation of the skin before surface electrode application, recording electrode type, orientation, shape and composition as well as interelectrode distance.

†Artifact filtering refers to an attempt made by the authors to account for and remove artifacts contaminating or obscuring the recorded EMG signals such as with the use of a filter.