Table 3.

Review of EEG-based motor intent detection studies in stroke patients.

	References	Studies	No. stroke subjects	Method	Accuracy mean (SD)%
Offline	Muralidharan et al., 2011b	Muralidharan et al., 2011	4	SMR	TPRmax = 70 FPR = 22 (9)
	Antelis et al., 2012	Antelis et al., 2012	4	SMR	71 (10)
	Lew et al., 2012	Lew et al., 2012	2a	MRCP	TPRmax = 79 (12) FPR = 10
	Niazi et al., 2013	Niazi et al., 2013	5a	MRCP	TPR = 60 (11) FPR/min = 4 (4)
	Ibáñez et al., 2014	Ibáñez et al., 2014	6a	SMR + MRCP	TPR = 82 (10) FPR/min = 1.5 (1)
	–	Current study [offline]	4	MRCP	TPRc = 82 (16) FPRc = 44 (18)
Online	Buch et al., 2008	Buch et al., 2008	8	SMR, [MEG]	73 (18), [median]
	Daly et al., 2008	Daly et al., 2008	3	SMR	82-98
	Ang et al., 2011	Ang et al., 2011	11	SMR	82 (−)b
	Gomez-Rodriguez et al., 2011	Rodriguez et al., 2011	2a	SMR	84 [AUC]
	–	Current study [online]	4	MRCP	TPR = 65 (18) FPR = 28 (36)

SMR, Sensorimotor or μ-rhythms; MRCP, Movement related cortical potentials; AUC, performance reported as area under the ROC curve.

^aStudy included both healthy and stroke participants. Here we mention only the number of stroke participants. If available, we report only the BMI accuracy that was obtained with stroke patients.

^bStudy reported both online and offline accuracies. Here we only consider online accuracy.

^cOverall TPR and FPR computed offline during BMI Calibration. Note that EMG gating was not used when computing offline accuracy.