Objective: All-night polysomnography （PSG） is a useful tool for evaluating sleep quality. However, it requires subjects to sleep in an unfamiliar environment, which can influence sleep quality. Moreover, PSG is labor-intensive, timeconsuming, and expensive. We evaluated a portable two-channel electroencephalogram （EEG） monitoring system and compared the signals obtained from the device with those of simultaneously recorded full PSG.
Methods: Signals obtained from two-channel EEG were comparing with simultaneously recorded full PSG signals. Sleep stages were scored using the American Academy of Sleep Medicine Manual for Scoring Sleep 2007. The epochby-epoch percent agreement and Cohen’s kappa coefficient were used for agreement evaluation of sleep stage in both devices.
Results: The participants were healthy Japanese volunteers （mean （standard deviation）: age: 20.9 （1.8） years; seven women and nine men）. In epoch-by-epoch comparison, the average agreement and kappa value of sleep stages between two-channel EEG and PSG were 0.83 （0.04） and 0.75 （0.05）, respectively. Kappa coefficients showed strong agreement for stage R （REM: rapid eye movement）, stage W （wake）, stage N3 （non-REM: NREM 3）, and stage N2 （NREM 2）（0.86（0.09）, 0.76（0.12）, 0.74（0.15） and 0.73（0.06）, respectively） and weak agreement for stage N1（NREM 1）（0.44（0.13））.
Conclusion: These results demonstrate that two-channel EEG facilitates home sleep monitoring and exhibits acceptable agreement with PSG. Therefore, this tool may be suitable for use in epidemiological and intervention studies.