C-STAGE, AUTOMATED SLEEP SCORING - DEVELOPMENT AND COMPARISON WITH HUMAN SLEEP SCORING FOR HEALTHY OLDER MEN AND WOMEN

Using the sleep records of 200 men and women (age 55-85 years), we hav e developed a human-assisted computer scoring system, C STAGE. The sys tem can have many applications, including quantitative electroencephal ographic (EEG) analysis during specific stages of sleep. C STAGE class ifies sleep/wake stages using power spectral analysis and other techni ques applied to one channel of EEG data. Here we report comparability data between C STAGE- and human-rated sleep-stage scoring using Rechts chaffen and Kales criteria for 70 normal subjects (a subset of the 200 ). Because the method was developed using these subjects, we also repo rt comparability data for an independent validation sample of 45 norma l older men and women. For waking measures, sleep stages 3 and 4, and total sleep time, C STAGE yielded ratings comparable with the human ra ter (r = 0.73-0.91; p < 0.001). For sleep stages 1 and 2 and REM sleep , C STAGE correlated less well with human ratings (r = 0.59-0.81; p < 0.001). Overall, these correlations compare well with other currently available computer stage-scoring methods. Epoch-by-epoch comparisons i n the validation sample revealed a mean proportion of agreement of 0.7 4 and a mean Kappa coefficient of 0.57, indicating the two methods pro vide reasonable agreement on an epoch-by-epoch basis. We conclude that C STAGE is a valid sleep/waking scoring system for healthy older adul ts.