Extended Data Fig. 8: Evaluation results for predicting the full-night qEEG summary from nocturnal breathing signals. | Nature Medicine

Extended Data Fig. 8: Evaluation results for predicting the full-night qEEG summary from nocturnal breathing signals.

From: Artificial intelligence-enabled detection and assessment of Parkinson’s disease using nocturnal breathing signals

Extended Data Fig. 8

a, One prediction sample of a full-night qEEG. The time resolution of the predicted qEEG is 1 second. b, Distribution of the prediction errors across four EEG bands (n = 6,660 nights from 5,652 subjects). The AI model made an unbiased (that is, median-unbiased) estimation of EEG prediction for all bands. On each box, the central line indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to 1.5 times the interquartile range. c, Cumulative distribution functions of the absolute prediction error across four EEG bands (n = 6,660 nights from 5,652 subjects).

Back to article page