Fig. 3: PD severity prediction from nocturnal breathing signals. | Nature Medicine

Fig. 3: PD severity prediction from nocturnal breathing signals.

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

Fig. 3

a, Severity prediction of the model with respect to MDS-UPDRS (two-sided t-test). The center line and the shadow denote the mean and 95% CI, respectively. b, Severity prediction distribution of the model with respect to the H&Y stage; a higher H&Y stage indicates increased PD severity (Kruskal–Wallis test). 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, Test–retest reliability of PD severity prediction as a function of the number of nights per subject. The dots and the shadow denote the mean and 95% CI, respectively. The model achieved a reliability of 0.97 (95% CI (0.95, 0.98)) with 12 nights of data. d, Correlation of the AI model predictions with MDS-UPDRS subpart I (two-sided t-test). e, Correlation of the AI model predictions with MDS-UPDRS subpart II (two-sided t-test). f, Correlation of the AI model predictions with MDS-UPDRS subpart III (two-sided t-test). g, Correlation of the AI model predictions MDS-UPDRS subpart IV (two-sided t-test). The center line and the shadow denote the mean and 95% CI, respectively. Data in all panels are from the wireless dataset (n = 53 subjects).

Back to article page