Fig. 4: Model evaluation for PD risk assessment before actual diagnosis, and disease progression tracking using longitudinal data. | Nature Medicine

Fig. 4: Model evaluation for PD risk assessment before actual diagnosis, and disease progression tracking using longitudinal data.

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

Fig. 4

a, Model prediction scores for the prodromal PD group (that is, undiagnosed individuals who were eventually diagnosed with PD) and the age- and gender-matched control group (one-tailed Wilcoxon rank-sum test). b, The AI model assessment of the change in MDS-UPDRS over 6 months (one-tailed one-sample Wilcoxon signed-rank test) and the clinician assessment of the change in MDS-UPDRS over the same period (one-tailed one-sample Wilcoxon signed-rank test). c, The AI model assessment of the change in MDS-UPDRS over 12 months (one-tailed one-sample Wilcoxon signed-rank test) and the clinician assessment of the change in MDS-UPDRS over the same period (one-tailed one-sample Wilcoxon signed-rank test). d, Continuous severity prediction across 1 year for the patient with maximum MDS-UPDRS increase (Kruskal–Wallis test; n = 365 nights from 1 September 2019 to 31 October 2020). For each box in a–d, 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.

Back to article page