Table 2.

Machine Learning Models for Prediction of Incident Clinical PD

		AUC (95%CI)
Model	Control group (n = 251)	Case 3-year (PD n = 25)	Case 5-year (PD n = 41)
1*	LB-Unknown (135)	0.64 (0.51–0.76)	0.61 (0.52–0.71)
	LB-No (32)
	LB-Yes (84)
2	LB-Unknown (135)	0.71 (0.59–0.83)	0.61 (0.51–0.71)
	LB-No (32)
3	LB-No (32)	0.79 (0.67–0.91)	0.73 (0.62–0.85)
4**	LB-Unknown (135)	0.82 (0.76–0.89)	0.77 (0.71–0.84)
	LB-No (32)

^*We also ran a model by using LB-Yes patients as cases and obtained AUC of 0.63 (0.56–0.70) for 3-year prediction window and AUC of 0.61 (0.55–0.68) for 5-year prediction window. ^**Controls with LB at autopsy (LB-Yes) were reannotated as PD for model development but excluded from tests of model performance. When we implemented Model 3, among 84 LB-Yes controls, 38 were classified as cases (PD) and 46 as controls). Using this evidence, in Model 4, we rebuilt a model by using these 38 as cases and 46 as controls. In addition, to compare the robustness of Model 4 with Model 3, we further excluded LB-Unknown patients in the AUC calculation and obtained an AUC of 0.91 (0.82–0.99) for 3-year prediction window and 0.80 (0.70–0.90) for 5-year prediction.