Table 4.

Performance of the Traditional Machine-Learning Model on the Kidney Precision Medicine Project

Description	Minimal	Mild	Moderate	Severe
Precision	0.52 ± 0.26	0.30 ± 0.40	0.26 ± 0.20	N/A
Recall/sensitivity	0.49 ± 0.30	0.07 ± 0.08	0.27 ± 0.20	N/A
Specificity	0.63 ± 0.24	0.88 ± 0.17	0.79 ± 0.23	N/A

The trained model based on weighted neighbor distance using compound hierarchy of algorithms representing morphology on The Ohio State University Wexner Medical Center data set was used to predict on the data obtained from the Kidney Precision Medicine Project. Performance of the trained model on the Kidney Precision Medicine Project data set is shown. Data are expressed as means SD.

N/A, not available.