Effect of the data collection time
We examined the effect of the length of data collection time on the model’s ability to predict revision ESS risk by using only Logistic Regression model (Figure 2), because the three classifiers performed with a similar pattern and with a fair discrimination ability (the highest AUC values between 0.7-0.8, see Figure 1). We found that the highest AUC value was obtained when the variable data were collected from the baseline visit until 12 months after the baseline ESS (.81, Figure 2, Table S7), as compared with data collection time from baseline visit until 0, 3 and 6 months after the baseline ESS.