Figure 3 | Scientific Reports

Figure 3

From: Development and validation of prognostic machine learning models for short- and long-term mortality among acutely admitted patients based on blood tests

Figure 3

Sensitivity of fifteen ML algorithms on training data, as measured by the Area Under the Receiver Operating Characteristic Curve (AUC) when using 1 to 27 variables as predictors in the machine learning algorithms. Figure 3a–d demonstrate the sensitivity for 3-day, 10-day, 30-day and 365-day mortality on training data, respectively. Among all models, the highest sensitivity is shown between 0.88 and 0.91 in Fig. 3a–c. In Fig. 3d, the highest sensitivity reached is 0.85. When using ten variables, the top 3 models achieved a sensitivity above 0.85–91 in Fig. 3a–d, The Sensitivity falls below 0.8 when using fewer than three variables for all models in Fig. 3a–d.

Back to article page