Maximizing Interpretability and Cost-Effectiveness of Surgical Site Infection (SSI) Predictive Models Using Feature-Specific Regularized Logistic Regression on Preoperative Temporal Data

Table 3

Evaluation of test data of SSI classification results using FLM for different values.

Name	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,	FLM,
Nr. features	4	10	21	29	29	40	37	34	33	44	34
AUC	0.966	0.967	0.961	0.961	0.961	0.961	0.961	0.962	0.961	0.962	0.962
AUPRC	0.882	0.872	0.880	0.874	0.874	0.875	0.873	0.873	0.871	0.877	0.874
Sensitivity	0.920	0.920	0.920	0.920	0.920	0.900	0.900	0.900	0.900	0.900	0.900
Specificity	0.871	0.865	0.893	0.893	0.893	0.920	0.910	0.900	0.900	0.920	0.900
PPV	0.667	0.657	0.701	0.701	0.701	0.780	0.763	0.734	0.726	0.763	0.726
NPV	0.975	0.975	0.975	0.975	0.975	0.971	0.970	0.970	0.970	0.970	0.970
Nr. predicted cases	69	70	65	65	65	59	61	62	62	59	62