Skip to main content
. 2020 Oct 17;9(10):giaa110. doi: 10.1093/gigascience/giaa110

Table 3:

Test set results

Antibiotic Lasso Cluster-lasso
Sensitivity Specificity bACC AUC Sensitivity Specificity bACC AUC
Amikacin 84.3 74.4 79.3 86.0 77.0 80.0 78.5 86.4
Aztreonam 69.6 80.0 74.8 83.3 67.2 80.0 73.6 82.0
Cefepime 77.0 60.4 68.7 69.1 78.3 54.7 66.5 69.6
Cefoxitin 51.7 92.8 72.2 74.6 55.2 94.2 74.7 76.0
Ceftazidime 77.7 98.4 88.1 94.3 60.8 96.0 78.4 92.8
Ciprofloxacin 91.3 91.2 91.2 96.6 92.1 89.8 90.9 96.7
Imipenem 65.3 99.0 82.2 87.2 65.6 98.3 81.9 85.4
Meropenem 66.0 97.7 81.8 81.1 66.3 97.7 82.0 78.8
Piperacillin/tazobactam 63.1 82.9 73.0 82.7 58.9 87.7 73.3 81.6
Tetracycline 64.8 93.5 79.2 82.4 64.5 94.8 79.7 82.8

This table summarizes the results obtained on the test dataset by the lasso and cluster-lasso models for the 10 antibiotics, in terms of sensitivity, specificity, balanced accuracy (bACC), and AUC.