Aernoud Fiolet

335 Text-mining in electronic healthcare records can be used for screening and data collection 2d. Overall test performance scores of variables Variable Sensitivity Specificity Positive predictive value Negative predictive value F1-score Sex 99.5 100.0 100.0 97.7 0.998 Current smoker 44.9 91.6 46.3 91.2 0.456 Hypertension 89.6 31.5 60.2 72.3 0.720 Diabetes 92.3 95.6 82.4 98.2 0.870 Insulin dependent diabetes 90.0 90.9 38.6 99.3 0.540 Renal function 93.7 67.4 96.8 50.0 0.952 Prior ACS 93.6 50.0 87.2 68.4 0.903 Prior PCI 98.6 52.4 93.5 84.6 0.960 CABG 91.4 98.4 88.3 98.9 0.898 AF 30.0 95.6 52.5 89.3 0.382 Gout 52.6 96.1 52.6 96.1 0.526 As or PI 98.6 1.9 69.3 37.5 0.814 Anticoagulant 97.3 90.8 65.2 99.5 0.781 No APT or OAC N/A 99.0 N/A 99.8 N/A Statin 93.7 7.9 92.5 9.4 0.931 Ezetimibe 78.6 89.6 69.7 92.3 0.739 ACE inhibitor 97.7 64.9 86.9 92.3 0.920 Beta blocker 97.9 21.4 72.8 82.9 0.835 Calcium channel blocker 91.2 76.9 59.6 95.9 0.721

RkJQdWJsaXNoZXIy ODAyMDc0