Aernoud Fiolet
333 Text-mining in electronic healthcare records can be used for screening and data collection 2b. Accuracy of variables per participating medical center Variable Hospital A, accuracy % Hospital B, accuracy % Hospital C, accuracy % Sex 98.5 100.0 100.0 Current smoker 85.8 83.1 85.3 Hypertension 64.2 52.5 63.8 Diabetes 97.0 88.1 95.4 Insulin dependent diabetes 90.3 93.2 90.6 Renal function 95.5 100.0 87.9 Prior ACS 85.8 79.7 84.4 Prior PCI 93.3 88.1 93.5 CABG 98.5 96.6 97.4 AF 88.8 88.1 85.0 Gout 93.3 91.5 92.8 As or PI 80.6 86.4 60.3 Anticoagulant 90.3 94.9 91.9 No APT or OAC 99.3 100.0 98.4 Statin 90.3 76.3 87.9 Ezetimibe 85.8 89.8 87.0 ACE inhibitor 89.6 94.9 86.0 Beta blocker 66.4 84.7 74.6 Calcium channel blocker 75.4 78.0 83.7
Made with FlippingBook
RkJQdWJsaXNoZXIy ODAyMDc0