Deep venous thrombosis and pulmonary embolism are diseases associated with significant morbidity and mortality.Known risk factors are attributed for only slight majority ofvenous thromboembolic disease (VTE) with the remainder of risk presumably related tounidentified genetic factors. We designed a general purpose Natural Language (NLP) algorithmto retrospectively capture both acute and historical cases of thromboembolic disease in a de identified electronic health record.Applying the NLP algorithm to a separate evaluation setfound a positive predictive value of 84.7% and sensitivity of 95.3% for an Fmeasure of 0.897,which was similar to the training set of 0.925.Use of the same algorithm on problem lists onlyin patients without VTE ICD9s was found to be the best means of capturing historical cases witha PPV of 83%.NLP of VTE ICD9 positive cases and nonICD9 positive problem lists providesan effective means for capture of both acute and historical cases of venous thromboembolic
【 预 览 】
附件列表
Files
Size
Format
View
A Natural Language Processing Algorithm to define a VenousThromboembolism Phenotype