Search

Background/purpose: We aim to evaluate the accuracy of the new prehospital notification criteria for patients with potential acute stroke in the prehospital setting.Methods: We conducted a retrospective observational study from March 2011 to February 2013 of potential acute stroke patients prenotified using the new criteria which were: (1) positive Cincinnati Prehospital Stroke Scale (CPSS); (2) symptom onset within 3 hours; and (3) blood glucose level > 60 mg/dL. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the new criteria were calculated and outcomes of acute stroke patients were reported. Data of all patients with stroke or transient ischemic attack (TIA) transported to the destination hospital were also obtained to evaluate the compliance of emergency medical technicians.Results: There were 2888 patients suspected of stroke by emergency medical technicians and 221 patients prenotified due to meeting the criteria. The PPV, NPV, sensitivity, and specificity of the new criteria were 76.9%, 96.6%, 64.9%, and 98.1%, respectively. Onset time > 3 hours (24/51, 47.1%) and seizure (27.5%) were the two most common conditions leading to false prenotification. Of all prenotified patients, 23.1% (51/221) received thrombolytic therapy. Hemorrhagic stroke or ischemic stroke with hemorrhagic transformation (53.8%) and minor symptoms or rapid recovery (26.9%) were the most common reasons excluding correctly prenotified patients from thrombolytic therapy.Conclusion: The accuracy of the new prehospital stroke criteria has higher PPV and specificity compared to previous CPSS validation studies. [ABSTRACT FROM AUTHOR]

In the proposed approach, financial data are collected automatically and are input to a GM(1,1) prediction model to forecast the future trends of the collected data. The forecast data are then reduced using a GM(1,N) model, classified using a K-Means clustering algorithm, and processed by a Rough Set (RS) classification module which selects appropriate investment stocks. Finally, a Minimum Variance At Risk (MVAR) scheme comprising five Grey Relational Analysis (GRA) models and the Markowitz means-variance (MV) model is employed to identify the efficient frontier which achieves the optimal trade off between the minimum risk and the maximum rate of return. [ABSTRACT FROM AUTHOR]