| Summary: | 碩士 === 國立臺北科技大學 === 電子工程系 === 107 === Emergency room is a very important medical department in hospitals. How to identify the patient's movement has always been an important issue. According to the statistic of the Ministry of Health and Welfare, the triage number of patients (level 1~3) is about 77% in the emergency room that the conditions of these patients is more serious. If there are no obvious symptoms, the hospital will ask them for hospital observation. However, these measurements of physiological parameters also cause the manpower wasting during the patient's staying in hospital. Therefore, this study designed a measuring system for wearable device that based on the Blood Pressure, Heart Rate, Oxygen saturation and Thermometer to help the long-term physiological parameters patients with monitoring in hospitals.
After using android phone for receiving the data and deleted the abnormal value, the data will be transported to the computer database for storage. Through the physiological parameters of the subject with classification algorithm to predict the emergency movements of patients, and establish a rating scale for the accurate model symptoms to help hospitals to administer the movement of emergency judgment. In conclusion, this study predicted the accuracy of all emergency movements was 0.63 and the ROC was 0.68. The accuracy of patients with urinary tract infection was 0.78 and the ROC was 0.83 and that is an excellent result with the classification. After establishing the emergency movement table, the point of hospitalization standard was 3 and the sensitivity was 87%, specificity was 63%, accuracy was 75%, which helps the emergency room to judge patient's movement.
|
|---|
