ระบบตรวจสอบตำแหน่งผู้นอนเพื่อใช้ตรวจจับการลุกออกจากเตียง

Abstract

Bed-exit monitoring is critical for establishing the mobility trend of patients. The development of an intelligent bed system can be designed to monitor and prevent the bed-exit and to help immediately, especially for the elderly patients. In this paper present intelligent sleeping position determination system for bed-exit monitoring, the detection of the subject using 100 lbs FlexiForce sensors based on the basic amplifier was focused. Four sensors were installed at the base of the bed to produce a voltage output and investigate the signs of emerging changes. An analyzed mattress was divided into three parts (middle, left, right) to detect the patient motions in a full coverage area before and after leaving the bed. In this case, the voltage variation would be inputs of Support Vector Machine (SVM) classifier to identify the bed-exit mode, i.e., exit or not. The performance evaluation of the classifier shows that middle, left and right are classified at 91.10%, 96.13% and 95.18% classification accuracy respectively. The overall accuracy is 94.12%. This indicates that the system can be used for the bed-exit monitoring to help the patients in time.