Using Air Cushion Pressure Sensor to Develop the Monitoring Infant Respiratory System combined with The Novel Active Safety Device

碩士 === 南臺科技大學 === 電機工程系 === 107 === There are about 3,500 sleep-related deaths among U.S. babies each year, including sudden infant death syndrome (SIDS), accidental suffocation, and deaths from unknown causes, and these causes of death are strictly related to the respiratory system. Therefore, this...

Full description

Bibliographic Details
Main Authors: HUANG,TENG-YUE, 黃騰閱
Other Authors: 黃基哲
Format: Others
Language:zh-TW
Published: 2019
Online Access:http://ndltd.ncl.edu.tw/handle/5uj3a6
Description
Summary:碩士 === 南臺科技大學 === 電機工程系 === 107 === There are about 3,500 sleep-related deaths among U.S. babies each year, including sudden infant death syndrome (SIDS), accidental suffocation, and deaths from unknown causes, and these causes of death are strictly related to the respiratory system. Therefore, this paper develops a non-invasive baby sleep-breathing monitoring system with an active safety device to prevent baby’s suffocating. When a baby is lying on the air cushion with a pressure sensor, the baby's weak vibrations of respiratory in the chest can be detected, and then the breathing signal is classified into three different modes, namely breathing mode, active mode, and abnormal mode. In an experimental test, the subjects are ten infants with an age of 7.7±2.249 months. We measure the breathing signal, synchronously from our system, and the standard measuring instrument. The confusion matrix is used to verify the ability of the system to identify the activity mode. For RR interval of the respiratory signal and the number of breaths per minute calculated from two systems, we can obtain the correlation with each other. The results show that the average sensitivity and the average positive predictive value of the identification are 0.9835 and 0.9402, respectively. The average sensitivity and the average positive predictive value of the breathing pattern are 0.9844 and 0.9926, respectively. The results show the high correlation of the identification of activity pattern. The correlation coefficient (r) between each other for the R-R interval and the number of breaths per minute are 0.896 and 0.976, respectively. Besides, this study designed an imitation baby that can simulate various breathing mode. It is used to simulate the occurrence of respiratory arrest to verify the mode discrimination and alarm reaction time of the system. After ten tests, the average reaction time was 20.36±1.0177 seconds. And for testing the active safety device, the device turn-on the pump immediately after the trigger, so that the 8 kg simulated baby leaves the mattress within 11 seconds, and after 9 seconds, the airbag inflated to 0.4 kg/cm2 pressure. The airbags lift the breathable cloth 2.8 cm high to generate the air camber and to provide enough air for the baby. This device provides the most immediate treatment before the caregiver arrives.