[Paper] Cheyne-Stokes Respiration Detection Method for Newborns with Apnea

Abstract

278 IntroductionIn recent years, the rate of low birth weight has been increasing. Low birth weight (under 2500 g) may be caused by preterm delivery or agenesis in utero. The mortality rate of low birth weight babies is high because their respiratory and other body functions are immature.Prevalent respiratory diseases in preterm babies include neonatal respiratory distress syndrome, episodes of apnea, and sudden infant death syndrome (SIDS). The incidence of these diseases increases with lower gestational age. People with CSR do not respond quickly enough to changes in blood levels of oxygen and carbon dioxide.CSR has a high prevalence among newborns, especially among preterm newborns, whose central nervous systems are immature. CSR is triggered by hypoxia associated with hypoventilation 3) , because respiratory control is unstable when the central nervous Abstract Cheyne-Stokes respiration (CSR) has a high prevalence among newborns, especially among preterm babies. Although doctors generally recognize the phenomenon, they are not able to assess the severity of CSR in individual infants. CSR is characterized by cyclical weakening and strengthening of respirations with apnea. In this study, we developed an apnea detection method, and a CSR detection method using detected apneic events. We detected apnea using two features of respiratory waveforms. The first feature is frequency information calculated from wavelet coefficients. The second is information based on the shape of the waveform.In our CSR detection method, we used a spurious periodicity feature to determine CSR sections. The waveform is calculated by a respiratory monitoring system that uses a fiber-grating vision sensor to measure the vertical motion of the infant's thoracic and abdominal regions during respiration. Our method is effective at detecting apnea (sensitivity: 94.3 %, specificity: 99.7 %). To solve these problems, we developed a respiratory monitoring system for newborns using a fiber-grating (FG) vision sensor 7) . The FG vision sensor is an active stereo 3D sensor that allows noncontact 3D measurements 8) . This system can calculate more accurate respiratory waveforms than traditional medical sensors. Using this system, we developed a novel method to estimate the severity of CSR.Previous work using only the respiratory waveform shape or only frequency information is not sufficiently robust for apnea detection in newborns. Moreover, using a single feature or detection function results in large trade-offs between detection accuracy and threshold sensitivity.In this paper, we propose methods to detect apnea and CSR. Duration of apnea in newborns is shorter than in adults and neonatal respiratory waveforms tend to be distorted. To increase the robustness of detection accuracy, we adopted three different apnea detection functions using features of respiratory frequency and waveform shape information. The use of two different features improved detection accuracy. Our method also reduces false negatives by employing three different functions to ...