Loading...

Masoudi, Samira | 2014

854 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45923 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Shamsollahi, Mohammad Bagher
  7. Abstract:
  8. Apnea-bradycardia is a medical term for prolonged respiratory pause accompanied with a heart rate reduction which is a common event among preterm infants. Repetition of apnea-bradycardia episodescompromises oxygenation and tissue perfusion and may lead to neurological impairment or even short-term morbi-mortality. Main solution to this breathing-related disorder is continues monitoring of infants in neonatal intensive care units in order to detect apnea-bradycardia event, generate an alarm and warn available nurse or physician to initiate quick nursing actions. Various studies have been done in this area and different methods are proposed which mainly focus on cardiac signal processing. This project is also dedicated to the same context. In this study, methods based on Markovian models are implemented based on time series implying information about apnea-bradycardia episodes. Furthermore using time-series as single and multidimensional observations in Hidden Markov Model and Hidden semi Markov Model against considering some sort of interactions between them as in Coupled Hidden Markov Model have been studied. Finally results of implementation of these three algorithms and KNN, using four different time-series obtained from neonatal cardiac signals are presented. These time series consist of RR series, R-wave amplitude, QRS complex duration and extracted respiratory signal. Respiratory signal is obtained from ECG, by the proposed method based on adaptive filter structure. Results show that the algorithm based on two-channel Coupled Hidden Markov Model using RR-series and extracted respiratory signal has a noticeable performance for apnea-bradycardia detection. It should also be mentioned that the recorded data applied in this study belongs to cardiac signals of preterm infants in Rennes-I University hospital, Rennes, France
  9. Keywords:
  10. Early Detection ; Hidden Markov Model ; Apnea ; Electrocardiogram ; Preterm Infants ; Hidden Semi-Markov Model ; Coupled Hidden Markov Model

 Digital Object List

 Bookmark

No TOC