Switching Kalman filter based methods for apnea bradycardia detection from ECG signals

• Published in: Physiological measurement 2015-09, Vol.36 (9), p.1763-1783
• Main Authors: Montazeri Ghahjaverestan, Nasim, Shamsollahi, Mohammad B, Ge, Di, Hernández, Alfredo I
• Format: Article
• Language: English
• Subjects: Algorithms; Apnea - diagnosis; Apnea - physiopathology; Bioengineering; bradycardia; Bradycardia - diagnosis; Bradycardia - physiopathology; Databases, Factual; Diagnosis, Computer-Assisted - methods; Early Diagnosis; ECG modelling; electrocardiography (ECG); Electrocardiography - methods; expectation-maximization (EM); Heart - physiopathology; Humans; Infant; Infant, Newborn; Infant, Premature; Kalman filter (KF); Life Sciences; Linear Models; Models, Cardiovascular; Observational Studies as Topic; Sensitivity and Specificity; switching Kalman filter (SKF)
• ISSN: 0967-3334; 1361-6579
• DOI: 10.1088/0967-3334/36/9/1763

Apnea bradycardia (AB) is an outcome of apnea occurrence in preterm infants and is an observable phenomenon in cardiovascular signals. Early detection of apnea in infants under monitoring is a critical challenge for the early intervention of nurses. In this paper, we introduce two switching Kalman filter (SKF) based methods for AB detection using electrocardiogram (ECG) signal. The first SKF model uses McSharry's ECG dynamical model integrated in two Kalman filter (KF) models trained for normal and AB intervals. Whereas the second SKF model is established by using only the RR sequence extracted from ECG and two AR models to be fitted in normal and AB intervals. In both SKF approaches, a discrete state variable called a switch is considered that chooses one of the models (corresponding to normal and AB) during the inference phase. According to the probability of each model indicated by this switch, the model with larger probability determines the observation label at each time instant. It is shown that the method based on ECG dynamical model can be effectively used for AB detection. The detection performance is evaluated by comparing statistical metrics and the amount of time taken to detect AB compared with the annotated onset. The results demonstrate the superiority of this method, with sensitivity and specificity 94.74 and 94.17, respectively. The presented approaches may therefore serve as an effective algorithm for monitoring neonates suffering from AB.