Heart rate based automatic seizure detection in the newborn

Abstract This work investigates the efficacy of heart rate (HR) based measures for patient-independent, automatic detection of seizures in newborns. Sixty-two time-domain and frequency-domain features were extracted from the neonatal heart rate signal. These features were classified using a sophisti...

Full description

Saved in:
Bibliographic Details
Published in:Medical engineering & physics 2010-10, Vol.32 (8), p.829-839
Main Authors: Doyle, O.M, Temko, A, Marnane, W, Lightbody, G, Boylan, G.B
Format: Article
Language:English
Subjects:
Artificial Intelligence
Automatic
Automation
Biological and medical sciences
Computerized, statistical medical data processing and models in biomedicine
Electrocardiography. Vectocardiography
Electrodiagnosis. Electric activity recording
Female
Headache. Facial pains. Syncopes. Epilepsia. Intracranial hypertension. Brain oedema. Cerebral palsy
Heart Rate
Humans
Infant, Newborn
Infant, Newborn, Diseases - diagnosis
Infant, Newborn, Diseases - physiopathology
Investigative techniques, diagnostic techniques (general aspects)
Linear Models
Male
Medical management aid. Diagnosis aid
Medical sciences
Nervous system (semeiology, syndromes)
Neurology
Newborn
Patient-independent
Probability
Radiology
Retrospective Studies
ROC Curve
Seizure detection
Seizures - diagnosis
Seizures - physiopathology
SVM
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract This work investigates the efficacy of heart rate (HR) based measures for patient-independent, automatic detection of seizures in newborns. Sixty-two time-domain and frequency-domain features were extracted from the neonatal heart rate signal. These features were classified using a sophisticated support vector machine (SVM) scheme. The performance was evaluated on a large dataset of 208 h from 14 newborn infants. It was shown that the HR can be useful for the detection of neonatal seizures for certain patients yielding an area under the receiver operating characteristic (ROC) curve of up to 82%. On evaluating the system using multiple patients an average ROC area of 0.59 with sensitivity of 60% and specificity of 60%, were obtained. Feature selection was performed and in the majority of patients the performance was degraded. Further analysis of the feature weights found significant variability in feature ranking across all patients. Overall, the patient-independent system presented here was seen to perform well in some patients (2 out of 14) but performed poorly when tested on the entire group.
ISSN:1350-4533
1873-4030
DOI:10.1016/j.medengphy.2010.05.010