An Automated System for Grading EEG Abnormality in Term Neonates with Hypoxic-Ischaemic Encephalopathy
- PDF / 892,234 Bytes
- 11 Pages / 593.972 x 792 pts Page_size
- 49 Downloads / 149 Views
An Automated System for Grading EEG Abnormality in Term Neonates with Hypoxic-Ischaemic Encephalopathy N. J. STEVENSON, I. KOROTCHIKOVA, A. TEMKO, G. LIGHTBODY, W. P. MARNANE, and G. B. BOYLAN Neonatal Brain Research Group, University College Cork, Cork, Ireland (Received 12 May 2012; accepted 20 November 2012; published online 4 December 2012) Associate Editor Leonidas D Iasemidis oversaw the review of this article.
is an evolving injury with a secondary injury occurring hours after the initial HI insult.2,16 Recent advances in treatment of neonatal HIE, principally the use of therapeutic hypothermia (TH), have rekindled interest in methods of monitoring the cortical function of the newborn with HIE in the neonatal intensive care unit (NICU).1,11 The electroencephalogram (EEG) is capable of passively monitoring neonatal cortical function. It is portable, provides minimal disturbance to the neonate, has a high time resolution and is capable of long duration recording in excess of 48 h. The visual interpretation, or grading, of the background EEG (EEG without seizure or artefact) has been shown to be a useful tool when monitoring the recovery of cortical activity after a HI injury.23,39 In fact, the normalisation of the EEG after a HI injury, a process whereby the EEG recovers through EEG grades from no electrical activity at the time of injury through periods of bursting activity to more continuous EEG with discernible sleep states, has been shown to correlate with outcome at 2 years of age.23 The aim of visual interpretation is to grade a period of background EEG, typically an hour, as normal or abnormal and then to grade the degree of abnormality. This visual interpretation incorporates EEG characteristics such as amplitude, continuity, frequency content, symmetry, synchrony, sleep state cycling and clinical information such as the gestational age of the neonate, suspected diagnosis and any administered medications.6,22,39 There are several grading or classification systems based on slightly different interpretations of these EEG and clinical characteristics and most interpretations correlate with neonatal outcome.44 The major impediment to the widespread use of the EEG in the NICU is that its interpretation is difficult,
Abstract—Automated analysis of the neonatal EEG has the potential to assist clinical decision making for neonates with hypoxic-ischaemic encephalopathy. This paper proposes a method of automatically grading the degree of abnormality in an hour long epoch of neonatal EEG. The automated grading system (AGS) was based on a multi-class linear classifier grading of short-term epochs of EEG which were converted into a long-term grading of EEG using a majority vote operation. The features used in the AGS were summary measurements of two sub-signals extracted from a quadratic time-frequency distribution: the amplitude modulation and instantaneous frequency. These sub-signals were based on a model of EEG as a multiplication of a coloured random process with a slowly varying pseudo-periodic waveform and may b
Data Loading...
