Monica Cusenza

Linear and non-linear parameterization of EEG during monitoring of carotid endarterectomy

In this paper, new quantitative linear (HLF ratio: high frequency/low frequency spectral power ratio) and non-linear parameters (ZC: zero crossing and FD: fractal dimension) which can assist the physician in real-time decision whether a shunt is required or not during intra-operative EEG monitoring of carotid endarterectomy are presented. The results obtained with the new parameters are compared with those achieved by other indexes proposed in the literature. The HLF ratio and ZC parameters yielded the best results with a 100% of correct identification of both shunt and no-shunt situations. The ZC can be also easily implemented in the real-time monitoring of EEG.

EEG fractal dimension combined with burst suppression ratio as a measure of depth of anesthesia

This paper proposes an alternative index for the monitoring of depth of anesthesia. The study was carried out on 6 EEG recordings acquired during surgery under general anesthesia. Induction was provided by propofol, anesthesia was then maintained either with propofol or with sevoflurane under BIS control. The proposed FDSR parameter, combination of Higuchi’s fractal dimension and burst suppression ratio, is able to distinguish among different clinical states and detected two episodes of intraoperative awareness not highlighted by BIS. Moreover, due to its low computational complexity, FDSR is more suitable than BIS for real-time implementation. In conclusion, FDSR is a promising index for the monitoring of depth of anesthesia and the reduction of the incidence of intraoperative awareness.

Predictive Value on Neurological Outcome of Early EEG Signal Analysis in Brain Injured Patients

Electrophysiological examinations constitute objective and accurate measures of cerebral function. They can be recorded at bedside, which is of major value in intensive care units. The objective of the present study considers the linear and non-linear analysis of resting electroencephalography (EEG) signals as predictors for poor outcome and differences between brain regions. We studied 24 brain injured patients (trauma, cerebral anoxia, intracranial haemorrhage, cerebral infections) and 12 healthy controls and compared their EEG power spectra, the zero crossing and fractal dimension splitting the brain in four regions: left hemisphere, right hemisphere, anterior cortex, posterior cortex. In this study early linear and non-linear parameters showed the ability to predict recovery of communicative skills in brain injured patients at 6 months.

Predictive value of early linear and non-linear analysis of EEG signals in brain damage on neurological outcome

ESICM LIVES 2013 26th Annual Congress Paris, France 5–9 October This supplement issue of the official ESICM/ESPNIC journal Intensive Care Medicine contains abstracts of scientific papers presented at the 26th Annual Congress of the European Society of Intensive Care Medicine. The abstracts appear in order of presentation from Monday 7 October to Wednesday 9 October 2013. The same abstract numbering is used in the Congress Final Programme. This supplement was not sponsored by outside commercial interests; it was funded entirely by the society’s own resources. DOI:10.1007/s00134-013-3095-5 123 26th ANNUAL CONGRESS—PARIS, FRANCE—5–9 OCTOBER 2013 26th ANNUAL CONGRESS—PARIS, FRANCE—5–9 OCTOBER 2013

Real-Time EEG Parameterization for Shunt Decision Supporting System During Carotid Endarterectomy

Intraoperative EEG monitoring during carotid endarterectomy (CEA) is the common operation used to reduce the risk of brain ischemia. Beside visual assessment of the EEG, some quantitative parameters, based on spectral information, have been recently suggested as additional criteria for shunt need decision. In this paper we explore spectral power-based parameters and some non linear parameters, like zero crossing (ZC) and beta coefficient, in order to find the parameter/s that could constitute a good decision support system in shunt decision. The results, compared with those supplied by the Brain Symmetry Index, suggest that the ZC represents the best parameter in a real time analysis of EEG during CEA.