Electroencephalographic analysis for the assessment of hepatic encephalopathy: comparison of non-parametric and parametric spectral estimation techniques.

OBJECTIVE: To compare electroencephalographic spectral analysis obtained by periodogram (calculated by means of Fast Fourier Transform) and autoregressive (AR) modelling for the assessment of hepatic encephalopathy. METHODS: The mean dominant frequency (MDF) and the relative power of delta, theta, alpha, and beta bands were computed by both techniques from the electroencephalograms (EEG) of 201 cirrhotics and were evaluated in the clinical and prognostic assessment of the patients. RESULTS: The values of all the five indexes computed by periodogram and AR modelling matched each other, but the latter provided stable values after the analysis of fewer epochs. Independently of the technique, the relative power of theta and alpha bands fitted the clinical data and had prognostic value. The relative power of beta and delta bands computed by AR modelling fitted more closely with clinical data fitted the clinical data more closely. CONCLUSIONS: The electroencephalographic spectral indexes obtained by periodogram and AR modelling were found to be, on average, undistinguishable, but the latter appeared less sensitive to noise and provided a more reliable assessment of low-power bands.

The EEG assessment of low-grade hepatic encephalopathy: comparison of an artificial neural network-expert system (ANNES) based evaluation with visual EEG readings and EEG spectral analysis.

OBJECTIVE: The EEG provides an objective staging of hepatic encephalopathy (HE), but its interpretation may be biased by inter-observer variability. This study aims at comparing an entirely automatic EEG classification of HE based on an artificial neural network-expert system procedure (ANNES) with visual and spectral analysis based EEG classifications. METHODS: Two hundred and thirty-eight consecutive cirrhotic patients underwent closed-eye EEG. They were followed up for up to one-year to detect bouts of overt HE and death. The EEG was classified by ANNES, qualitative visual reading, main basic rhythm frequency and spectral analysis. The classifications were assessed on the basis of: (i) match with liver function, (ii) prognostic value and (iii) repeatability. RESULTS: All classifications were found to be related to the severity of liver failure, with cognitive findings and a history of previous bouts of HE. All of them had prognostic value on the occurrence of overt HE and on survival. The ANNES based classification was more repeatable than the qualitative visual one, and had the advantage of detecting low power EEG, but its efficiency in analyzing low-grade alterations was questionable. CONCLUSIONS: An entirely automatic - ANNES based - EEG classification of HE can improve the repeatability of EEG assessment, but further improvement of the device is required to classify mild alterations. SIGNIFICANCE: The ANNES based EEG grading of HE needs further improvements to be recommended in clinical practice, but it is already sufficient for detecting normal and clearly altered EEG tracings.

Electroencephalographic staging of hepatic encephalopathy by an artificial neural network and an expert system.

AIM OF THE STUDY: To provide an objective EEG assessment of hepatic encephalopathy (HE), we set up and tested an entirely automatic procedure based on an artificial neural network-expert system software (ANNESS). PATIENTS AND METHODS: A training set sample of 50 EEG (group A) and a test sample of 50 EEG (group B) of 100 cirrhotic patients were considered. The EEGs had been visually classified by an expert electroencephalographer, using a modified five-degree Parsons-Simith classification of HE. The efficiency of the ANNESS, trained in group A, was tested in group B. RESULTS: Both the ANNESS and the visually-based classifications were found to be correlated to liver insufficiency, as assessed by the Child-Pugh score (Spearman's coefficient rho=0.485, P<0.0001; rho=0.489, P<0.0001, respectively) and by the biochemical indexes of hepatic function (bilirubin: rho=0.31 vs. 0.27; albumin: rho=-0.13 vs. -0.18; prothrombin time rho=-0.35 vs. -0.52). The classifications were found to be correlated to each other (rho=0.84 P<0.0001, Cohen's kappa=0.55). However, the ANNESS overestimated grade 2 EEG alterations. CONCLUSION: An ANNESS-based classification of EEG in HE provided data comparable with a visually-based classification, except for mild alterations (class 2) that tended to be overestimated. Further optimization of automatic EEG staging of HE is desirable, as well as a prospective clinical evaluation.