Georgopoulos AP et al. (2007) Synchronous neural interactions assessed by magnetoencephalography: a functional biomarker for brain disorders. J Neural Eng 4: 349–355

In a 2006 paper, Georgopoulos et al. hypothesized that synchronous neural interactions observed at high temporal resolution using magnetoencephalography (MEG) could have utility as functional biomarkers of brain disorders. They have now tested this idea by comparing normal brain activity with that of patients diagnosed with a variety of functional brain disorders including Alzheimer's disease, schizophrenia, multiple sclerosis, chronic alcoholism and Sjögren's syndrome.

The authors used MEG to record magnetic field strength from 248 axial gradiometers while subjects fixated on a spot of light for 45–60 s. Autoregressive integrative moving average (ARIMA) modeling of raw data from 52 subjects generated stationary residuals from which synchronous cross-correlations and partial correlations were calculated. Linear discriminant analysis was used to extract relevant information and classify patterns of neural activity that were consistent among patient groups, yielding a number of disease-specific predictor subsets that gave 100% correct classification considerably more often than would be expected by chance (z = 8.78; P <10−50). Cross-validation of the predictor subsets was performed in a further 46 patients who were successfully classified into their respective patient groups more than 90% of the time. Excellent (100% correct) classification was obtained in a total sample of 142 subjects. The predictive value of this method will continue to improve as more subjects and different disease groups are included in the analysis.

As a clinical application, MEG could represent a quick, simple and noninvasive alternative to behavioral examinations in the screening or differential diagnosis of functional brain disorders.