In search of biomarkers for schizophrenia using electroencephalography

The diagnostic process for schizophrenia is mainly clinical and has to be performed by an experienced psychiatrist, relying mainly on clinical signs and symptoms. Current neurophysiological measurements can distinguish groups of healthy controls and groups of schizophrenia patients. Individual classification based on neurophysiological measurements only shows moderate accuracy. In this study, we wanted to examine whether it is possible to distinguish controls and patients individually with a good accuracy. To this end we used a combination of features from different test paradigms, in particular the auditory and visual P300 and the mismatch negativity. We selected 54 patients and 54 controls, matched for age and gender, from the data available at the UPC Kortenberg. The EEG-data were high- and low-pass filtered, epoched, artefacts were rejected and the epochs were averaged. Features (latencies and amplitudes of component peaks) were extracted from the averaged signals. The resulting dataset was used to train and test classification algorithms. Here we applied Naïve Bayes and Decision Tree (without and with AdaBoost). A combination of three evoked potentials allowed us to accurately classify individual subjects as either control or patient. For the three investigated classifiers a total accuracy of more than 80%, a sensitivity of above 82% and a specificity of at least 78% was found.