Nonlinear EEG biomarker profiles for autism and absence epilepsy

摘要

BackgroundAlthough autism and epilepsy are considered to be different disorders, epileptiform EEG activity is common in people with autism even when overt seizures are not present. The relatively high comorbidity between autism and all epilepsy syndromes suggests the possibility of common underlying neurophysiological mechanisms. Although many different epilepsies may be comorbid with autism, absence epilepsy is a generalized epilepsy syndrome with seizures that appear as staring spells, with no motor signs and no focal lesions, making it more difficult to diagnose. Application of nonlinear methods for EEG signal analysis may enable characterization of brain activity that can help to delineate neurophysiological commonalities and differences between autism and epilepsy. Multiscale entropy and recurrence quantitative analysis (RQA) were computed from EEG signals derived from children with autism or absence epilepsy and compared with the goal of finding significant and potentially clinically useful biomarkers neurophysiological differences between these two childhood disorders. MethodsMultiscale entropy and a multiscale version of RQA were computed from EEG data obtained from 92 children were collected in two different settings at Boston Children’s Hospital. Short segments of alert resting state EEG were selected for analysis. A complexity index derived from entropy and RQA methods was computed from each of 19 standard EEG channels for all subjects using publicly available software. Statistical comparisons were made between the groups. Machine learning classifiers were also used to determine which derived features were most significantly different among the groups, and to determine classification specificity and sensitivity. ResultsSignificant differences were found between absence, autism, and control groups in a number of different scalp locations and the values of complexity index. Autism values appeared to be intermediate between epilepsy and control in many locations, and differences between controls and absence patients were more widely distributed across scalp locations. Classification algorithms were able to distinguish absence epilepsy and autism cases from controls with high (>95%) accuracy. Importantly, two independent control groups, although they were derived from different settings and with different equipment were statistically indistinguishable. ConclusionsSignficant neurophysiological differences were found between absence, autism, and control cases. In most scalp regions, autism values were intermediate between the control values and absence values, suggesting several future research studies. Nonlinear EEG signal analysis, together with classification methods, may provide complementary information to visual EEG analysis and clinical assessment in epilepsy and autism, and may provide useful information for research on pediatric neurodevelopmental and neurological disorders. Additional research may enable neurophysiological biomarker profiles to be derived from these techniques for clinical use.