Validation of fetal auditory evoked cortical responses to enhance the assessment of early brain development using fetal MEG measurements
Liviu Moraru, Reza Sameni, Uwe Schneider, Jens Haueisen, Ekkehard Schleußner, Dirk Hoyer · Physiological Measurement · 2011
The maturation of fetal auditory evoked cortical responses (fAECRs) is an important aspect of developmental medicine, but their reliable identification is limited due to the technical restrictions in prenatal diagnosis. The signal-to-noise ratio of the fAECRs extracted exclusively from fetal magnetoencephalography is a known issue which limits their analysis as markers of brain development. The objective of this work was to develop a signal analysis strategy to address these problems and find appropriate processing steps. In this study, a group of 147 normal fetuses with gestations between 26 and 41 weeks underwent auditory evoked response testing. We combine different approaches that address data cleaning, fAECR determination and statistical fAECR validation to reduce the uncertainty in the detection of the auditory evoked responses. For the statistical validation of the evoked responses, we use parameters computed from bootstrap-based test statistics and the correlation between different averaging modes. Appropriate thresholds for those parameters are identified using linear regression analyses by looking at the maximum correlation coefficients. The results show that by using different validation parameters, the selected fAECRs conduct to similar regression slopes with an average of −13.6 ms/week gestational age which agree with previous studies. Our novel processing framework provides an objective way to identify and eliminate non-physiological variation in the data induced by artifacts. This approach has the potential to produce more reliable data needed in clinical studies for fetal brain maturation as well as extending the investigations to high-risk groups.