Spatio-temporal multiple source localization by wavelet-type decomposition of evoked potentials

Characterizing somatosensory evoked potential sources with dipole models: Advantages and limitations

Several methods have been developed to investigate the cerebral generators of scalp somatosensory evoked potentials (SEPs), because simple visual inspection of the electroencephalographic signal does not allow for immediate identification of the active brain regions. When the neurons fired by the afferent inputs are closely grouped, as usually occurs in SEP generation, they can be represented as a dipole, that is, as a linear source with two opposite poles. Several techniques for dipolar source modeling, which use different algorithms, have been employed to build source models of early, middle-latency, and late cognitive SEPs. Modifications of SEP dipolar activities after experimental maneuvers or in pathological conditions have also been observed. Although the effectiveness of dipolar source analysis should not be overestimated due to the intrinsic limitations of the approach, dipole modeling provides a means to assess SEPs in terms of cerebral sources and voltage fields that they produce over the head.

Auditory evoked potentials in multiple sclerosis: correlation with magnetic resonance imaging

The present study addresses issues regarding the location of neural sources (i.e. generators) of human auditory evoked potentials (AEPs), and the pattern of neural conduction in the auditory pathway. AEPs were recorded from fifteen patients with multiple sclerosis (MS) and compared to normals. The recordings included auditory brainstem responses (ABRs), mid-latency responses (MLRs), and long-latency responses (LLRs). AEP latency abnormalities were related to the locus of demyelinating lesions, as determined by magnetic resonance imaging (MRI) scans. The data demonstrated several anatomical patterns relating abnormal ABR wave intervals and abnormal MRI signals. From these patterns specific loci for ABR neural sources in the brainstem might be postulated. In addition, the earlier the ABR waves, the more unilateral the abnormalities appeared, suggesting bilateral sources for later waves. The MLRs were highly correlated with ABR wave V and were associated with greater abnormality in MRI signals in midbrain and forebrain regions. In general, patients with abnormal LLRs also had widespread AEP and MRI abnormalities, supporting a multiple source approach for the N1 wave of the LLRs. The observation that LLRs were only abnormal in the presence of bilateral ABR abnormalities suggests a cross wiring which would serve as a compensatory mechanism for unilateral disturbances. The AEP data showed dissociation between early and late wave abnormalities, thus supporting parallel channels for neural conduction in the central auditory system. Such a model calls for some degree of independence of AEP generators along the auditory pathway.