Analysis of scalp-recorded potentials (EEG or evoked potential examination) is of great importance for studying bioelectrical processes in the human brain both in normal and pathological cases. Investigating potential distributions in the relation to individual topography of the cortex is helpful in treatment and surgery planning. We developed a method providing visual presentation of electrical potentials measured in EEG examination superimposed on a patient's cortex surface computed from MRI data. Analysis of such images contributes into diagnostics of common neurological disorders (for example: epilepsy, Parkinson's disease, any mechanical damage). Integration of EEG and MRI data sets requires their alignment. This paper focuses on applying a new registration technique for the alignment of these two data sets. The method does not require any external fiducial markers to be fixed on a patient's head, what makes it convenient for numerous examinations performed in clinic. Alignment bases on geometrical features derived from intrinsic data of an image or an object. To show the example of usage of the technique for assisting in neurological diagnostics we present results of SEP (sensory evoked potential) examinations performed for one pathological case and a control group.