Measurement of the conductivity of the skull, temporarily removed during epilepsy surgery

For EEG source localization of brain activity, exact knowledge of the conductivity of the skull is essential, because it is low as compared to the other conductivities in the head. Simulation studies have shown that the depth of a source as determined while assuming a wrong skull conductivity may differ from its actual position by up to two centimeters[l]. A conductivity ratio between skin, skull, and brain of 1:1/80:1 is usually used, where the skull conductivity is 0.0048 S/m. This value has been reported by Rush and Driscoll in 1968[2] and has been questioned recently[3]. In that paper, the conductivity of post mortem skull material was measured and, additionally, an estimate was made based on a whole-head electrical impedance tomography (HIT) type of measurements. All data on human skull conductivity currently available is either based on post mortem material or on indirect HIT measurements. The drawback of using the first approach is that the bone is not in its natural condition. HIT measurements are performed I'M vivo, but accurate knowledge of skull geometry is needed, since in a model skull thickness and conductivity are interchangeable to some extent (although this may not be a problem if the identical model is used in the inverse problem[4]). During epilepsy surgery, a part of the skull is removed to gain access to the brain. This provides a unique opportunity to directly measure the skull conductivity, in vitro, but on a fresh skull part at body temperature, still saturated with its natural contents. We present a method to estimate the conductivity of this skull part, using an electrical impedance tomography approach.