Abstracts

Rationale: Electroencephalography (EEG) has been used in veterinary medicine for many years to detect pathological interictal and ictal activity. Although a ten-twenty system for electrode placement has never been developed in dogs due to variations in head sizes and morphology, several methods have been proposed to establish coherence between anatomical landmarks and electrode positioning. The purpose of this study was to evaluate an innovative and non-invasive method of positioning subdermal EEG electrodes in close proximity to the desired cortical areas in the dog (i.e frontal and temporal lobes). Methods: A research dog was anesthetized to undergo brain MRI. Prior to MRI scanning, a dental bite block with laterally located fiducial markers was placed and secured in the dog s mouth. A 3D T1 weighted volume was acquired with the dog placed in sphinx position. Post-MRI, 3D image reconstructions of the brain, skull and skin were performed using Brainsight software. The dog was registered to his MRI scan by identifying homologous fiducial marker points using an optical position sensor and a neuronavigation pointer that displayed real-time images on a computer screen. Subdermal wiring EEG electrodes (SWE) were positioned in approximate locations under the scalp using an electrode montage routinely performed on a clinical basis without the use of the neuronavigation (Method 1). Localization of each electrode was then established to identify its position on the head and the underlying anatomical brain region using neuronavigation. The same electrode montage was then performed a second time using the navigation system to intentionally locate the SWE in direct proximity to desired cortical regions represented by electrode convention (ie. F3 = left frontal lobe) (Method 2). This second methodology was used as the gold standard for determination of accuracy in electrode positioning. The two techniques of electrode placement were then compared based on the overall mean electrode placement error represented in x, y and z between Methods 1 and 2 (Fig.1). Results: The technique used in Method 1 demonstrated a variation in electrode positioning when compared to the gold standard Method 2. A total of 8 electrode sites were used for the analysis. The mean electrode placement error for all SWE was 14.85 /- 7.69 mm when comparing both methods. Frontal (F3 and F4) and parietal (P3 and P4) electrodes were most closely associated with the corresponding cortical lobes whereas occipital (O1 and O2) and temporal (T3 and T4) electrodes were the most divergent. In Method 1, three out of eight electrodes were placed over the wrong cortical anatomical region. Conclusions: The use of a brain MRI compatible navigation system is an important technique to position EEG electrodes over their corresponding cortical anatomical lobe and strongly suggests that recorded EEG activity originates from the desired cortical area. This application may become useful for clinical and research purposes and may be extended to other species for comparative studies.