Issue No. 01 - January (1996 vol. 16)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/38.481564
A quarter of epilepsy patients in the United States, approximately 125,000 with an estimated annual increase of 5,000, have seizures uncontrolled by medication and are candidates for surgical treatment. Planning of epilepsy surgery often includes invasive, intracranial EEG recordings to localize the region of seizure onset for resection. Such invasive procedures impose considerable costs and definite risks, ranging from $40,000 to $80,000 per session and 6 to 8 percent morbidity rate. In this article, we propose an alternative approach that is noninvasive and costs about $3,000 to $4,000 per patient. This article describes computing tools developed to quantitate and combine different biomedical information of four neuroimaging modalities, that is, magnetic resonance imaging (MRI), positron emission tomography (PET), magnetic resonance spectroscopy (MRS), and magnetoencephalography (MEG). Multimodal image fusion aims to decrease the need for invasive intracranial EEG monitoring. As a result, it provides major cost savings, reduced patient risks, and increased availablility for epilepsy surgery. Selected patient cases from the Northern California Comprehensive Epilepsy Center are also provided to illustrate the clinical application of these techniques.
Multimodal brain imaging, image registration, noninvasive surgical planning, minimum invasive medicine, medical image databases, PACS
Kenneth D. Laxer, Robert C. Knowlton, Randy A. Hawkins, Stephen T.C. Wong, "Multimodal Image Fusion for Noninvasive Epilepsy Surgery Planning", IEEE Computer Graphics and Applications, vol. 16, no. , pp. 30-38, January 1996, doi:10.1109/38.481564