Brain mapping: Validation and applications of structural and functional brain imaging

The accurate correlation of complementary structural and functional images of the human brain (brain mapping) is now an essential component of studies of brain functional anatomy. This paper describes a generalized validation procedure for multi-modal image registration in an individual subject. We first measured the distortions of the magnetic resonance imaging (MRI) and positron emission tomography (PET) scanners used for structural and functional imaging, and implemented distortion corrections. We then designed and built a multi-modal validation phantom with markers that are visible in MRI, PET, and SPECT. We scanned the phantom in MRI and PET, corrected the resulting images for distortion, correlated the images using a well-known registration algorithm, and determined the accuracy of the resulting fit. The procedure is general enough for validating most single-subject, multi-modal correlation protocols, and the results are useful in designing automated correlation procedures.