STIMULATION-ACTIVATION STUDY USING (H2O)-O-15 AND PET - PRINCIPLE AND METHOD

: KANNO I, FUJITA H, IIDA H, LAW I

Publisher: BIOMED RES FOUND

: 1992

: Biomedical Research

: BIOMEDICAL RESEARCH-TOKYO

: BIOMED RES-TOKYO

: 13

: 45

: 50

: 6

: 0388-6107

The theoretical principle and method using H-2 O-15 and PET was described. Owing to a short tracer half-life (123 s), we can repeat measurements at a short interval for in vivo investigation of the higher brain functions. The method calculates CBF from time-integration of the H-2 O-15 distribution in brain tissue measured by PET. We have searched the method to optimize single to noise ratio in the temporal and spatial domains to maximize the signal-to-noise ratio. Optimal scantime of accumulation after bolus injection of H-2 O-15 was found to be from 90 to 120 s. A slow input method for temporal smooth of CBF fluctuation did not show any benefit compared to the bolus injection method. In spatial domain, an increase in full-width at half-maximum (FWHM) of reconstruction filter reduced statistical noise more rapidly than the signal deterioration. A phantom experiment directly showed that the larger FWHM gave a higher statistical t-value. In averaging inter-subject data, we transform individual brain into a standard brain atlas. The standardization allows us anatomical registration of activation foci. The Talairach atlas based on the anterior-commissure posterior-commissure (AC-PC) line is most commonly used. This adjusts brain size linearly in x, y and z directions. Non-linear techniques have also been developed by several other groups. Finally, we have described several topics from our experiences with stimulation-activation studies. Somatosensory and motor stimulation clearly activates circumscribed areas. But the language tasks on the other hand provides many weak foci in both hemispheres.