A METHOD FOR ASSESSING THE SIGNIFICANCE OF ABNORMALITIES IN HMPAO BRAIN SPECT IMAGES

Methods: A normal atlas for HMPAO rCBF SPECT images was obtained from images of 53 normal controls. Following image registration and normali zation, a mean image was extracted, while images representing correlat ed normal deviants were identified using principal component analysis. These images formed the ''building blocks'' of the atlas. For subsequ ent images, the atlas was used to create a ''nearest normal equivalent '' image, which was compared to a residual standard deviation image to determine the significance of deviations in the new image. Results: I mages from 30 patients (10 with Alzheimer's disease; 12 with single or multiple infarcts; and 8 normals) were analyzed. Conclusion: Using an optimal decision level, 10/10 patients with Alzheimer's disease and 1 1/12 patients with infarcts were correctly identified, with only one f alse-positive resulting. We utilized a database of images obtained fro m normal controls to create a normal atlas.