@article {Ishii2010, author = {Kazunari Ishii and Kohei Hanaoka and Masahiro Okada and Seishi Kumano and Yoshihiro Komeya and Norio Tsuchiya and Makoto Hosono and Takamichi Murakami}, title = {Clinical impact of CT attenuation correction by SPECT/CT in brain perfusion images}, volume = {52}, number = {supplement 1}, pages = {2010--2010}, year = {2011}, publisher = {Society of Nuclear Medicine}, abstract = {2010 Objectives To elucidate the regional differences between brain perfusion single photon emission computed tomography (SPECT) images reconstructed with a uniform attenuation correction using Chang{\textquoteright}s method (AC-Chang) and a non-uniform attenuation correction with CT in the SPECT/CT instrument (AC-CT). Methods Twenty-eight consecutive patients with brain disease examined by SPECT/CT brain perfusion imaging were selected and images were reconstructed with AC-Chang and with AC-CT. The AC-Chang and AC-CT reconstructed images were then compared by voxel-based analysis using three-dimensional stereotactic surface projections. Results Counts in the frontal area of the AC-Chang images were significantly higher than in the AC-CT images, while the counts in the margin of the frontal lobe and posterior margin of the parietal, occipital cortices and cerebellum of the AC-Chang images were significantly lower than those of the AC-CT images. Relative frontal perfusion was 5.0\% higher and relative cerebellar perfusion was 4.6\% lower in the AC-Chang image relative to the AC-CT image, on average. Conclusions We have demonstrated the frontal dominant hyper-perfusion and parieto-occipital and cerebellar hypo-perfusion in brain SPECT images reconstructed with AC-Chang compared to those reconstructed with AC-CT. We suggest that to obtain an accurate attenuation-corrected brain perfusion SPECT image, attenuation correction by Chang{\textquoteright}s method is inadequate and should be performed by CT using a SPECT/CT instrument}, issn = {0161-5505}, URL = {https://jnm.snmjournals.org/content/52/supplement_1/2010}, eprint = {https://jnm.snmjournals.org/content}, journal = {Journal of Nuclear Medicine} }