Related Experiment Video
Updated: Aug 6, 2026
Non-invasive Imaging and Analysis of Cerebral Ischemia in Living Rats Using Positron Emission Tomography with 18F-FDG
Published on: December 28, 2014
Effect of resolution recovery on statistical analysis of brain perfusion SPECT: Evaluation in a disease-insert
N Matsutomo1, M Fukami2, T Yamamoto2
1Department of Radiological Technology, Faculty of Health Science and Technology, Kawasaki University of Medical Welfare, 288 Matsushima, Kurashiki-City, Okayama 701-0193, Japan.
Introduction:
The purpose of this study was to use a Hoffman brain phantom with an insert simulating posterior cingulate cortex (PCC) hypoperfusion to assess how Z-score-based statistical outputs change when a filtered back projection (FBP)-based normal database (NDB) is applied to ordered subset expectation maximization (OSEM) brain perfusion single-photon emission computed tomography (SPECT) images reconstructed with resolution recovery.
Methods:
SPECT/computed tomography was performed with an Infinia 8 Hawkeye 4 system equipped with an extended low-energy general-purpose collimator. A custom PCC insert was fabricated using three-dimensional printing and mounted in the Hoffman phantom. Data were acquired with and without the insert, filled with iodine-123 solution. Images were reconstructed by FBP without attenuation correction, OSEM without attenuation correction, and OSEM with resolution recovery. Z-scores and Z-summation values were computed with a FBP NDB without attenuation correction and a FBP NDB with Chang attenuation correction.
Results:
Resolution recovery produced a sharper cortical outline with higher apparent cortical contrast. With the NDB without attenuation correction, Z-scores were higher across most regions with OSEM than FBP and were highest with resolution recovery, with several regions shifting from around the Z = 2 cutoff to above it. With the Chang-corrected NDB, Z-scores were higher overall and differences among reconstruction conditions widened. Z-summation in the precuneus was highest with resolution recovery, whereas the PCC peaked with OSEM without resolution recovery.
Conclusion:
When resolution recovery-processed images are analyzed using a conventional FBP NDB, Z-score-based indices shift in a region-dependent manner, and their level varies with NDB settings, including attenuation correction.
Implications For Practice:
The reconstruction protocol and NDB should be matched whenever possible, because mismatches may alter Z-score maps and affect cutoff-based interpretation of dementia-related hypoperfusion patterns.
