Different partial volume correction methods lead to different conclusions: An (18)F-FDG-PET study of aging

Area of Science:

• Neuroscience
• Medical Imaging
• Gerontology

Background:

• Brain metabolism changes with age, but studies yield conflicting results.
• Partial volume correction (PVC) methods are used to improve accuracy in (18)F-FDG-PET imaging.
• The impact of various PVC methods on aging brain metabolism studies is not fully understood.

Purpose of the Study:

• To compare the effects of different partial volume correction (PVC) methods on assessing age-related brain metabolism changes using (18)F-FDG-PET.
• To evaluate the sensitivity of PVC methods to parameter selection.
• To determine the influence of cerebrospinal fluid (CSF) modeling on detecting age-related metabolic decline.

Main Methods:

• Cross-sectional study of 99 subjects (aged 65-87) from the Harvard Aging Brain study.
• Utilized (18)F-FDG-PET imaging with four PVC methods: no correction (NoPVC), Meltzer (MZ), Müller-Gärtner (MG), and symmetric geometric transfer matrix (SGTM).
• Assessed sensitivity to parameter settings for MZ and MG, and evaluated the impact of excluding CSF.

Main Results:

• PVC methods produced a wide range of conclusions regarding age-related metabolic changes.
• NoPVC showed significant bias, overestimating age effects.
• SGTM identified significant age-related metabolic decreases (approx. 7% per decade) in frontal, cingulate, visual, and insular cortices.
• Excluding CSF from MG and SGTM models reduced the power to detect age-related changes.

Conclusions:

• The choice of PVC method critically impacts the interpretation of aging brain metabolism studies.
• Standard methods like NoPVC and MZ may be biased or insensitive.
• SGTM, when accounting for CSF, offers a more reliable assessment of age-related metabolic changes.
• Careful consideration of PVC methods and CSF is essential for accurate interpretation of aging brain metabolism data.