How processing choices effect repeatability in BOLD-CVR imaging

Area of Science:

• Neuroimaging
• Clinical Biomarkers
• Cerebrovascular Physiology

Background:

• Cerebrovascular reactivity (CVR) is a key clinical biomarker.
• Accurate CVR quantification is essential but challenging due to methodological variability.
• Lack of a gold standard complicates CVR measurement.

Purpose of the Study:

• To systematically evaluate the impact of different processing strategies on CVR map repeatability.
• To provide practical insights for consistent CVR data analysis across studies.
• To guide researchers in optimizing CVR quantification methods.

Main Methods:

• BOLD-CVR imaging was performed on 24 healthy volunteers.
• Multiple CVR paradigms were used: CO2 inhalation, breath-hold, and resting-state.
• Test-retest imaging was conducted to assess repeatability using spatial intraclass correlation.

Main Results:

• Optimal CVR processing strategies differed across paradigms.
• Processing choices, such as motion confound inclusion and temporal filtering, significantly affected repeatability.
• Interactions between processing steps influenced CVR map consistency.

Conclusions:

• Methodological choices in CVR data processing critically impact repeatability.
• Understanding these impacts is crucial for ensuring reliable and consistent CVR measurements.
• This study offers practical guidance for researchers to improve CVR analysis consistency.