Extracting VOIs from brain PET images.

Area of Science:

• Neuroimaging
• Medical Physics
• Radiochemistry

Background:

• Accurate delineation of volumes of interest (VOI) is crucial for quantitative analysis of Positron Emission Tomography (PET) data.
• Traditional manual VOI definition is time-consuming and prone to inter-observer variability.
• Existing automated methods may not fully leverage functional PET data or integrate anatomical validation.

Purpose of the Study:

• To develop and evaluate a semi-automatic system for VOI determination in brain PET imaging.
• To assess the efficiency and accuracy of the system compared to manual methods.
• To utilize functional PET data for VOI extraction with anatomical image cross-validation.

Main Methods:

• A semi-automatic system employing user-selectable thresholds and 3D flood-fill algorithms for VOI surface extraction.
• Functional PET images were used for initial VOI determination.
• Extracted VOIs were validated against anatomical images.
• The system was evaluated using brain 18F-DOPA PET studies targeting the striatum.

Main Results:

• The semi-automatic system achieved VOI determination comparable in accuracy to manual methods.
• The time required for target extraction was reduced to approximately one-third of the manual method.
• The system successfully delineated the striatum in FDOPA-PET studies.

Conclusions:

• The developed semi-automatic system provides an efficient and accurate alternative for VOI analysis in brain PET imaging.
• This approach offers significant time savings without compromising the quality of results.
• The method is particularly applicable to functional PET studies like FDOPA-PET for striatal analysis.