Does applying resolution recovery to normal databases confer an advantage over conventional 3D-stereotactic surface...

Area of Science:

• Nuclear Medicine
• Neuroimaging

Background:

• Accurate interpretation of brain SPECT scans relies on robust normal databases (NDBs).
• Existing NDBs may not fully leverage advanced reconstruction techniques like resolution recovery (RR).

Purpose of the Study:

• To evaluate a novel NDB derived from SPECT/CT data using advanced reconstruction methods.
• To compare the performance of the novel NDB against conventional methods using simulated and patient data.

Main Methods:

• Generation of a novel NDB using SPECT/CT data from healthy brains.
• Data reconstruction via three-dimensional ordered subset expectation maximization (3D-OSEM) with scatter correction (SC), attenuation correction (AC), and RR.
• Analysis of Z, extent, and error scores using simulated hypoperfusion data and patient data from the precuneus, cuneus, and posterior cingulate.

Main Results:

• Simulated data showed decreased Z-scores with the novel NDB and RR, but increased extent scores in the posterior cingulate.
• The novel NDB without RR reduced error scores by 15% compared to conventional NDBs.
• Patient data revealed decreased Z-scores and extent scores with the novel NDB and RR, though posterior cingulate extent scores improved.

Conclusions:

• The novel NDB shows potential, particularly in improving posterior cingulate analysis in SPECT imaging.
• The application of RR with the novel NDB necessitates a re-evaluation of current Z-score thresholds for accurate interpretation.
• Further research is needed to optimize the use of RR in conjunction with advanced NDBs for brain SPECT analysis.