Related Experiment Videos
[Study of the normal database that affects 3D-SSP Z-score mapping]
Yasushi Yamamoto1, Masuo Haramoto, Akio Komatsu
1Radiological Center, Shimane University Hospital.
Nihon Hoshasen Gijutsu Gakkai Zasshi
|May 8, 2004
Summary
Matching image reconstruction conditions for phantom normal databases (PNDB) and signal data is crucial for accurate 3D-SSP analysis. Using different devices or reconstruction methods significantly increases false-positive rates.
Area of Science:
- Medical Imaging
- Quantitative Analysis
Background:
- 3D-Single Photon SPECT (3D-SSP) analysis relies on accurate image reconstruction.
- Variations in reconstruction parameters and devices can impact Z-score mapping accuracy.
Purpose of the Study:
- To investigate the influence of image reconstruction conditions and devices on 3D-SSP analysis.
- To determine optimal strategies for creating and utilizing phantom normal databases (PNDB).
Main Methods:
- Utilized the 3D Hoffman brain phantom with IRIX and E.CAM devices.
- Employed filtered back-projection (FBP) and ordered-subset expectation maximization (OSEM) reconstruction methods.
- Varied Butterworth filter cutoff frequencies and attenuation correction application.
Main Results:
- High Z-scores observed when PNDB and signal data used OSEM reconstruction.
- Lowering Butterworth cutoff frequency with FBP increased Z-score and signal size, reducing false positives.
- Inconsistent reconstruction methods or devices between PNDB and signal data led to higher false-positive rates.
Conclusions:
- Matching image reconstruction conditions between PNDB and signal data is essential for reliable 3D-SSP analysis.
- Phantom normal databases are not interchangeable between different imaging devices.