Improving PET Imaging with New Methods
Author Information
Author(s): Elhamiasl Masoud, Jolivet Frederic, Rezaei Ahmadreza, Fieseler Michael, Schäfers Klaus, Nuyts Johan, Schramm Georg, Boada Fernando
Primary Institution: KU Leuven, Belgium
Hypothesis
Can we improve the quality of PET imaging by jointly estimating activity, attenuation, and motion without external hardware?
Conclusion
The proposed methods enable the reconstruction of a single, high-quality activity image that is motion-corrected and free from attenuation artifacts.
Supporting Evidence
- The proposed methods improved lesion-to-background contrast significantly.
- Image quality matched that of static acquisitions without motion artifacts.
- Both methods were evaluated on data from state-of-the-art PET scanners.
Takeaway
This study found new ways to take pictures of the body that help doctors see better by fixing problems caused by breathing during the scan.
Methodology
The methods were evaluated using data from a phantom and three clinical PET/CT datasets, comparing image quality and lesion uptake values.
Potential Biases
The ADMM-based method was sensitive to parameter tuning, which could lead to suboptimal performance.
Limitations
The study was limited to single bed position data and used a static scatter estimate instead of gated scatter estimation.
Participant Demographics
Data included a phantom and three patients with varying weights and lesion locations.
Want to read the original?
Access the complete publication on the publisher's website