Xin, Lyu (2025) Optimizations of Q. Clear image reconstruction method for brain 18F PET/CT studies. Project Report. Universiti Sains Malaysia. (Submitted)
![[img]](http://eprints.usm.my/style/images/fileicons/application_pdf.png) |
PDF
Download (826kB) |
Abstract
Brain ¹⁸F-FDG PET/CT is a crucial tool in assessing cerebral metabolism and diagnosing neurological conditions. However, dynamic acquisitions with short frame durations suffer from low count statistics and high noise levels, posing challenges for both image quality and quantification. The Q. Clear (BSREM) reconstruction algorithm introduces a β penalization parameter to improve image clarity, yet the optimal β setting for each frame duration remains unclear. Methods: A phantom study was conducted using a Hoffman 3D brain phantom filled with 185 MBq of ¹⁸F-FDG. List-mode PET data were acquired and divided into five frame durations: 10 s, 30 s, 2 min, 5 min, and 30 min. Images were reconstructed using Q. Clear with 20 β values (ranging from 50 to 1000). Image quality was assessed using recovery coefficient (RC%) and signal-to-noise ratio (SNR). A Bayesian optimization framework was applied to identify the optimal β under three prioritization strategies: RC%-priority, SNR-priority, and balanced. Results and Conclusion: High β values (≥950) were optimal for 10 s and 30 s frames to suppress severe noise, while mid-range values (β ≈ 650) achieved the best trade-off between anatomical clarity and quantitative stability for 2–30 min frames. Based on both visual interpretability and quantitative performance, β = 650 is recommended as a general-purpose reconstruction setting for Q. Clear in dynamic brain PET imaging, with potential for implementation in future adaptive protocols.
| Item Type: | Monograph (Project Report) |
|---|---|
| Uncontrolled Keywords: | radiation |
| Subjects: | R Medicine R Medicine > RC Internal medicine > RC71-78.7 Examination. Diagnosis Including raiography |
| Divisions: | Kampus Kesihatan (Health Campus) > Pusat Pengajian Sains Kesihatan (School of Health Sciences) > Monograph |
| Depositing User: | MUHAMMAD AKIF AIMAN AB SHUKOR |
| Date Deposited: | 15 Oct 2025 08:10 |
| Last Modified: | 18 Nov 2025 08:46 |
| URI: | http://eprints.usm.my/id/eprint/62987 |
Actions (login required)
 |
View Item |
