Related Experiment Video
Updated: Jul 12, 2026
Creating Dynamic Images of Short-lived Dopamine Fluctuations with lp-ntPET: Dopamine Movies of Cigarette Smoking
Published on: August 6, 2013
Improving diagnostic precision in amyloid brain PET imaging through data-driven motion correction
Hye Lim Park1, Sonya Youngju Park2, Mingeon Kim3
1Division of Nuclear Medicine, Department of Radiology, Eunpyeong St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
Background:
Head motion during brain positron emission tomography (PET)/computed tomography (CT) imaging degrades image quality, resulting in reduced reading accuracy. We evaluated the performance of a head motion correction algorithm using 18F-flutemetamol (FMM) brain PET/CT images.
Methods:
FMM brain PET/CT images were retrospectively included, and PET images were reconstructed using a motion correction algorithm: (1) motion estimation through 3D time-domain signal analysis, signal smoothing, and calculation of motion-free intervals using a Merging Adjacent Clustering method; (2) estimation of 3D motion transformations using the Summing Tree Structural algorithm; and (3) calculation of the final motion-corrected images using the 3D motion transformations during the iterative reconstruction process. All conventional and motion-corrected PET images were visually reviewed by two readers. Image quality was evaluated using a 3-point scale, and the presence of amyloid deposition was interpreted as negative, positive, or equivocal. For quantitative analysis, we calculated the uptake ratio (UR) of 5 specific brain regions, with the cerebellar cortex as a reference region. The results of the conventional and motion-corrected PET images were statistically compared.
Results:
In total, 108 sets of FMM brain PET images from 108 patients (34 men and 74 women; median age, 78 years) were included. After motion correction, image quality significantly improved (p < 0.001), and there were no images of poor quality. In the visual analysis of amyloid deposition, higher interobserver agreements were observed in motion-corrected PET images for all specific regions. In the quantitative analysis, the UR difference between the conventional and motion-corrected PET images was significantly higher in the group with head motion than in the group without head motion (p = 0.016).
Conclusions:
The motion correction algorithm provided better image quality and higher interobserver agreement. Therefore, we suggest that this algorithm be adopted as a routine post-processing protocol in amyloid brain PET/CT imaging and applied to brain PET scans with other radiotracers.
Insights
A new head motion correction algorithm significantly improved image quality and accuracy in amyloid brain PET/CT scans. This method enhances diagnostic reliability for detecting amyloid deposition, suggesting its routine use.
Area of Science:
- Neuroimaging
- Radiochemistry
- Medical Physics
Background:
- Head motion during brain PET/CT imaging compromises image quality and diagnostic accuracy.
- 18F-flutemetamol (FMM) PET/CT is used for amyloid deposition detection.
- Accurate imaging is crucial for diagnosing neurodegenerative diseases.
Purpose of the Study:
- To evaluate a novel head motion correction algorithm for FMM brain PET/CT.
- To assess the algorithm's impact on image quality and amyloid detection accuracy.
- To determine the algorithm's potential for routine clinical application.
Main Methods:
- Retrospective analysis of 108 FMM brain PET/CT scans.
- Application of a 3D motion correction algorithm involving signal analysis and iterative reconstruction.
- Visual review for amyloid deposition and quantitative uptake ratio (UR) analysis.
Main Results:
- Motion correction significantly improved image quality (p < 0.001).
- Enhanced interobserver agreement in visual assessment of amyloid deposition.
- Increased quantitative differences in UR in patients with head motion.
Conclusions:
- The motion correction algorithm effectively enhances image quality and diagnostic agreement.
- The algorithm is recommended for routine post-processing in amyloid PET/CT.
- Potential applicability to other brain PET radiotracers.
