Abstract
Image reconstruction from low-count PET projection data is challenging because the inverse problem is ill-posed. Inspired by the kernel methods for machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by maximum likelihood or penalized likelihood image reconstruction. Computer simulation shows that the proposed approach can achieve a higher signal-to-noise ratio for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings - International Symposium on Biomedical Imaging |
| Pages | 1162-1165 |
| Number of pages | 4 |
| DOIs | |
| State | Published - 2013 |
| Event | 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 - San Francisco, CA, United States Duration: Apr 7 2013 → Apr 11 2013 |
Other
| Other | 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 |
|---|---|
| Country | United States |
| City | San Francisco, CA |
| Period | 4/7/13 → 4/11/13 |
Fingerprint
ASJC Scopus subject areas
- Biomedical Engineering
- Radiology Nuclear Medicine and imaging
Cite this
PET image reconstruction using kernel method. / Wang, Guobao; Qi, Jinyi.
Proceedings - International Symposium on Biomedical Imaging. 2013. p. 1162-1165 6556686.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - PET image reconstruction using kernel method
AU - Wang, Guobao
AU - Qi, Jinyi
PY - 2013
Y1 - 2013
N2 - Image reconstruction from low-count PET projection data is challenging because the inverse problem is ill-posed. Inspired by the kernel methods for machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by maximum likelihood or penalized likelihood image reconstruction. Computer simulation shows that the proposed approach can achieve a higher signal-to-noise ratio for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising.
AB - Image reconstruction from low-count PET projection data is challenging because the inverse problem is ill-posed. Inspired by the kernel methods for machine learning, this paper proposes a kernel based method that models PET image intensity in each pixel as a function of a set of features obtained from prior information. The kernel-based image model is incorporated into the forward model of PET projection data and the coefficients can be readily estimated by maximum likelihood or penalized likelihood image reconstruction. Computer simulation shows that the proposed approach can achieve a higher signal-to-noise ratio for dynamic PET image reconstruction than the conventional maximum likelihood method with and without post-reconstruction denoising.
UR - http://www.scopus.com/inward/record.url?scp=84881620294&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84881620294&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2013.6556686
DO - 10.1109/ISBI.2013.6556686
M3 - Conference contribution
AN - SCOPUS:84881620294
SN - 9781467364546
SP - 1162
EP - 1165
BT - Proceedings - International Symposium on Biomedical Imaging
ER -