TY - JOUR
T1 - Automatic correction of dental artifacts in PET/MRI
AU - Ladefoged, Claes N.
AU - Andersen, Flemming L.
AU - Keller, Sune
AU - Beyer, Thomas
AU - Law, Ian
AU - Højgaard, Liselotte
AU - Darkner, Sune
AU - Lauze, Francois Bernard
PY - 2015
Y1 - 2015
N2 - A challenge when using current magnetic resonance (MR)-based attenuation correction in positron emission tomography/MR imaging (PET/MRI) is that the MRIs can have a signal void around the dental fillings that is segmented as artificial air-regions in the attenuation map. For artifacts connected to the background, we propose an extension to an existing active contour algorithm to delineate the outer contour using the non-attenuation corrected PET image and the original attenuation map. We propose a combination of two different methods for differentiating the artifacts within the body from the anatomical air-regions by first using a template of artifact regions, and second, representing the artifact regions with a combination of active shape models and k-nearest-neighbors. The accuracy of the combined method has been evaluated using 25 18 F-fluorodeoxyglucose PET/MR patients. Results showed that the approach was able to correct an average of 97+/- 3% of the artifact areas.
AB - A challenge when using current magnetic resonance (MR)-based attenuation correction in positron emission tomography/MR imaging (PET/MRI) is that the MRIs can have a signal void around the dental fillings that is segmented as artificial air-regions in the attenuation map. For artifacts connected to the background, we propose an extension to an existing active contour algorithm to delineate the outer contour using the non-attenuation corrected PET image and the original attenuation map. We propose a combination of two different methods for differentiating the artifacts within the body from the anatomical air-regions by first using a template of artifact regions, and second, representing the artifact regions with a combination of active shape models and k-nearest-neighbors. The accuracy of the combined method has been evaluated using 25 18 F-fluorodeoxyglucose PET/MR patients. Results showed that the approach was able to correct an average of 97+/- 3% of the artifact areas.
U2 - 10.1117/1.JMI.2.2.024009
DO - 10.1117/1.JMI.2.2.024009
M3 - Journal article
C2 - 26158104
VL - 2
JO - Journal of Medical Imaging
JF - Journal of Medical Imaging
SN - 2329-4302
IS - 2
M1 - 024009
ER -