Improved 2.5D-UNet pseudo-CT generation for attenuation correction in PET imaging using 18F-fluorodeoxyglucose and 11C-acetoacetate radiotracers
Abstract
Background Attenuation in PET imaging can lead to quantification errors, underestimation of lesions and misinterpretation of the extent of disease. Commonly, this correction is made using a CT image that provides additional anatomical information. Unfortunately, using CT significantly increases the radiation dose received by patients, increases the cost of imaging equipment, and can cause misregistration errors. To overcome these limitations, deep-learning based methods were proposed to generate a pseudo-CT from the non-attenuation-corrected (NAC) PET image, which can then be used to correct for attenuation. Methods In this work, an improved UNet 2.5D architecture has been developed in which three consecutive PET image slices are used to provide more contextual information to generate a single CT slice. A dataset of 189 NAC head PET scans from patients injected with F-fluorodeoxyglucose or C-acetoacetate, and the corresponding CT images were used for training. Results The generated pseudo-CT of all patients showed an overall improvement with the proposed UNet 2.5D architecture. Compared to a standard UNet 2D architecture, the soft tissue dice score increased from (0.905 ± 0.015) to (0.922 ± 0.017), along with the SSIM score increasing from (0.889 ± 0.023) to (0.921 ± 0.019). 3D forward projections comparing the CT and pseudo-CT total attenuation projection values also highlight the improved attenuation maps generated by the proposed 2.5D deep learning architecture. Conclusion The proposed 2.5D UNet, using adjacent PET image slices for added context, outperforms the 2D UNet in generating accurate pseudo-CTs from NAC PET images, enabling effective and fast attenuation correction without additional radiation or imaging.
Related
- Deep Learning Trained Without Target-Domain Labels Improves Risk Stratification Over Expert Coronary Calcium Reading on Routine Attenuation Correction CT
- Assessing effects of low-dose CT attenuation correction on quantification and image quality on a LAFOV PET system with a modified NEMA-IQ phantom.
- Myocardial Perfusion and Fluorodeoxyglucose Uptake Patterns in Cardiac Sarcoidosis Stratified by Diagnostic Category and Systemic Involvement
- PET-based molecular imaging in neurodegeneration and dementia: current and emerging radiotracers.
- CO083 | IN VITRO CORRECTION OF SEVERE FACTOR V DEFICIENCY USING A NOVEL PLASMA-DERIVED FV CONCENTRATE: A WHOLE BLOOD THROMBIN GENERATION STUDY
BibTeX
@article{stgeorges2026improved,
title = {{Improved 2.5D-UNet pseudo-CT generation for attenuation correction in PET imaging using 18F-fluorodeoxyglucose and 11C-acetoacetate radiotracers}},
author = {Alexandre St-Georges and Alexi Houle and Thomas Cenci and Jean-Sébastien Giroux and Maxime Toussaint and Christian Thibaudeau and Etienne Auger and Étienne Croteau and Stephen Cunnane and Roger Lecomte and Jean-Baptiste Michaud},
journal = {EJNMMI Physics},
year = {2026},
month = sep,
doi = {10.1186/s40658-026-00909-w},
url = {https://nantes-universite.hal.science/hal-05745975v2},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Improved 2.5D-UNet pseudo-CT generation for attenuation correction in PET imaging using 18F-fluorodeoxyglucose and 11C-acetoacetate radiotracers
AU - Alexandre St-Georges
AU - Alexi Houle
AU - Thomas Cenci
AU - Jean-Sébastien Giroux
AU - Maxime Toussaint
AU - Christian Thibaudeau
AU - Etienne Auger
AU - Étienne Croteau
AU - Stephen Cunnane
AU - Roger Lecomte
AU - Jean-Baptiste Michaud
PY - 2026
DA - 2026/09/01
JO - EJNMMI Physics
DO - 10.1186/s40658-026-00909-w
UR - https://nantes-universite.hal.science/hal-05745975v2
AB - Background Attenuation in PET imaging can lead to quantification errors, underestimation of lesions and misinterpretation of the extent of disease. Commonly, this correction is made using a CT image that provides additional anatomical information. Unfortunately, using CT significantly increases the radiation dose received by patients, increases the cost of imaging equipment, and can cause misregistration errors. To overcome these limitations, deep-learning based methods were proposed to generate a pseudo-CT from the non-attenuation-corrected (NAC) PET image, which can then be used to correct for attenuation. Methods In this work, an improved UNet 2.5D architecture has been developed in which three consecutive PET image slices are used to provide more contextual information to generate a single CT slice. A dataset of 189 NAC head PET scans from patients injected with F-fluorodeoxyglucose or C-acetoacetate, and the corresponding CT images were used for training. Results The generated pseudo-CT of all patients showed an overall improvement with the proposed UNet 2.5D architecture. Compared to a standard UNet 2D architecture, the soft tissue dice score increased from (0.905 ± 0.015) to (0.922 ± 0.017), along with the SSIM score increasing from (0.889 ± 0.023) to (0.921 ± 0.019). 3D forward projections comparing the CT and pseudo-CT total attenuation projection values also highlight the improved attenuation maps generated by the proposed 2.5D deep learning architecture. Conclusion The proposed 2.5D UNet, using adjacent PET image slices for added context, outperforms the 2D UNet in generating accurate pseudo-CTs from NAC PET images, enabling effective and fast attenuation correction without additional radiation or imaging.
SN - 2197-7364
ER -