AUTHOR=Abrahamsen Bendik Skarre , Knudtsen Ingerid Skjei , Eikenes Live , Bathen Tone Frost , Elschot Mattijs 

TITLE=Pelvic PET/MR attenuation correction in the image space using deep learning

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1220009

DOI=10.3389/fonc.2023.1220009

ISSN=2234-943X

ABSTRACT=Introduction: The 5-class Dixon-based PET/MR attenuation correction (AC) model which adds bone information to the 4-class model by registration major bones from a bone atlas has been shown to be error-prone. In this study, we introduce a novel method of accounting for bone in pelvic PET/MR AC by directly predicting the errors in the PET image space caused by the lack of bone in 4-class Dixon-based attenuation correction. Methods: A convolutional neural network was trained to predict the 4-class AC error map relative to CT-based attenuation correction. Dixon MR images and the 4-class attenuation correction µ-map were used as input to the models. CT and PET/MR examinations for 22 patients ([ 18 F]FDG) were used for training and validation and 17 patients were used for testing (6 [ 18 F]PSMA-1007 11 [ 68 Ga]Ga-PSMA-11). A quantitative analysis of PSMA uptake using voxel-based and lesions-based error metrics was used to assess performance. Results: In the voxel-based analysis the proposed model reduced median root mean squared percentage error from 12.1% and 8.6% for the 4-and 5-class Dixon-based AC methods respectively to 6.2%. Median absolute percentage error in SUV max in bone lesions improved from 20.0% and 7.0% for 4-and 5-class Dixon-based AC methods to 3.8%. The proposed method reduces the voxel-based error and SUV max errors in bone lesions when compared to the 4-class and 5-class Dixon-based AC models.