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

Bendik Skarre Abrahamsen; Ingerid Skjei Knudtsen; Live Eikenes; Tone Frost Bathen; Mattijs Elschot

doi:10.3389/fonc.2023.1220009

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

Front Oncol. 2023 Aug 24:13:1220009. doi: 10.3389/fonc.2023.1220009. eCollection 2023.

Authors

Bendik Skarre Abrahamsen¹, Ingerid Skjei Knudtsen¹, Live Eikenes¹, Tone Frost Bathen^{1

2}, Mattijs Elschot^{1

2}

Affiliations

¹ Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway.
² Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway.

Abstract

Introduction: The five-class Dixon-based PET/MR attenuation correction (AC) model, which adds bone information to the four-class model by registering 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 four-class Dixon-based attenuation correction.

Methods: A convolutional neural network was trained to predict the four-class AC error map relative to CT-based attenuation correction. Dixon MR images and the four-class attenuation correction µ-map were used as input to the models. CT and PET/MR examinations for 22 patients ([¹⁸F]FDG) were used for training and validation, and 17 patients were used for testing (6 [¹⁸F]PSMA-1007 and 11 [⁶⁸Ga]Ga-PSMA-11). A quantitative analysis of PSMA uptake using voxel- and lesion-based error metrics was used to assess performance.

Results: In the voxel-based analysis, the proposed model reduced the median root mean squared percentage error from 12.1% and 8.6% for the four- and five-class Dixon-based AC methods, respectively, to 6.2%. The median absolute percentage error in the maximum standardized uptake value (SUV_max) in bone lesions improved from 20.0% and 7.0% for four- and five-class Dixon-based AC methods to 3.8%.

Conclusion: The proposed method reduces the voxel-based error and SUV_max errors in bone lesions when compared to the four- and five-class Dixon-based AC models.

Keywords: MRAC; PET/MR; artificial intelligence frontiers; attenuation correction; deep learning; prostate cancer; pseudo-CT.

Grants and funding

This work was supported by Norwegian Cancer Society and Prostatakreft-foreningen (Grant Number 215951), the Liaison Committee between the Central Norway Regional Health Authority and the Norwegian University of Science and Technology (Grant Numbers 90265300) and 180N – Norwegian Nuclear Medicine Consortium.