Format

Send to

Choose Destination
Eur J Nucl Med Mol Imaging. 2019 Jul 1. doi: 10.1007/s00259-019-04380-x. [Epub ahead of print]

Novel adversarial semantic structure deep learning for MRI-guided attenuation correction in brain PET/MRI.

Arabi H1, Zeng G2, Zheng G2,3, Zaidi H4,5,6,7.

Author information

1
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland.
2
Institute for Surgical Technology and Biomechanics, University of Bern, CH-3014, Bern, Switzerland.
3
School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China.
4
Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, CH-1211, Geneva 4, Switzerland. habib.zaidi@hcuge.ch.
5
Geneva University Neurocenter, Geneva University, CH-1205, Geneva, Switzerland. habib.zaidi@hcuge.ch.
6
Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center Groningen, 9700 RB, Groningen, Netherlands. habib.zaidi@hcuge.ch.
7
Department of Nuclear Medicine, University of Southern Denmark, 500, Odense, Denmark. habib.zaidi@hcuge.ch.

Abstract

OBJECTIVE:

Quantitative PET/MR imaging is challenged by the accuracy of synthetic CT (sCT) generation from MR images. Deep learning-based algorithms have recently gained momentum for a number of medical image analysis applications. In this work, a novel sCT generation algorithm based on deep learning adversarial semantic structure (DL-AdvSS) is proposed for MRI-guided attenuation correction in brain PET/MRI.

MATERIALS AND METHODS:

The proposed DL-AdvSS algorithm exploits the ASS learning framework to constrain the synthetic CT generation process to comply with the extracted structural features from CT images. The proposed technique was evaluated through comparison to an atlas-based sCT generation method (Atlas), previously developed for MRI-only or PET/MRI-guided radiation planning. Moreover, the commercial segmentation-based approach (Segm) implemented on the Philips TF PET/MRI system was included in the evaluation. Clinical brain studies of 40 patients who underwent PET/CT and MR imaging were used for the evaluation of the proposed method under a two-fold cross validation scheme.

RESULTS:

The accuracy of cortical bone extraction and CT value estimation were investigated for the three different methods. Atlas and DL-AdvSS exhibited similar cortical bone extraction accuracy resulting in a Dice coefficient of 0.78 ± 0.07 and 0.77 ± 0.07, respectively. Likewise, DL-AdvSS and Atlas techniques performed similarly in terms of CT value estimation in the cortical bone region where a mean error (ME) of less than -11 HU was obtained. The Segm approach led to a ME of -1025 HU. Furthermore, the quantitative analysis of corresponding PET images using the three approaches assuming the CT-based attenuation corrected PET (PETCTAC) as reference demonstrated comparative performance of DL-AdvSS and Atlas techniques with a mean standardized uptake value (SUV) bias less than 4% in 63 brain regions. In addition, less that 2% SUV bias was observed in the cortical bone when using Atlas and DL-AdvSS approaches. However, Segm resulted in 14.7 ± 8.9% SUV underestimation in the cortical bone.

CONCLUSION:

The proposed DL-AdvSS approach demonstrated competitive performance with respect to the state-of-the-art atlas-based technique achieving clinically tolerable errors, thus outperforming the commercial segmentation approach used in the clinic.

KEYWORDS:

Attenuation correction; Brain imaging; Deep learning; PET/MRI; Quantitative imaging

PMID:
31264170
DOI:
10.1007/s00259-019-04380-x

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center