MRI-guided attenuation correction in torso PET/MRI: Assessment of segmentation-, atlas-, and deep learning-based approaches in the presence of outliers

被引:12
作者
Arabi, Hossein [1 ]
Zaidi, Habib [1 ,2 ,3 ,4 ]
机构
[1] Geneva Univ Hosp, Div Nucl Med & Mol Imaging, CH-1211 Geneva, Switzerland
[2] Univ Geneva, Neuroctr, Geneva, Switzerland
[3] Univ Groningen, Univ Med Ctr Groningen, Dept Nucl Med & Mol Imaging, Groningen, Netherlands
[4] Univ Southern Denmark, Dept Nucl Med, Odense, Denmark
基金
瑞士国家科学基金会;
关键词
body truncation; deep learning; metal artifact; PET; MRI; quantitative PET; CT IMAGES; GENERATION;
D O I
10.1002/mrm.29003
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Purpose We compare the performance of three commonly used MRI-guided attenuation correction approaches in torso PET/MRI, namely segmentation-, atlas-, and deep learning-based algorithms. Methods Twenty-five co-registered torso F-18-FDG PET/CT and PET/MR images were enrolled. PET attenuation maps were generated from in-phase Dixon MRI using a three-tissue class segmentation-based approach (soft-tissue, lung, and background air), voxel-wise weighting atlas-based approach, and a residual convolutional neural network. The bias in standardized uptake value (SUV) was calculated for each approach considering CT-based attenuation corrected PET images as reference. In addition to the overall performance assessment of these approaches, the primary focus of this work was on recognizing the origins of potential outliers, notably body truncation, metal-artifacts, abnormal anatomy, and small malignant lesions in the lungs. Results The deep learning approach outperformed both atlas- and segmentation-based methods resulting in less than 4% SUV bias across 25 patients compared to the segmentation-based method with up to 20% SUV bias in bony structures and the atlas-based method with 9% bias in the lung. The deep learning-based method exhibited superior performance. Yet, in case of sever truncation and metallic-artifacts in the input MRI, this approach was outperformed by the atlas-based method, exhibiting suboptimal performance in the affected regions. Conversely, for abnormal anatomies, such as a patient presenting with one lung or small malignant lesion in the lung, the deep learning algorithm exhibited promising performance compared to other methods. Conclusion The deep learning-based method provides promising outcome for synthetic CT generation from MRI. However, metal-artifact and body truncation should be specifically addressed.
引用
收藏
页码:686 / 701
页数:16
相关论文
共 49 条
[1]   Evaluation of whole-body MR to CT deformable image registration [J].
Akbarzadeh, A. ;
Gutierrez, D. ;
Baskin, A. ;
Ay, M. R. ;
Ahmadian, A. ;
Alam, N. Riahi ;
Loevblad, K. O. ;
Zaidi, H. .
JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS, 2013, 14 (04) :238-253
[2]   Deep learning-based metal artefact reduction in PET/CT imaging [J].
Arabi, Hossein ;
Zaidi, Habib .
EUROPEAN RADIOLOGY, 2021, 31 (08) :6384-6396
[3]   Applications of artificial intelligence and deep learning in molecular imaging and radiotherapy [J].
Arabi, Hossein ;
Zaidi, Habib .
EUROPEAN JOURNAL OF HYBRID IMAGING, 2020, 4 (01)
[4]   Truncation compensation and metallic dental implant artefact reduction in PET/MRI attenuation correction using deep learning-based object completion [J].
Arabi, Hossein ;
Zaidi, Habib .
PHYSICS IN MEDICINE AND BIOLOGY, 2020, 65 (19)
[5]   Deep learning-guided estimation of attenuation correction factors from time-of-flight PET emission data [J].
Arabi, Hossein ;
Zaidi, Habib .
MEDICAL IMAGE ANALYSIS, 2020, 64
[6]   Deep learning-guided joint attenuation and scatter correction in multitracer neuroimaging studies [J].
Arabi, Hossein ;
Bortolin, Karin ;
Ginovart, Nathalie ;
Garibotto, Valentina ;
Zaidi, Habib .
HUMAN BRAIN MAPPING, 2020, 41 (13) :3667-3679
[7]   Novel adversarial semantic structure deep learning for MRI-guided attenuation correction in brain PET/MRI [J].
Arabi, Hossein ;
Zeng, Guodong ;
Zheng, Guoyan ;
Zaidi, Habib .
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2019, 46 (13) :2746-2759
[8]   Comparative study of algorithms for synthetic CT generation from MRI: Consequences for MRI-guided radiation planning in the pelvic region [J].
Arabi, Hossein ;
Dowling, Jason A. ;
Burgos, Ninon ;
Han, Xiao ;
Greer, Peter B. ;
Koutsouvelis, Nikolaos ;
Zaidi, Habib .
MEDICAL PHYSICS, 2018, 45 (11) :5218-5233
[9]   Comparison of atlas-based techniques for whole-body bone segmentation [J].
Arabi, Hossein ;
Zaidi, Habib .
MEDICAL IMAGE ANALYSIS, 2017, 36 :98-112
[10]   Whole-body bone segmentation from MRI for PET/MRI attenuation correction using shape- based averaging [J].
Arabi, Hossein ;
Zaidi, Habib .
MEDICAL PHYSICS, 2016, 43 (11) :5848-5861