Zero-Echo-Time and Dixon Deep Pseudo-CT (ZeDD CT): Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI

• Published in: Journal of Nuclear Medicine Vol. 59; no. 5; pp. 852 - 858
• Main Authors: Leynes, Andrew P, Yang, Jaewon, Wiesinger, Florian, Kaushik, Sandeep S, Shanbhag, Dattesh D, Seo, Youngho, Hope, Thomas A, Larson, Peder E Z
• Format: Journal Article
• Language: English
• Published: United States Society of Nuclear Medicine 01.05.2018
• Subjects: Accuracy; Artificial neural networks; Attenuation; Bone lesions; Bones; Computed tomography; Error analysis; Error correction; Image acquisition; Lesions; Machine learning; Magnetic resonance imaging; Medical imaging; Neural networks; NMR; Nuclear magnetic resonance; Patients; Physics/Instrumentation; Positron emission; Tomography; multiparametric MRI; deep learning; synthetic CT; MRAC; convolutional neural networks
• ISSN: 0161-5505; 1535-5667; 2159-662X
• DOI: 10.2967/jnumed.117.198051

Accurate quantification of uptake on PET images depends on accurate attenuation correction in reconstruction. Current MR-based attenuation correction methods for body PET use a fat and water map derived from a 2-echo Dixon MRI sequence in which bone is neglected. Ultrashort-echo-time or zero-echo-time (ZTE) pulse sequences can capture bone information. We propose the use of patient-specific multiparametric MRI consisting of Dixon MRI and proton-density-weighted ZTE MRI to directly synthesize pseudo-CT images with a deep learning model: we call this method ZTE and Dixon deep pseudo-CT (ZeDD CT). Twenty-six patients were scanned using an integrated 3-T time-of-flight PET/MRI system. Helical CT images of the patients were acquired separately. A deep convolutional neural network was trained to transform ZTE and Dixon MR images into pseudo-CT images. Ten patients were used for model training, and 16 patients were used for evaluation. Bone and soft-tissue lesions were identified, and the SUV was measured. The root-mean-squared error (RMSE) was used to compare the MR-based attenuation correction with the ground-truth CT attenuation correction. In total, 30 bone lesions and 60 soft-tissue lesions were evaluated. The RMSE in PET quantification was reduced by a factor of 4 for bone lesions (10.24% for Dixon PET and 2.68% for ZeDD PET) and by a factor of 1.5 for soft-tissue lesions (6.24% for Dixon PET and 4.07% for ZeDD PET). ZeDD CT produces natural-looking and quantitatively accurate pseudo-CT images and reduces error in pelvic PET/MRI attenuation correction compared with standard methods.