Deep learning–based attenuation correction for whole-body PET — a multi-tracer study with 18F-FDG, 68 Ga-DOTATATE, and 18F-Fluciclovine

• Published in: European journal of nuclear medicine and molecular imaging Vol. 49; no. 9; pp. 3086 - 3097
• Main Authors: Toyonaga, Takuya, Shao, Dan, Shi, Luyao, Zhang, Jiazhen, Revilla, Enette Mae, Menard, David, Ankrah, Joseph, Hirata, Kenji, Chen, Ming-Kai, Onofrey, John A., Lu, Yihuan
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.07.2022; Springer Nature B.V
• Subjects: Advanced Image Analyses (Radiomics and Artificial Intelligence); Algorithms; Attenuation; Cardiology; Computed tomography; Deep learning; Fluorine isotopes; Image analysis; Image processing; Image reconstruction; Imaging; Machine learning; Medicine; Medicine & Public Health; Neural networks; Nuclear Medicine; Oncology; Original Article; Orthopedics; Positron emission; Positron emission tomography; Radiology; Tomography; Tracers; Tumors; Deep learning; Image synthesis; MLAA; Attenuation correction; PET
• ISSN: 1619-7070; 1619-7089; 1619-7089
• DOI: 10.1007/s00259-022-05748-2

A novel deep learning (DL)-based attenuation correction (AC) framework was applied to clinical whole-body oncology studies using 18 F-FDG, 68  Ga-DOTATATE, and 18 F-Fluciclovine. The framework used activity (λ-MLAA) and attenuation (µ-MLAA) maps estimated by the maximum likelihood reconstruction of activity and attenuation (MLAA) algorithm as inputs to a modified U-net neural network with a novel imaging physics-based loss function to learn a CT-derived attenuation map (µ-CT). Methods Clinical whole-body PET/CT datasets of 18 F-FDG ( N  = 113), 68  Ga-DOTATATE ( N  = 76), and 18 F-Fluciclovine ( N  = 90) were used to train and test tracer-specific neural networks. For each tracer, forty subjects were used to train the neural network to predict attenuation maps (µ-DL). µ-DL and µ-MLAA were compared to the gold-standard µ-CT. PET images reconstructed using the OSEM algorithm with µ-DL (OSEM DL ) and µ-MLAA (OSEM MLAA ) were compared to the CT-based reconstruction (OSEM CT ). Tumor regions of interest were segmented by two radiologists and tumor SUV and volume measures were reported, as well as evaluation using conventional image analysis metrics. Results µ-DL yielded high resolution and fine detail recovery of the attenuation map, which was superior in quality as compared to µ-MLAA in all metrics for all tracers. Using OSEM CT as the gold-standard, OSEM DL provided more accurate tumor quantification than OSEM MLAA for all three tracers, e.g., error in SUV max for OSEM MLAA vs. OSEM DL : − 3.6 ± 4.4% vs. − 1.7 ± 4.5% for 18 F-FDG ( N  = 152), − 4.3 ± 5.1% vs. 0.4 ± 2.8% for 68  Ga-DOTATATE ( N  = 70), and − 7.3 ± 2.9% vs. − 2.8 ± 2.3% for 18 F-Fluciclovine ( N  = 44). OSEM DL also yielded more accurate tumor volume measures than OSEM MLAA , i.e., − 8.4 ± 14.5% (OSEM MLAA ) vs. − 3.0 ± 15.0% for 18 F-FDG, − 14.1 ± 19.7% vs. 1.8 ± 11.6% for 68  Ga-DOTATATE, and − 15.9 ± 9.1% vs. − 6.4 ± 6.4% for 18 F-Fluciclovine. Conclusions The proposed framework provides accurate and robust attenuation correction for whole-body 18 F-FDG, 68  Ga-DOTATATE and 18 F-Fluciclovine in tumor SUV measures as well as tumor volume estimation. The proposed method provides clinically equivalent quality as compared to CT in attenuation correction for the three tracers.