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

Toyonaga, Takuya; Shao, Dongguo; Shi, Luyao; Zhang, Jiazhen; Revilla, Enette Mae; Ménard, David; Ankrah, Joseph; Hirata, Kohji; Chen, Ming-Kai; Onofrey, John A.; Lu, Yihuan

doi:10.1007/s00259-022-05748-2

Cited by 12 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although direct attenuation/scatter correction in the image domain has a number of advantages, the generation of pseudo µ-maps (synthetic CT) from non-attenuation corrected images or MR images would provide an explainable AC map to verify/detect errors/drawbacks within PET attenuation and scatter correction procedures [ 20 , 66 , 72 , 73 ]. The suboptimal performance of direct AC approaches cannot be easily depicted from the resulting PET-AC images (local under/over estimation of radiotracer uptake).…”

Section: Discussionmentioning

confidence: 99%

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

Shiri

Sadr

Akhavan

et al. 2022

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

Purpose Attenuation correction and scatter compensation (AC/SC) are two main steps toward quantitative PET imaging, which remain challenging in PET-only and PET/MRI systems. These can be effectively tackled via deep learning (DL) methods. However, trustworthy, and generalizable DL models commonly require well-curated, heterogeneous, and large datasets from multiple clinical centers. At the same time, owing to legal/ethical issues and privacy concerns, forming a large collective, centralized dataset poses significant challenges. In this work, we aimed to develop a DL-based model in a multicenter setting without direct sharing of data using federated learning (FL) for AC/SC of PET images. Methods Non-attenuation/scatter corrected and CT-based attenuation/scatter corrected (CT-ASC) 18F-FDG PET images of 300 patients were enrolled in this study. The dataset consisted of 6 different centers, each with 50 patients, with scanner, image acquisition, and reconstruction protocols varying across the centers. CT-based ASC PET images served as the standard reference. All images were reviewed to include high-quality and artifact-free PET images. Both corrected and uncorrected PET images were converted to standardized uptake values (SUVs). We used a modified nested U-Net utilizing residual U-block in a U-shape architecture. We evaluated two FL models, namely sequential (FL-SQ) and parallel (FL-PL) and compared their performance with the baseline centralized (CZ) learning model wherein the data were pooled to one server, as well as center-based (CB) models where for each center the model was built and evaluated separately. Data from each center were divided to contribute to training (30 patients), validation (10 patients), and test sets (10 patients). Final evaluations and reports were performed on 60 patients (10 patients from each center). Results In terms of percent SUV absolute relative error (ARE%), both FL-SQ (CI:12.21–14.81%) and FL-PL (CI:11.82–13.84%) models demonstrated excellent agreement with the centralized framework (CI:10.32–12.00%), while FL-based algorithms improved model performance by over 11% compared to CB training strategy (CI: 22.34–26.10%). Furthermore, the Mann–Whitney test between different strategies revealed no significant differences between CZ and FL-based algorithms (p-value > 0.05) in center-categorized mode. At the same time, a significant difference was observed between the different training approaches on the overall dataset (p-value < 0.05). In addition, voxel-wise comparison, with respect to reference CT-ASC, exhibited similar performance for images predicted by CZ (R2 = 0.94), FL-SQ (R2 = 0.93), and FL-PL (R2 = 0.92), while CB model achieved a far lower coefficient of determination (R2 = 0.74). Despite the strong correlations between CZ and FL-based methods compared to reference CT-ASC, a slight underestimation of predicted voxel values was observed. Conclusion Deep learning-based models provide promising results toward quantitative PET image reconstruction. Specifically, we developed two FL models and compared their performance with center-based and centralized models. The proposed FL-based models achieved higher performance compared to center-based models, comparable with centralized models. Our work provided strong empirical evidence that the FL framework can fully benefit from the generalizability and robustness of DL models used for AC/SC in PET, while obviating the need for the direct sharing of datasets between clinical imaging centers.

show abstract

Section: Discussionmentioning

confidence: 99%

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

Shiri

Sadr

Akhavan

et al. 2022

Eur J Nucl Med Mol Imaging

View full text Add to dashboard Cite

show abstract

“…Such multi-tracer approaches have been successfully implemented to explore the diversity and spatial heterogeneity of multiple pathomechanisms in rodent models of other neurodegenerative diseases [133,134]. Like for IVM, AI algorithms can also be used in PET imaging to improve the quantitative measurement of each radiotracer concentration depending on its localization in the body [135]. Therefore, it can be envisioned that the relative densities of the most relevant cells for neuroinflammatory lesions will soon be dynamically examined by PET during the course of the pathology and at every location of the CNS instead of being limited to the superficial SC layers with IVM.…”

Section: Animal Models and Whole-body Preclinical Imaging: Developmen...mentioning

confidence: 99%

Contribution of Intravital Neuroimaging to Study Animal Models of Multiple Sclerosis

et al. 2023

View full text Add to dashboard Cite

Multiple sclerosis (MS) is a complex and long-lasting neurodegenerative disease of the central nervous system (CNS), characterized by the loss of myelin within the white matter and cortical fibers, axonopathy, and inflammatory responses leading to consequent sensory-motor and cognitive deficits of patients. While complete resolution of the disease is not yet a reality, partial tissue repair has been observed in patients which offers hope for therapeutic strategies. To address the molecular and cellular events of the pathomechanisms, a variety of animal models have been developed to investigate distinct aspects of MS disease. Recent advances of multiscale intravital imaging facilitated the direct in vivo analysis of MS in the animal models with perspective of clinical transfer to patients. This review gives an overview of MS animal models, focusing on the current imaging modalities at the microscopic and macroscopic levels and emphasizing the importance of multimodal approaches to improve our understanding of the disease and minimize the use of animals.

show abstract

“…The MLAA algorithm was compared to the maximum likelihood expectation maximization (MLEM) reconstructions with CT-based AC on 23 torso 18 F-FDG patient scans and the joint estimation results were found to be within clinical acceptable accuracy [19]. In addition, compared to the gold-CT-derived attenuation map, deep learning (DL) and convolutional neural network (CNN) were applied to predict the CT attenuation map for attenuation correction in quantitative evaluations from MLAA pre-computed results, where 279 and 100 datasets are used to provide accurate and robust attenuation correction [22,23]. Of course, there are also corresponding studies to improve the aspect of AC in PET/MR [24][25][26][27][28][29].…”

Section: Introductionmentioning

confidence: 98%

A Quantitative Clinical Evaluation of Simultaneous Reconstruction of Attenuation and Activity in Time-of-Flight PET

Zhang

Wang

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Purpose:The maximum likelihood activity and attenuation (MLAA) reconstruction algorithm has been proposed to jointly estimate tracer activity and attenuation at the same time, and proven to be a promising solution to the CT attenuation correction (CT-AC) artifacts in PET images. This study aimed to perform a quantitative evaluation of the MLAA method.Methods: A uniform cylinder phantom filled with 18F-FDG solution was scanned to optimize the reconstruction parameters for the implemented MLAA algorithm. 67 patients underwent whole-body 18F-FDG PET/CT scan were retrospectively recruited. PET images were reconstructed using MLAA and clinical standard OSEM algorithm with CT-AC (CT-OSEM). The mean and maximum standardized uptake values (SUVmean and SUVmax) in regions of interest (ROIs) of organs, high uptake lesions and metal implants affected regions were quantitatively analyzed.Results: In quantitative analysis, SUVs in patient’s organ ROIs between two methods showed R2 ranging from 0.91 to 0.98 and k ranging from 0.90 to 1.06, and the average SUVmax and SUVmean differences between two methods were within 10%, except for the lung ROI, which was 10.5% and 16.73% respectively. The average SUVmax and SUVmean differences of a total of 117 high uptake lesions were 7.25% and 7.10% respectively. 20 patients were identified to have apparent respiration motion artifacts in the liver in CT-OSEM images, and the SUVs differences between two methods measured at dome of the liver were significantly larger than measured at middle part of the liver. 10 regions with metal AC mismatch artifacts were identified in CT-OSEM images and the average SUVmean and SUVmax differences in metal implants affected regions were reported to be 52.90% and 56.20% respectively. Conclusions: PET images reconstructed using MLAA are clinically acceptable in terms of image quality as well as quantification and it is a useful tool in clinical practice, especially when CT-AC may cause artifacts.

show abstract

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

Cited by 12 publications

References 29 publications

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

Contribution of Intravital Neuroimaging to Study Animal Models of Multiple Sclerosis

A Quantitative Clinical Evaluation of Simultaneous Reconstruction of Attenuation and Activity in Time-of-Flight PET

Contact Info

Product

Resources

About