Direct parametric reconstruction in dynamic PET using deep image prior and a novel parameter magnification strategy

• Published in: Computers in biology and medicine Vol. 194; p. 110487
• Main Authors: Hong, Xiaotong, Wang, Fanghu, Sun, Hao, Arabi, Hossein, Lu, Lijun
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.08.2025
• Subjects: Algorithms; Compartmental model; Deep image prior; Deep Learning; Dynamic PET; Humans; Image Processing, Computer-Assisted - methods; Internal Medicine; Multiparametric imaging; Other; Parameter magnification; Phantoms, Imaging; Positron-Emission Tomography - methods; Deep image prior; Dynamic PET; Multiparametric imaging; Compartmental model; Parameter magnification
• ISSN: 0010-4825; 1879-0534; 1879-0534
• DOI: 10.1016/j.compbiomed.2025.110487

Multiple parametric imaging in positron emission tomography (PET) is challenging due to the noisy dynamic data and the complex mapping to kinetic parameters. Although methods like direct parametric reconstruction have been proposed to improve the image quality, limitations persist, particularly for nonlinear and small-value micro-parameters (e.g., k2, k3). This study presents a novel unsupervised deep learning approach to reconstruct and improve the quality of these micro-parameters. We proposed a direct parametric image reconstruction model, DIP-PM, integrating deep image prior (DIP) with a parameter magnification (PM) strategy. The model employs a U-Net generator to predict multiple parametric images using a CT image prior, with each output channel subsequently magnified by a factor to adjust the intensity. The model was optimized with a log-likelihood loss computed between the measured projection data and forward projected data. Two tracer datasets were simulated for evaluation: 82Rb data using the 1-tissue compartment (1 TC) model and 18F-FDG data using the 2-tissue compartment (2 TC) model, with 10-fold magnification applied to the 1 TC k2 and the 2 TC k3, respectively. DIP-PM was compared to the indirect method, direct algorithm (OTEM) and the DIP method without parameter magnification (DIP-only). Performance was assessed on phantom data using peak signal-to-noise ratio (PSNR), normalized root mean square error (NRMSE) and structural similarity index (SSIM), as well as on real 18F-FDG scan from a male subject. For the 1 TC model, OTEM performed well in K1 reconstruction, but both indirect and OTEM methods showed high noise and poor performance in k2. The DIP-only method suppressed noise in k2, but failed to reconstruct fine structures in the myocardium. DIP-PM outperformed other methods with well-preserved detailed structures, particularly in k2, achieving the best metrics (PSNR: 19.00, NRMSE: 0.3002, SSIM: 0.9289). For the 2 TC model, traditional methods exhibited high noise and blurred structures in estimating all nonlinear parameters (K1, k2, k3), while DIP-based methods significantly improved image quality. DIP-PM outperformed all methods in k3 (PSNR: 21.89, NRMSE: 0.4054, SSIM: 0.8797), and consequently produced the most accurate 2 TC Ki images (PSNR: 22.74, NRMSE: 0.4897, SSIM: 0.8391). On real FDG data, DIP-PM also showed evident advantages in estimating K1, k2 and k3 while preserving myocardial structures. The results underscore the efficacy of the DIP-based direct parametric imaging in generating and improving quality of PET parametric images. This study suggests that the proposed DIP-PM method with the parameter magnification strategy can enhance the fidelity of nonlinear micro-parameter images. [Display omitted] •DIP-PM is an effective unsupervised deep learning method for reconstructing multiple parametric images in dynamic PET.•DIP-PM enhances the estimation of nonlinear and/or small-value parameters by magnifying their gradients in optimization.•We investigated DIP-PM in reconstructing nonlinear parametric images of the 1-tissue 2-tissue compartmental models.•DIP-PM achieved superior performance in preserving structural details and suppressing noise in nonlinear parametric images.•This study suggests that DIP-PM provides a promising approach for multi-latent-variable estimation challenges in inverse problems.