Impact of Tissue Classification in MRI-Guided Attenuation Correction on Whole-Body Patlak PET/MRI.

Purpose: The aim of this work is to investigate the impact of tissue classification in magnetic resonance imaging (MRI)-guided positron emission tomography (PET) attenuation correction (AC) for whole-body (WB) Patlak net uptake rate constant (K) imaging in PET/MRI studies.

Procedures: WB dynamic PET/CT data were acquired for 14 patients. The CT images were utilized to generate attenuation maps (μ-map) of continuous attenuation coefficient values (A). The μ-map were then segmented into four tissue classes (μ-map), namely background (air), lung, fat, and soft tissue, where a predefined A was assigned to each class. To assess the impact of bone for AC, the bones in the μ-map were then assigned a predefined soft tissue A (0.1 cm) to produce an AC μ-map without bones (μ-map). Thereafter, both WB static SUV and dynamic PET images were reconstructed using μ-map, μ-map, and μ-map (PET PET, and PET), respectively. WB indirect and direct parametric K images were generated using Patlak graphical analysis. Malignant lesions were delineated on PET images with an automatic segmentation method that uses an active contour model (MASAC). Then, the quantitative metrics of the metabolically active tumor volume (MATV), target-to-background (TBR), contrast-to-noise ratio (CNR), peak region-of-interest (ROI), maximum region-of-interest (ROI), mean region-of-interest (ROI), and metabolic volume product (MVP) were analyzed. The Wilcoxon test was conducted to assess the difference between PET and PET against PET for all images. The same test was also adopted to compare the differences between SUV, indirect K, and direct K images for each evaluated AC method.

Results: No significant differences in MATV, TBR, and CNR were observed between PET and PET for either SUV or K images. PET significantly overestimated ROI, ROI, ROI, as well as MVP scores compared with PET in both SUV and K images. SUV images exhibited the highest median relative errors for PET with respect to PET (RE): 6.91 %, 6.55 %, 5.90 %, and 6.56 % for ROI, ROI, ROI, and MVP, respectively. On the contrary, K images showed slightly reduced RE (indirect 5.52 %, 5.95 %, 4.43 %, and 5.70 %, direct 6.61 %, 6.33 %, 5.53 %, and 4.96 %) for ROI, ROI, ROI, and MVP, respectively. A higher TBR was observed on indirect and direct K images relative to SUV, while direct K images demonstrated the highest CNR.

Conclusions: Four-tissue class AC may impact SUV and K parameter estimation but only to a limited extent, thereby suggesting that WB Patlak K imaging for dynamic WB PET/MRI studies is feasible. Patlak K imaging can enhance TBR, thereby facilitating lesion segmentation and quantification. However, patient-specific A for each tissue class should be used when possible to address the high inter-patient variability of A distributions.