Mingzan Zhuang1,2, Nicolas A Karakatsanis3, Rudi A J O Dierckx1, Habib Zaidi4,5,6,7. 1. Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center, 9700 RB, Groningen, The Netherlands. 2. Department of Radiation Oncology, Affiliated Hospital of Yangzhou University, Yangzhou, 225012, Jiangsu, China. 3. Division of Radiopharmaceutical Sciences, Department of Radiology, Weill Cornell Medical College, Cornell University, New York, NY, 10021, USA. 4. Department of Nuclear Medicine and Molecular Imaging, University of Groningen, University Medical Center, 9700 RB, Groningen, The Netherlands. habib.zaidi@hcuge.ch. 5. Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, 1211, Geneva, Switzerland. habib.zaidi@hcuge.ch. 6. Geneva University Neurocenter, University of Geneva, 1205, Geneva, Switzerland. habib.zaidi@hcuge.ch. 7. Department of Nuclear Medicine, University of Southern Denmark, 500, Odense, Denmark. habib.zaidi@hcuge.ch.
Abstract
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 (Ki) 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-mapCTAC) of continuous attenuation coefficient values (Acoeff). The μ-mapCTAC were then segmented into four tissue classes (μ-map4-classes), namely background (air), lung, fat, and soft tissue, where a predefined Acoeff was assigned to each class. To assess the impact of bone for AC, the bones in the μ-mapCTAC were then assigned a predefined soft tissue Acoeff (0.1 cm-1) to produce an AC μ-map without bones (μ-mapno-bones). Thereafter, both WB static SUV and dynamic PET images were reconstructed using μ-mapCTAC, μ-map4-classes, and μ-mapno-bones (PETCTAC, PET4-classes, and PETno-bones), respectively. WB indirect and direct parametric Ki 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 (ROIpeak), maximum region-of-interest (ROImax), mean region-of-interest (ROImean), and metabolic volume product (MVP) were analyzed. The Wilcoxon test was conducted to assess the difference between PET4-classes and PETno-bones against PETCTAC for all images. The same test was also adopted to compare the differences between SUV, indirect Ki, and direct Ki images for each evaluated AC method. RESULTS: No significant differences in MATV, TBR, and CNR were observed between PET4-classes and PETCTAC for either SUV or Ki images. PET4-classes significantly overestimated ROIpeak, ROImax, ROImean, as well as MVP scores compared with PETCTAC in both SUV and Ki images. SUV images exhibited the highest median relative errors for PET4-classes with respect to PETCTAC (RE4-classes): 6.91 %, 6.55 %, 5.90 %, and 6.56 % for ROIpeak, ROImax, ROImean, and MVP, respectively. On the contrary, Ki images showed slightly reduced RE4-classes (indirect 5.52 %, 5.95 %, 4.43 %, and 5.70 %, direct 6.61 %, 6.33 %, 5.53 %, and 4.96 %) for ROIpeak, ROImax, ROImean, and MVP, respectively. A higher TBR was observed on indirect and direct Ki images relative to SUV, while direct Ki images demonstrated the highest CNR. CONCLUSIONS: Four-tissue class AC may impact SUV and Ki parameter estimation but only to a limited extent, thereby suggesting that WB Patlak Ki imaging for dynamic WB PET/MRI studies is feasible. Patlak Ki imaging can enhance TBR, thereby facilitating lesion segmentation and quantification. However, patient-specific Acoeff for each tissue class should be used when possible to address the high inter-patient variability of Acoeff distributions.
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 (Ki) 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-mapCTAC) of continuous attenuation coefficient values (Acoeff). The μ-mapCTAC were then segmented into four tissue classes (μ-map4-classes), namely background (air), lung, fat, and soft tissue, where a predefined Acoeff was assigned to each class. To assess the impact of bone for AC, the bones in the μ-mapCTAC were then assigned a predefined soft tissue Acoeff (0.1 cm-1) to produce an AC μ-map without bones (μ-mapno-bones). Thereafter, both WB static SUV and dynamic PET images were reconstructed using μ-mapCTAC, μ-map4-classes, and μ-mapno-bones (PETCTAC, PET4-classes, and PETno-bones), respectively. WB indirect and direct parametric Ki 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 (ROIpeak), maximum region-of-interest (ROImax), mean region-of-interest (ROImean), and metabolic volume product (MVP) were analyzed. The Wilcoxon test was conducted to assess the difference between PET4-classes and PETno-bones against PETCTAC for all images. The same test was also adopted to compare the differences between SUV, indirect Ki, and direct Ki images for each evaluated AC method. RESULTS: No significant differences in MATV, TBR, and CNR were observed between PET4-classes and PETCTAC for either SUV or Ki images. PET4-classes significantly overestimated ROIpeak, ROImax, ROImean, as well as MVP scores compared with PETCTAC in both SUV and Ki images. SUV images exhibited the highest median relative errors for PET4-classes with respect to PETCTAC (RE4-classes): 6.91 %, 6.55 %, 5.90 %, and 6.56 % for ROIpeak, ROImax, ROImean, and MVP, respectively. On the contrary, Ki images showed slightly reduced RE4-classes (indirect 5.52 %, 5.95 %, 4.43 %, and 5.70 %, direct 6.61 %, 6.33 %, 5.53 %, and 4.96 %) for ROIpeak, ROImax, ROImean, and MVP, respectively. A higher TBR was observed on indirect and direct Ki images relative to SUV, while direct Ki images demonstrated the highest CNR. CONCLUSIONS: Four-tissue class AC may impact SUV and Ki parameter estimation but only to a limited extent, thereby suggesting that WB Patlak Ki imaging for dynamic WB PET/MRI studies is feasible. Patlak Ki imaging can enhance TBR, thereby facilitating lesion segmentation and quantification. However, patient-specific Acoeff for each tissue class should be used when possible to address the high inter-patient variability of Acoeff distributions.
Authors: Arman Rahmim; Martin A Lodge; Nicolas A Karakatsanis; Vladimir Y Panin; Yun Zhou; Alan McMillan; Steve Cho; Habib Zaidi; Michael E Casey; Richard L Wahl Journal: Eur J Nucl Med Mol Imaging Date: 2018-09-29 Impact factor: 9.236
Authors: Florent Tixier; Dennis Vriens; Catherine Cheze-Le Rest; Mathieu Hatt; Jonathan A Disselhorst; Wim J G Oyen; Lioe-Fee de Geus-Oei; Eric P Visser; Dimitris Visvikis Journal: J Nucl Med Date: 2016-03-10 Impact factor: 10.057
Authors: Nicolas A Karakatsanis; Martin A Lodge; Abdel K Tahari; Y Zhou; Richard L Wahl; Arman Rahmim Journal: Phys Med Biol Date: 2013-09-30 Impact factor: 3.609