PURPOSE: We compare the quantitative accuracy of magnetic resonance imaging (MRI)-based attenuation correction (AC) using the 3-class attenuation map (PET-MRAC3c) implemented on the Ingenuity TF PET/MRI and the 4-class attenuation map (PET-MRAC4c) similar to the approach used on the Siemens mMR PET/MR considering CT-based attenuation-corrected PET images (PET-CTAC) as standard of reference. PROCEDURES: Fourteen patients with malignant tumors underwent whole-body sequential 2-deoxy-2-[(18)F]fluoro-D-glucose ((18)F-FDG) positron emission tomography (PET)/X-ray computed tomography (CT) and PET/MR imaging. A 3-class attenuation map was obtained from segmentation of T1-weighted MR images followed by assignment of attenuation coefficients (air 0 cm(-1), lung 0.022 cm(-1), soft tissue 0.096 cm(-1)), whereas a 4-class attenuation map was derived from a MR Dixon sequence (air 0 cm(-1), lung 0.018 cm(-1), fat 0.086 cm(-1), soft tissue 0.096 cm(-1)). Additional adipose tissue class and inner body air cavities (e.g., sinus and abdomen) were also considered. Different attenuation coefficients were assigned to the lungs since the two techniques were implemented as they were proposed without any modification. Standardized uptake value (SUV)mean and SUVmax metrics were calculated for volumes of interest in various organs/tissues and malignant lesions. Well-established metrics were used for the analysis of SUVs estimated using both PET-MRAC techniques and PET-CTAC including relative error, Spearman rank correlation, and Bland and Altman analysis. RESULTS: PET-MRAC3c and PET-MRAC4c revealed significant underestimation of SUV for normal organs (-17.4 ± 8.5 and -22.0 ± 6.8%, respectively) compared to PET-CTAC. Lesions' SUV presented the same trend with larger underestimation for PET-MRAC4c (-9.2 ± 6.1%) compared to PET-MRAC3c (-3.9 ± 9.0). The different attenuation coefficients assigned to the lungs with both techniques resulted in significant positive bias on PET-MRAC3c (18.6 ± 15.3%) and low negative bias on PET-MRAC4c (-0.5 ± 13.3%). Both approaches yielded the largest differences in and near bony structures. Despite the large bias, there was good correlation between PET-MRAC3c (R = 0.97, P < 0.01) and PET-CTAC, and PET-MRAC4c (R = 0.97, P < 0.01) and PET-CTAC, respectively. CONCLUSIONS: PET-MRAC3c resulted in significant systematic positive bias in the lungs owing to the lower attenuation coefficient used and negative bias in other regions. PET-MRAC4c slightly underestimated tracer uptake in the lungs and led to even larger negative bias than PET-MRAC3c in other body regions. The presence of artifacts in the MRAC might lead to misinterpretation of clinical studies. As such, the attenuation map needs to be checked for artifacts as part of the reading procedure to avoid misinterpretation of SUV measurements.
PURPOSE: We compare the quantitative accuracy of magnetic resonance imaging (MRI)-based attenuation correction (AC) using the 3-class attenuation map (PET-MRAC3c) implemented on the Ingenuity TF PET/MRI and the 4-class attenuation map (PET-MRAC4c) similar to the approach used on the Siemens mMR PET/MR considering CT-based attenuation-corrected PET images (PET-CTAC) as standard of reference. PROCEDURES: Fourteen patients with malignant tumors underwent whole-body sequential 2-deoxy-2-[(18)F]fluoro-D-glucose ((18)F-FDG) positron emission tomography (PET)/X-ray computed tomography (CT) and PET/MR imaging. A 3-class attenuation map was obtained from segmentation of T1-weighted MR images followed by assignment of attenuation coefficients (air 0 cm(-1), lung 0.022 cm(-1), soft tissue 0.096 cm(-1)), whereas a 4-class attenuation map was derived from a MR Dixon sequence (air 0 cm(-1), lung 0.018 cm(-1), fat 0.086 cm(-1), soft tissue 0.096 cm(-1)). Additional adipose tissue class and inner body air cavities (e.g., sinus and abdomen) were also considered. Different attenuation coefficients were assigned to the lungs since the two techniques were implemented as they were proposed without any modification. Standardized uptake value (SUV)mean and SUVmax metrics were calculated for volumes of interest in various organs/tissues and malignant lesions. Well-established metrics were used for the analysis of SUVs estimated using both PET-MRAC techniques and PET-CTAC including relative error, Spearman rank correlation, and Bland and Altman analysis. RESULTS: PET-MRAC3c and PET-MRAC4c revealed significant underestimation of SUV for normal organs (-17.4 ± 8.5 and -22.0 ± 6.8%, respectively) compared to PET-CTAC. Lesions' SUV presented the same trend with larger underestimation for PET-MRAC4c (-9.2 ± 6.1%) compared to PET-MRAC3c (-3.9 ± 9.0). The different attenuation coefficients assigned to the lungs with both techniques resulted in significant positive bias on PET-MRAC3c (18.6 ± 15.3%) and low negative bias on PET-MRAC4c (-0.5 ± 13.3%). Both approaches yielded the largest differences in and near bony structures. Despite the large bias, there was good correlation between PET-MRAC3c (R = 0.97, P < 0.01) and PET-CTAC, and PET-MRAC4c (R = 0.97, P < 0.01) and PET-CTAC, respectively. CONCLUSIONS: PET-MRAC3c resulted in significant systematic positive bias in the lungs owing to the lower attenuation coefficient used and negative bias in other regions. PET-MRAC4c slightly underestimated tracer uptake in the lungs and led to even larger negative bias than PET-MRAC3c in other body regions. The presence of artifacts in the MRAC might lead to misinterpretation of clinical studies. As such, the attenuation map needs to be checked for artifacts as part of the reading procedure to avoid misinterpretation of SUV measurements.
Authors: K Soejima; K Yamaguchi; E Kohda; K Takeshita; Y Ito; H Mastubara; T Oguma; T Inoue; Y Okubo; K Amakawa; H Tateno; T Shiomi Journal: Am J Respir Crit Care Med Date: 2000-04 Impact factor: 21.405
Authors: André Salomon; Andreas Goedicke; Bernd Schweizer; Til Aach; Volkmar Schulz Journal: IEEE Trans Med Imaging Date: 2010-11-29 Impact factor: 10.048
Authors: Katherine A Zukotynski; Frederic H Fahey; Mehmet Kocak; Abass Alavi; Terence Z Wong; S Ted Treves; Barry L Shulkin; Daphne A Haas-Kogan; Jeffrey R Geyer; Sridhar Vajapeyam; James M Boyett; Larry E Kun; Tina Young Poussaint Journal: J Nucl Med Date: 2011-01-13 Impact factor: 10.057
Authors: Ciprian Catana; Andre van der Kouwe; Thomas Benner; Christian J Michel; Michael Hamm; Matthias Fenchel; Bruce Fischl; Bruce Rosen; Matthias Schmand; A Gregory Sorensen Journal: J Nucl Med Date: 2010-09 Impact factor: 10.057
Authors: Gaspar Delso; Sebastian Fürst; Björn Jakoby; Ralf Ladebeck; Carl Ganter; Stephan G Nekolla; Markus Schwaiger; Sibylle I Ziegler Journal: J Nucl Med Date: 2011-11-11 Impact factor: 10.057
Authors: Pengjiang Qian; Yangyang Chen; Jung-Wen Kuo; Yu-Dong Zhang; Yizhang Jiang; Kaifa Zhao; Rose Al Helo; Harry Friel; Atallah Baydoun; Feifei Zhou; Jin Uk Heo; Norbert Avril; Karin Herrmann; Rodney Ellis; Bryan Traughber; Robert S Jones; Shitong Wang; Kuan-Hao Su; Raymond F Muzic Journal: IEEE Trans Med Imaging Date: 2019-08-16 Impact factor: 10.048
Authors: Jorge D Oldan; Shetal N Shah; Richard C Brunken; Frank P DiFilippo; Nancy A Obuchowski; Michael A Bolen Journal: J Nucl Cardiol Date: 2015-06-13 Impact factor: 5.952
Authors: Kuan-Hao Su; Lingzhi Hu; Christian Stehning; Michael Helle; Pengjiang Qian; Cheryl L Thompson; Gisele C Pereira; David W Jordan; Karin A Herrmann; Melanie Traughber; Raymond F Muzic; Bryan J Traughber Journal: Med Phys Date: 2015-08 Impact factor: 4.071
Authors: Valentina Brancato; Pasquale Borrelli; Vincenzo Alfano; Marco Picardi; Mario Mascalchi; Emanuele Nicolai; Marco Salvatore; Marco Aiello Journal: Med Phys Date: 2021-09-13 Impact factor: 4.506