Vision 20/20: Magnetic resonance imaging‐guided attenuation correction in PET/MRI: Challenges, solutions, and opportunities

Abstract

Attenuation correction is an essential component of the long chain of data correction techniques required to achieve the full potential of quantitative positron emission tomography (PET) imaging. The development of combined PET/magnetic resonance imaging (MRI) systems mandated the widespread interest in developing novel strategies for deriving accurate attenuation maps with the aim to improve the quantitative accuracy of these emerging hybrid imaging systems. The attenuation map in PET/MRI should ideally be derived from anatomical MR images; however, MRI intensities reflect proton density and relaxation time properties of biological tissues rather than their electron density and photon attenuation properties. Therefore, in contrast to PET/computed tomography, there is a lack of standardized global mapping between the intensities of MRI signal and linear attenuation coefficients at 511 keV. Moreover, in standard MRI sequences, bones and lung tissues do not produce measurable signals owing to their low proton density and short transverse relaxation times. MR images are also inevitably subject to artifacts that degrade their quality, thus compromising their applicability for the task of attenuation correction in PET/MRI. MRI-guided attenuation correction strategies can be classified in three broad categories: (i) segmentation-based approaches, (ii) atlasregistration and machine learning methods, and (iii) emission/transmission-based approaches. This paper summarizes past and current state-of-the-art developments and latest advances in PET/MRI attenuation correction. The advantages and drawbacks of each approach for addressing the challenges of MR-based attenuation correction are comprehensively described. The opportunities brought by both MRI and PET imaging modalities for deriving accurate attenuation maps and improving PET quantification will be elaborated. Future prospects and potential clinical applications of these techniques and their integration in commercial systems will also be discussed. C 2016 American Association of Physicists in Medicine. [http://dx

Comparison of atlas-based techniques for whole-body bone segmentation

“…Burgos et al (2014 ) demonstrated superior performance of atlas-based methods in CT synthesis and PET quantitative accuracy compared to a segmentation method using an UTE MRI sequence in brain imaging. Likewise, Mehranian et al (2016 ) demonstrated that atlas-based methods provide the most accurate attenuation maps compared to simultaneous activity-attenuation estimation and state-of-the-art 3-class segmentation method. In whole-body imaging, Hofmann et al (2011 ) proposed an atlas-based method combined with a pattern recognition technique, which resulted in less than 10% uptake error on average, thus outperforming standard segmentation methods in whole-body imaging.…”

“…Beside the precious anatomical information provided by CT or MRI, additional information that can be extracted from these images, such as attenuation properties of body tissues and motion information can be exploited for correction of emission data and quantitative PET image reconstruction. However, MRIguided attenuation correction in whole-body PET/MRI proved to be a challenging issue and has therefore remained an active and open research question during the last decade ( Mehranian et al, 2016 ). Commercially available PET/MR scanners employ tissue classification methods, which rely on segmentation of MR images into tissue classes and assigning uniform linear attenuation coefficients to each tissue class ( Martinez-Moller et al, 2009 ;Arabi et al, 2015 ).…”

PET/MRI: Reliability/Reproducibility of SUV Measurements

“…The bulk of quantitative PET/MRI research to date focused on addressing the challenges of MRI-guided PET attenuation correction. Three categories of MRI-guided attenuation correction techniques have emerged [66]. This includes (1) segmentation-based approaches, which segment MR images into different tissue classes and assign predefined attenuation coefficients to each class, (2) atlas-based and machine learning techniques in which co-registered MR-CT Atlas pairs are used to derive a pseudo-CT image or to learn a mapping function that predicts the pseudo-CT from actual patient's MRI, and (3) the recently revisited joint emission and attenuation reconstruction algorithms or maximum likelihood reconstruction of attenuation and activity (MLAA), in which the attenuation map is estimated from emission or transmission data.…”

“…An additional complexity arises from the attenuation and scattering of annihilation photons by objects present in the field of view, which may also induce SUV underestimation if not accounted for. This includes patient's bed, MRI radiofrequency body or surface coils, and patient positioning aids [66]. Transmission or CT scanning-based predetermination of attenuation maps for rigid objects (bed, body coils, etc.)…”

Quantitative analysis of MRI-guided attenuation correction techniques in time-of-flight brain PET/MRI

“…The quantitative analysis using clinical whole-body studies showed that the proposed MLAA-AC method resulted in − 10.2% quantification error in bony structures compared to − 18.4% induced by the 4-class MRAC method. For a more detailed survey of the strategies devised to address the challenges of AC in PET/MRI, interested readers are referred to recent reviews on the topic (Mehranian et al, 2016, Bezrukov et al, 2013.…”