PET/MRI: a frontier in era of complementary hybrid imaging

Abstract

With primitive approaches, the diagnosis and therapy were operated at the cellular, molecular, or even at the genetic level. As the diagnostic techniques are more concentrated towards molecular level, multi modal imaging becomes specifically essential. Multi-modal imaging has extensive applications in clinical as well as in pre-clinical studies. Positron Emission Tomography (PET) has flourished in the field of nuclear medicine, which has motivated it to fuse with Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) for PET/CT and PET/MRI respectively. However, the challenges in PET/CT are due to the inability of simultaneous acquisition and reduced soft tissue contrast, which has led to the development of PET/MRI. Also, MRI offers the better soft tissue contrast over CT. Hence, fusion of PET and MRI results in combining structural information with functional image from PET. Yet, it has many technical challenges due to the interference between the modalities. Also, it must be resolved with various approaches for addressing the shortcomings of each system and improvise on the image quantification system. This review elaborates on the various challenges in the present PET/MRI system and the future directions of the hybrid modality. Also, the different data acquisition and analysis techniques of PET/MRI system are discussed with enhanced details on the software tools.

Keywords: Attenuation correction; Hybrid modality; Image reconstruction; Multi-modal imaging; PET/MRI.

Fig. 1

Shows the different configurations of PET/MRI (a) Sequential design, (b) Simultaneous PET insert MRI scanner and (c) Simultaneous fully integrated system

Fig. 2

Challenges in fusing PET and MRI

Fig. 3

Shows (a) MR-compatible PET detector module, (b) complete MR-compatible PET scanner with 16 detector modules, shielding material (removed on left side), and carbon fibre tube for support and (c) Axial placement of PET insert inside 7 T MRI scanner (This research was originally published in C. Catana, Y. Wu, M. S. Judenhofer, J. Qi, B. J. Pichler, and S. R. Cherry, “Simultaneous Acquisition of Multislice PET and MR Images: Initial Results with a MR-Compatible PET Scanner, (Catana et al. 2006)” J. Nucl. Med., vol. 47, no. 12, pp. 1968–1976, 2006.© by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 4

LSO-APD detector assembly: 10 × 10 LSO array (centre right) was coupled by custom-made light guide to 3 × 3 APD array (bottom right) (This research was originally published in B. J. Pichler et al., “Performance test of an LSO-APD detector in a 7-T MRI scanner for simultaneous PET/MRI, (Pichler et al. 2006)” J. Nucl. Med., vol. 47, no. 4, pp. 639–647, 2006. © by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 5

Illustration of PET/MRI integration showing the placement of the PET gantry, RF coil and shielding box (This research was originally published in K. J. Hong et al., “A prototype MR insertable brain PET using tileable GAPD arrays, (Hong et al. 2013)” Med. Phys., vol. 40,no. 4, 2013)

Fig. 6

Various techniques of attenuation correction and reconstruction algorithms

Fig. 7

Set of images in a 48-year-old patient with a liver metastasis from sigmoid cancer. High liver uptake is found both in PETAC_CT (a) and PETAC_MR (d), whereas image fusion with low-dose CT (b) and Dixon T1w (e) allow for better anatomical correlation. In low-dose CT due to its low soft tissue contrast no anatomical correlate for the liver metastasis could be found (c). Figures from (f) to (I) present the different sets created from the raw data of the Dixon sequence ((f) T1w in-phase, (g) T1w out-of-phase, (h) water-only and (i) fat-only). The complementary value of different reconstructions can be appreciated as the liver metastases are outlined with different quality in T1w in-phase (f), T1w out-of-phase (g) and fat-only (i).No correlate can be found in the water-only image (h) (This research was originally published in M. Eiber et al., “Value of a Dixon-based MR/PET attenuation correction sequence for the localization and evaluation of PET-positive lesions, (Eiber et al. 2011)” Eur. J. Nucl. Med. Mol. Imaging, vol. 38, no. 9, pp. 1691–1701, 2011.© by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 8

Transverse slices of uncorrected (a) and corrected (b) R2 maps and segmented MR (c) and CT (d) images of phantom (This research was originally published inV. Keereman, Y. Fierens, T. Broux, Y. De Deene, M. Lonneux, and S. Vandenberghe, “MRI-based attenuation correction for PET/MRI using ultrashort echo time sequences, (Catana et al. 2010)” J. Nucl. Med., vol. 51, no. 5, pp. 812–818, 2010.© by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 9

Transverse and sagittal sections through MRI and attenuation maps generated from the tissue atlas (a) and measured templates (b) (This research was originally published in I. B. Malone, R. E. Ansorge, G. B. Williams, P. J. Nestor, T. A. Carpenter, and T. D. Fryer, “Attenuation correction methods suitable for brain imaging with a PET/MRI scanner: a comparison of tissue atlas and template attenuation map approaches., (Malone et al. 2011)” J. Nucl. Med., vol. 52, no. 7, pp. 1142–9, 2011. © by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 10

Comparison of uncorrected (a), gated (b), and corrected (c) sagittal PET image slice featuring lesions A7 and A8 in patient A. Multiple lesions and inflammatory areas with increased 18F-FDG uptake in lung show enhanced delineation (This research was originally published in Würslin C, Schmidt H, Martirosian P, Brendle C, Boss A, Schwenzer NF, et al., “Respiratory motion correction in oncologic PET using T1-weighted MR imaging on a simultaneous whole-body PET/MR system, (Würslin et al. 2013)” J. Nucl. Med. Soc Nuclear Med; 2013;54:464–71. © by the Society of Nuclear Medicine and Molecular Imaging, Inc.)

Fig. 11

Comparison between visual and SPM analysis of 18F-FDG PET images. (a) Hypermetabolism in bilateral frontal lobes (solid arrow) was detected by both visual and SPM analysis. (b) Hypometabolism in left temporal lobe (solid arrow) was detected by both visual and SPM analysis. (c) Hypometabolism was found in left rolandic area (dashed arrow) by visual assessment, and hypermetabolic region was further identified in right rolandic area (solid arrow) by SPM analysis. (d) Hypometabolic region undetected by visual assessment was identified in left mesial temporal lobe (solid arrow) by SPM analysis. (This research was originally published in Y. Zhu et al., “Glucose Metabolic Profile by Visual Assessment Combined with Statistical Parametric Mapping Analysis in Pediatric Patients with Epilepsy, (Zhu et al. 2017)” J. Nucl. Med., vol. 58, no. 8, pp. 1293–1299, 2017.© by the Society of Nuclear Medicine and Molecular Imaging)