MR-based attenuation correction for PET/MRI neurological studies with continuous-valued attenuation coefficients for bone through a conversion from R2* to CT-Hounsfield units

A review of PET attenuation correction methods for PET-MR

Despite being thirteen years since the installation of the first PET-MR system, the scanners constitute a very small proportion of the total hybrid PET systems installed. This is in stark contrast to the rapid expansion of the PET-CT scanner, which quickly established its importance in patient diagnosis within a similar timeframe. One of the main hurdles is the development of an accurate, reproducible and easy-to-use method for attenuation correction. Quantitative discrepancies in PET images between the manufacturer-provided MR methods and the more established CT- or transmission-based attenuation correction methods have led the scientific community in a continuous effort to develop a robust and accurate alternative. These can be divided into four broad categories: (i) MR-based, (ii) emission-based, (iii) atlas-based and the (iv) machine learning-based attenuation correction, which is rapidly gaining momentum. The first is based on segmenting the MR images in various tissues and allocating a predefined attenuation coefficient for each tissue. Emission-based attenuation correction methods aim in utilising the PET emission data by simultaneously reconstructing the radioactivity distribution and the attenuation image. Atlas-based attenuation correction methods aim to predict a CT or transmission image given an MR image of a new patient, by using databases containing CT or transmission images from the general population. Finally, in machine learning methods, a model that could predict the required image given the acquired MR or non-attenuation-corrected PET image is developed by exploiting the underlying features of the images. Deep learning methods are the dominant approach in this category. Compared to the more traditional machine learning, which uses structured data for building a model, deep learning makes direct use of the acquired images to identify underlying features. This up-to-date review goes through the literature of attenuation correction approaches in PET-MR after categorising them. The various approaches in each category are described and discussed. After exploring each category separately, a general overview is given of the current status and potential future approaches along with a comparison of the four outlined categories.

Deep-learning synthesized pseudo-CT for MR high-resolution pediatric cranial bone imaging (MR-HiPCB)

PURPOSE

CT is routinely used to detect cranial abnormalities in pediatric patients with head trauma or craniosynostosis. This study aimed to develop a deep learning method to synthesize pseudo-CT (pCT) images for MR high-resolution pediatric cranial bone imaging to eliminating ionizing radiation from CT.

METHODS

3D golden-angle stack-of-stars MRI were obtained from 44 pediatric participants. Two patch-based residual UNets were trained using paired MR and CT patches randomly selected from the whole head (NetWH) or in the vicinity of bone, fractures/sutures, or air (NetBA) to synthesize pCT. A third residual UNet was trained to generate a binary brain mask using only MRI. The pCT images from NetWH (pCTNetWH ) in the brain area and NetBA (pCTNetBA ) in the nonbrain area were combined to generate pCTCom . A manual processing method using inverted MR images was also employed for comparison.

RESULTS

pCTCom (68.01 ± 14.83 HU) had significantly smaller mean absolute errors (MAEs) than pCTNetWH (82.58 ± 16.98 HU, P < 0.0001) and pCTNetBA (91.32 ± 17.2 HU, P < 0.0001) in the whole head. Within cranial bone, the MAE of pCTCom (227.92 ± 46.88 HU) was significantly lower than pCTNetWH (287.85 ± 59.46 HU, P < 0.0001) but similar to pCTNetBA (230.20 ± 46.17 HU). Dice similarity coefficient of the segmented bone was significantly higher in pCTCom (0.90 ± 0.02) than in pCTNetWH (0.86 ± 0.04, P < 0.0001), pCTNetBA (0.88 ± 0.03, P < 0.0001), and inverted MR (0.71 ± 0.09, P < 0.0001). Dice similarity coefficient from pCTCom demonstrated significantly reduced age dependence than inverted MRI. Furthermore, pCTCom provided excellent suture and fracture visibility comparable to CT.

CONCLUSION

MR high-resolution pediatric cranial bone imaging may facilitate the clinical translation of a radiation-free MR cranial bone imaging method for pediatric patients.

Multimodal image synthesis based on disentanglement representations of anatomical and modality specific features, learned using uncooperative relativistic GAN

Growing number of methods for attenuation-coefficient map estimation from magnetic resonance (MR) images have recently been proposed because of the increasing interest in MR-guided radiotherapy and the introduction of positron emission tomography (PET) MR hybrid systems. We propose a deep-network ensemble incorporating stochastic-binary-anatomical encoders and imaging-modality variational autoencoders, to disentangle image-latent spaces into a space of modality-invariant anatomical features and spaces of modality attributes. The ensemble integrates modality-modulated decoders to normalize features and image intensities based on imaging modality. Besides promoting disentanglement, the architecture fosters uncooperative learning, offering ability to maintain anatomical structure in a cross-modality reconstruction. Introduction of a modality-invariant structural consistency constraint further enforces faithful embedding of anatomy. To improve training stability and fidelity of synthesized modalities, the ensemble is trained in a relativistic generative adversarial framework incorporating multiscale discriminators. Analyses of priors and network architectures as well as performance validation were performed on computed tomography (CT) and MR pelvis datasets. The proposed method demonstrated robustness against intensity inhomogeneity, improved tissue-class differentiation, and offered synthetic CT in Hounsfield units with intensities consistent and smooth across slices compared to the state-of-the-art approaches, offering median normalized mutual information of 1.28, normalized cross correlation of 0.97, and gradient cross correlation of 0.59 over 324 images.

Attenuation Coefficient Estimation for PET/MRI With Bayesian Deep Learning Pseudo-CT and Maximum-Likelihood Estimation of Activity and Attenuation

A major remaining challenge for magnetic resonance-based attenuation correction methods (MRAC) is their susceptibility to sources of magnetic resonance imaging (MRI) artifacts (e.g., implants and motion) and uncertainties due to the limitations of MRI contrast (e.g., accurate bone delineation and density, and separation of air/bone). We propose using a Bayesian deep convolutional neural network that in addition to generating an initial pseudo-CT from MR data, it also produces uncertainty estimates of the pseudo-CT to quantify the limitations of the MR data. These outputs are combined with the maximum-likelihood estimation of activity and attenuation (MLAA) reconstruction that uses the PET emission data to improve the attenuation maps. With the proposed approach uncertainty estimation and pseudo-CT prior for robust MLAA (UpCT-MLAA), we demonstrate accurate estimation of PET uptake in pelvic lesions and show recovery of metal implants. In patients without implants, UpCT-MLAA had acceptable but slightly higher root-mean-squared-error (RMSE) than Zero-echotime and Dixon Deep pseudo-CT when compared to CTAC. In patients with metal implants, MLAA recovered the metal implant; however, anatomy outside the implant region was obscured by noise and crosstalk artifacts. Attenuation coefficients from the pseudo-CT from Dixon MRI were accurate in normal anatomy; however, the metal implant region was estimated to have attenuation coefficients of air. UpCT-MLAA estimated attenuation coefficients of metal implants alongside accurate anatomic depiction outside of implant regions.

Deep learning for Dixon MRI-based attenuation correction in PET/MRI of head and neck cancer patients

Background

Quantitative whole-body PET/MRI relies on accurate patient-specific MRI-based attenuation correction (AC) of PET, which is a non-trivial challenge, especially for the anatomically complex head and neck region. We used a deep learning model developed for dose planning in radiation oncology to derive MRI-based attenuation maps of head and neck cancer patients and evaluated its performance on PET AC.

Methods

Eleven head and neck cancer patients, referred for radiotherapy, underwent CT followed by PET/MRI with acquisition of Dixon MRI. Both scans were performed in radiotherapy position. PET AC was performed with three different patient-specific attenuation maps derived from: (1) Dixon MRI using a deep learning network (PET_Deep). (2) Dixon MRI using the vendor-provided atlas-based method (PET_Atlas). (3) CT, serving as reference (PET_CT). We analyzed the effect of the MRI-based AC methods on PET quantification by assessing the average voxelwise error within the entire body, and the error as a function of distance to bone/air. The error in mean uptake within anatomical regions of interest and the tumor was also assessed.

Results

The average (± standard deviation) PET voxel error was 0.0 ± 11.4% for PET_Deep and −1.3 ± 21.8% for PET_Atlas. The error in mean PET uptake in bone/air was much lower for PET_Deep (−4%/12%) than for PET_Atlas (−15%/84%) and PET_Deep also demonstrated a more rapidly decreasing error with distance to bone/air affecting only the immediate surroundings (less than 1 cm). The regions with the largest error in mean uptake were those containing bone (mandible) and air (larynx) for both methods, and the error in tumor mean uptake was −0.6 ± 2.0% for PET_Deep and −3.5 ± 4.6% for PET_Atlas.

Conclusion

The deep learning network for deriving MRI-based attenuation maps of head and neck cancer patients demonstrated accurate AC and exceeded the performance of the vendor-provided atlas-based method both overall, on a lesion-level, and in vicinity of challenging regions such as bone and air.

MRI classification using semantic random forest with auto-context model

BACKGROUND

It is challenging to differentiate air and bone on MR images of conventional sequences due to their low contrast. We propose to combine semantic feature extraction under auto-context manner into random forest to improve reasonability of the MRI segmentation for MRI-based radiotherapy treatment planning or PET attention correction.

METHODS

We applied a semantic classification random forest (SCRF) method which consists of a training stage and a segmentation stage. In the training stage, patch-based MRI features were extracted from registered MRI-CT training images, and the most informative elements were selected via feature selection to train an initial random forest. The rest sequence of random forests was trained by a combination of MRI feature and semantic feature under an auto-context manner. During segmentation, the MRI patches were first fed into these random forests to derive patch-based segmentation. By using patch fusion, the final end-to-end segmentation was obtained.

RESULTS

The Dice similarity coefficient (DSC) for air, bone and soft tissue classes obtained via proposed method were 0.976±0.007, 0.819±0.050 and 0.932±0.031, compared to 0.916±0.099, 0.673±0.151 and 0.830±0.083 with random forest (RF), and 0.942±0.086, 0.791±0.046 and 0.917±0.033 with U-Net. SCRF also outperformed the competing methods in sensitivity and specificity for all three structure types.

CONCLUSIONS

The proposed method accurately segmented bone, air and soft tissue. It is promising in facilitating advanced MR application in diagnosis and therapy.

Deep learning-based T1-enhanced selection of linear attenuation coefficients (DL-TESLA) for PET/MR attenuation correction in dementia neuroimaging

PURPOSE

The accuracy of existing PET/MR attenuation correction (AC) has been limited by a lack of correlation between MR signal and tissue electron density. Based on our finding that longitudinal relaxation rate, or R1 , is associated with CT Hounsfield unit in bone and soft tissues in the brain, we propose a deep learning T1 -enhanced selection of linear attenuation coefficients (DL-TESLA) method to incorporate quantitative R1 for PET/MR AC and evaluate its accuracy and longitudinal test-retest repeatability in brain PET/MR imaging.

METHODS

DL-TESLA uses a 3D residual UNet (ResUNet) for pseudo-CT (pCT) estimation. With a total of 174 participants, we compared PET AC accuracy of DL-TESLA to 3 other methods adopting similar 3D ResUNet structures but using UTE R 2 ∗ , or Dixon, or T1 -MPRAGE as input. With images from 23 additional participants repeatedly scanned, the test-retest differences and within-subject coefficient of variation of standardized uptake value ratios (SUVR) were compared between PET images reconstructed using either DL-TESLA or CT for AC.

RESULTS

DL-TESLA had (1) significantly lower mean absolute error in pCT, (2) the highest Dice coefficients in both bone and air, (3) significantly lower PET relative absolute error in whole brain and various brain regions, (4) the highest percentage of voxels with a PET relative error within both ±3% and ±5%, (5) similar to CT test-retest differences in SUVRs from the cerebrum and mean cortical (MC) region, and (6) similar to CT within-subject coefficient of variation in cerebrum and MC.

CONCLUSION

DL-TESLA demonstrates excellent PET/MR AC accuracy and test-retest repeatability.

Photoacoustic computed tomography for functional human brain imaging [Invited]

The successes of magnetic resonance imaging and modern optical imaging of human brain function have stimulated the development of complementary modalities that offer molecular specificity, fine spatiotemporal resolution, and sufficient penetration simultaneously. By virtue of its rich optical contrast, acoustic resolution, and imaging depth far beyond the optical transport mean free path (∼1 mm in biological tissues), photoacoustic computed tomography (PACT) offers a promising complementary modality. In this article, PACT for functional human brain imaging is reviewed in its hardware, reconstruction algorithms, in vivo demonstration, and potential roadmap.

Attenuation correction for human PET/MRI studies

Attenuation correction has been one of the main methodological challenges in the integrated positron emission tomography and magnetic resonance imaging (PET/MRI) field. As standard transmission or computed tomography approaches are not available in integrated PET/MRI scanners, MR-based attenuation correction approaches had to be developed. Aspects that have to be considered for implementing accurate methods include the need to account for attenuation in bone tissue, normal and pathological lung and the MR hardware present in the PET field-of-view, to reduce the impact of subject motion, to minimize truncation and susceptibility artifacts, and to address issues related to the data acquisition and processing both on the PET and MRI sides. The standard MR-based attenuation correction techniques implemented by the PET/MRI equipment manufacturers and their impact on clinical and research PET data interpretation and quantification are first discussed. Next, the more advanced methods, including the latest generation deep learning-based approaches that have been proposed for further minimizing the attenuation correction related bias are described. Finally, a future perspective focused on the needed developments in the field is given.

Transcranial photoacoustic computed tomography based on a layered back-projection method

A major challenge of transcranial human brain photoacoustic computed tomography (PACT) is correcting for the acoustic aberration induced by the skull. Here, we present a modified universal back-projection (UBP) method, termed layered UBP (L-UBP), that can de-aberrate the transcranial PA signals by accommodating the skull heterogeneity into conventional UBP. In L-UBP, the acoustic medium is divided into multiple layers: the acoustic coupling fluid layer between the skull and detectors, the skull layer, and the brain tissue layer, which are assigned different acoustic properties. The transmission coefficients and wave conversion are considered at the fluid-skull and skull-tissue interfaces. Simulations of transcranial PACT using L-UBP were conducted to validate the method. Ex vivo experiments with a newly developed three-dimensional PACT system with 1-MHz center frequency demonstrated that L-UBP can substantially improve the image quality compared to conventional UBP.

MR Image-Based Attenuation Correction of Brain PET Imaging: Review of Literature on Machine Learning Approaches for Segmentation

Recent emerging hybrid technology of positron emission tomography/magnetic resonance (PET/MR) imaging has generated a great need for an accurate MR image-based PET attenuation correction. MR image segmentation, as a robust and simple method for PET attenuation correction, has been clinically adopted in commercial PET/MR scanners. The general approach in this method is to segment the MR image into different tissue types, each assigned an attenuation constant as in an X-ray CT image. Machine learning techniques such as clustering, classification and deep networks are extensively used for brain MR image segmentation. However, only limited work has been reported on using deep learning in brain PET attenuation correction. In addition, there is a lack of clinical evaluation of machine learning methods in this application. The aim of this review is to study the use of machine learning methods for MR image segmentation and its application in attenuation correction for PET brain imaging. Furthermore, challenges and future opportunities in MR image-based PET attenuation correction are discussed.

Characterization of hardware-related spatial distortions for IR-PETRA pulse sequence using a brain specific phantom

OBJECTIVE

Inversion recovery-pointwise encoding time reduction with radial acquisition (IR-PETRA) is an effective magnetic resonance (MR) pulse sequence in generating pseudo-CTs. The hardware-related spatial-distortion (HRSD) in MR images potentially deteriorates the accuracy of pseudo-CTs. Thus, we aimed at characterizing HRSD for IR-PETRA.

MATERIALS AND METHODS

gross-HRSD_overall (Euclidean-sum of gross-HRSD_i (i = x, y, z)) for IR-PETRA was assessed using a brain-specific phantom for two MR scanners (1.5 T-Aera and 3.0 T-Prisma). Moreover, hardware imperfections were analyzed by determining gradient-nonlinearity spatial-distortion (GNSD) and B₀-inhomogeneity spatial-distortion (B₀ISD) for magnetization-prepared rapid acquisition gradient-echo (MP-RAGE) which has well-known distortion characteristics.

RESULTS

In 3.0 T, maximum of gross-GNSD_overall (Euclidean-sum of gross-GNSD_i) and gross-B₀ISD for MP-RAGE was 2.77 mm and 0.57 mm, respectively. For this scanner, the mean and maximum of gross-HRSD_overall for IR-PETRA were 0.63 ± 0.38 mm and 1.91 mm, respectively. In 1.5 T, maximum of gross-GNSD_overall and gross-B₀ISD for MP-RAGE was 3.41 mm and 0.78 mm, respectively. The mean and maximum of gross-HRSD_overall for IR-PETRA were 1.02 ± 0.50 mm and 3.12 mm, respectively.

DISCUSSION

The spatial accuracy of MR images, besides being impacted by hardware performance, scanner capabilities, and imaging parameters, is mainly affected by its imaging strategy and data acquisition scheme. In 3.0 T, even without applying vendor correction algorithms, spatial accuracy of IR-PETRA image is sufficient for generating pseudo-CTs. In 1.5 T, distortion-correction is required to provide this accuracy.

MR-based PET attenuation correction using a combined ultrashort echo time/multi-echo Dixon acquisition

PURPOSE

To develop a magnetic resonance (MR)-based method for estimation of continuous linear attenuation coefficients (LACs) in positron emission tomography (PET) using a physical compartmental model and ultrashort echo time (UTE)/multi-echo Dixon (mUTE) acquisitions.

METHODS

We propose a three-dimensional (3D) mUTE sequence to acquire signals from water, fat, and short T₂ components (e.g., bones) simultaneously in a single acquisition. The proposed mUTE sequence integrates 3D UTE with multi-echo Dixon acquisitions and uses sparse radial trajectories to accelerate imaging speed. Errors in the radial k-space trajectories are measured using a special k-space trajectory mapping sequence and corrected for image reconstruction. A physical compartmental model is used to fit the measured multi-echo MR signals to obtain fractions of water, fat, and bone components for each voxel, which are then used to estimate the continuous LAC map for PET attenuation correction.

RESULTS

The performance of the proposed method was evaluated via phantom and in vivo human studies, using LACs from computed tomography (CT) as reference. Compared to Dixon- and atlas-based MRAC methods, the proposed method yielded PET images with higher correlation and similarity in relation to the reference. The relative absolute errors of PET activity values reconstructed by the proposed method were below 5% in all of the four lobes (frontal, temporal, parietal, and occipital), cerebellum, whole white matter, and gray matter regions across all subjects (n = 6).

CONCLUSIONS

The proposed mUTE method can generate subject-specific, continuous LAC map for PET attenuation correction in PET/MR.

Validation of MR-Based Attenuation Correction of a Newly Released Whole-Body Simultaneous PET/MR System

The aim of this study was to validate quantitative performance of a newly released simultaneous positron emission tomography (PET)/magnetic resonance imaging (MRI) scanner, by using MR-based attenuation correction (MRAC), both in phantom study and in patient study. PET/MRI image uniformities of a phantom under different hardware configurations were tested and compared. Thirty patients were examined with 2-deoxy-2-[¹⁸F]fluoro-D-glucose (¹⁸F-FDG) PET/computed tomography (CT) and subsequent PET/MRI. PET images from PET/MRI were corrected with MRAC (PET_MR), CT-based attenuation maps (μ-maps, PET_CT), and segmented CT μ-maps (PET_CTSeg) derived from PET/CT. Standardized uptake values (SUVs) were compared among the 3 sets of PET in main organs (bone, liver and lung) and in 52 FDG-avid lesions, including soft-tissue lesions and bone lesions. The result showed that PET imaging uniformities of PET/MRI under different configurations were good (<8.8%). The SUV differences among the 3 sets of PET varied with organs and lesion types. In detail, the mean relative differences of SUV between PET_MR and PET_CT were as follows: -18.8%, bone (SUV_mean); -8.0%, liver (SUV_mean); -12.2%, lung (SUV_mean); -18.1%, bone lesions (SUV_mean); -13.3%, bone lesions (SUV_max); -8.2%, soft-tissue lesions (SUV_mean); and -7.3%, soft-tissue lesions (SUV_max). The mean relative differences between PET_MR and PET_CTSeg were as follows: -19.0%, bone (SUV_mean); -3.5%, liver (SUV_mean); -3.3%, lung (SUV_mean); -19.3%, bone lesions (SUV_mean); -17.5%, bone lesions (SUV_max); -5.5%, soft-tissue lesions (SUV_mean); and -4.4%, soft-tissue lesions (SUV_max). The differences of SUV between PET_MR and PET_CT were larger than those between PET_MR and PET_CTSeg, in both soft tissue and soft-tissue lesions (P < 0.001), but not in bone or bone lesions. In conclusion, MRAC in the newly released PET/MR system is accurate in most tissues, with SUV deviations being generally less than 10%, compared to PET/CT. In bone, however, underestimations can be substantial, which may be partially attributed to segmentation of the MR-based μ-maps.

Evaluation of a deep learning-based pelvic synthetic CT generation technique for MRI-based prostate proton treatment planning

The purpose of this work is to validate the application of a deep learning-based method for pelvic synthetic CT (sCT) generation that can be used for prostate proton beam therapy treatment planning. We propose to integrate dense block minimization into 3D cycle-consistent generative adversarial networks (cycleGAN) framework to effectively learn the nonlinear mapping between MRI and CT pairs. A cohort of 17 patients with co-registered CT and MR pairs were used to test the deep learning-based sCT generation method by leave-one-out cross-validation. Image quality between the sCT and CT images, gamma analysis passing rate, dose-volume metrics, distal range displacement, and the individual pencil beam Bragg peak shift between sCT- and CT-based proton plans were evaluated. The average mean absolute error (MAE) was 51.32  ±  16.91 HU. The relative differences of the statistics of the PTV dose-volume histogram (DVH) metrics in between sCT and CT were generally less than 1%. Mean values of dose difference, absolute dose difference (in percent of the prescribed dose) were  -0.07%  ±  0.07% and 0.23%  ±  0.08%. Mean gamma analysis pass rate of 1 mm/1%, 2 mm/2%, 3 mm/3% criteria with 10% dose threshold were 92.39%  ±  5.97%, 97.95%  ±  2.95% and 98.97%  ±  1.62% respectively. The median, mean and standard deviation of absolute maximum range differences were 0.09 cm and 0.23  ±  0.25 cm. The median and mean Bragg peak shifts among the 17 patients were 0.09 cm and 0.18  ±  0.07 cm. The image similarity, dosimetric and distal range agreement between sCT and original CT suggests the feasibility of further development of an MRI-only workflow for prostate proton radiotherapy.

Impact of non-uniform attenuation correction in a dynamic [18F]-FDOPA brain PET/MRI study

Background

PET (positron emission tomography) biokinetic modelling relies on accurate quantitative data. One of the main corrections required in PET imaging to obtain high quantitative accuracy is tissue attenuation correction (AC). Incorrect non-uniform PET-AC may result in local bias in the emission images, and thus in relative activity distributions and time activity curves for different regions. MRI (magnetic resonance imaging)-based AC is an active area of research in PET/MRI neuroimaging, where several groups developed in the last few years different methods to calculate accurate attenuation (μ-)maps. Some AC methods have been evaluated for different PET radioisotopes and pathologies. However, AC in PET/MRI has scantly been investigated in dynamic PET studies where the aim is to get quantitative kinetic parameters, rather than semi-quantitative parameters from static PET studies. In this work, we investigated the impact of AC accuracy in PET image absolute quantification and, more importantly, in the slope of the Patlak analysis based on the simplified reference tissue model, from a dynamic [¹⁸F]-fluorodopa (FDOPA) PET/MRI study. In the study, we considered the two AC methods provided by the vendor and an in-house AC method based on the dual ultrashort time echo MRI sequence, using as reference a multi-atlas-based AC method based on a T1-weighted MRI sequence.

Results

Non-uniform bias in absolute PET quantification across the brain, from − 20% near the skull to − 10% in the central region, was observed using the two vendor’s μ-maps. The AC method developed in-house showed a − 5% and 1% bias, respectively. Our study resulted in a 5–9% overestimation of the PET kinetic parameters with the vendor-provided μ-maps, while our in-house-developed AC method showed < 2% overestimation compared to the atlas-based AC method, using the cerebellar cortex as reference region. The overestimation obtained using the occipital pole as reference region resulted in a 7–10% with the vendor-provided μ-maps, while our in-house-developed AC method showed < 6% overestimation.

Conclusions

PET kinetic analyses based on a reference region are especially sensitive to the non-uniform bias in PET quantification from AC inaccuracies in brain PET/MRI. Depending on the position of the reference region and the bias with respect to the analysed region, kinetic analyses suffer different levels of bias. Considering bone in the μ-map can potentially result in larger errors, compared to the absence of bone, when non-uniformities in PET quantification are introduced.

Flexible Prediction of CT Images From MRI Data Through Improved Neighborhood Anchored Regression for PET Attenuation Correction

Given the complicated relationship between the magnetic resonance imaging (MRI) signals and the attenuation values, the attenuation correction in hybrid positron emission tomography (PET)/MRI systems remains a challenging task. Currently, existing methods are either time-consuming or require sufficient samples to train the models. In this paper, an efficient approach for predicting pseudo computed tomography (CT) images from T1- and T2-weighted MRI data with limited data is proposed. The proposed approach uses improved neighborhood anchored regression (INAR) as a baseline method to pre-calculate projected matrices to flexibly predict the pseudo CT patches. Techniques, including the augmentation of the MR/CT dataset, learning of the nonlinear descriptors of MR images, hierarchical search for nearest neighbors, data-driven optimization, and multi-regressor ensemble, are adopted to improve the effectiveness of the proposed approach. In total, 22 healthy subjects were enrolled in the study. The pseudo CT images obtained using INAR with multi-regressor ensemble yielded mean absolute error (MAE) of 92.73 ± 14.86 HU, peak signal-to-noise ratio of 29.77 ± 1.63 dB, Pearson linear correlation coefficient of 0.82 ± 0.05, dice similarity coefficient of 0.81 ± 0.03, and the relative mean absolute error (rMAE) in PET attenuation correction of 1.30 ± 0.20% compared with true CT images. Moreover, our proposed INAR method, without any refinement strategies, can achieve considerable results with only seven subjects (MAE 106.89 ± 14.43 HU, rMAE 1.51 ± 0.21%). The experiments prove the superior performance of the proposed method over the six innovative methods. Moreover, the proposed method can rapidly generate the pseudo CT images that are suitable for PET attenuation correction.

Multiatlas Fusion with a Hybrid CT Number Correction Technique for Subject-Specific Pseudo-CT Estimation in the Context of MRI-Only Radiation Therapy

OBJECTIVE

To propose a hybrid multiatlas fusion and correction approach to estimate a pseudo-computed tomography (pCT) image from T2-weighted brain magnetic resonance (MR) images in the context of MRI-only radiotherapy.

MATERIALS AND METHODS

A set of eleven pairs of T2-weighted MR and CT brain images was included. Using leave-one-out cross-validation, atlas MR images were registered to the target MRI with multimetric, multiresolution deformable registration. The subsequent deformations were applied to the atlas CT images, producing uncorrected pCT images. Afterward, a three-dimensional hybrid CT number correction technique was used. This technique uses information about MR intensity, spatial location, and tissue label from segmented MR images with the fuzzy c-means algorithm and combines them in a weighted fashion to correct Hounsfield unit values of the uncorrected pCT images. The corrected pCT images were then fused into a final pCT image.

RESULTS

The proposed hybrid approach proved to be performant in correcting Hounsfield unit values in terms of qualitative and quantitative measures. Average correlation was 0.92 and 0.91 for the proposed approach by taking the mean and the median, respectively, compared with 0.86 for the uncorrected unfused version. Average values of dice similarity coefficient for bone were 0.68 and 0.72 for the fused corrected pCT images by taking the mean and the median, respectively, compared with 0.65 for the uncorrected unfused version indicating a significant bone estimation improvement.

CONCLUSION

A hybrid fusion and correction method is presented to estimate a pCT image from T2-weighted brain MR images.

Ultrashort echo time magnetic resonance fingerprinting (UTE-MRF) for simultaneous quantification of long and ultrashort T2 tissues

PURPOSE

To demonstrate an ultrashort echo time magnetic resonance fingerprinting (UTE-MRF) method that allows quantifying relaxation times for muscle and bone in the musculoskeletal system and generating bone enhanced images that mimic CT scans.

METHODS

A fast imaging steady-state free precession MRF sequence with half pulse excitation and half projection readout was designed to sample fast T₂ decay signals. Varying echo time (TE) of a sinusoidal pattern was applied to enhance sensitivity for tissues with short and ultrashort T₂ values. The performance of UTE-MRF was evaluated via simulations, phantom, and in vivo experiments.

RESULTS

A minimal TE of 0.05 ms was achieved. Simulations indicated the sinusoidal TE sampling increased T₂ quantification accuracy in the cortical bone and tendon but had little impact on long T₂ muscle quantifications. For the rubber phantom, the averaged relaxometries from UTE-MRF (T₁ = 162 ms and T₂ = 1.07 ms) compared well with the gold standard (T₁ = 190 ms and T 2 ∗ = 1.03 ms). For the long T₂ agarose phantom, the linear regression slope between UTE-MRF and gold standard was 1.07 (R² = 0.991) for T₁ and 1.04 (R² = 0.994) for T₂ . In vivo experiments showed the detection of the cortical bone (averaged T₂ = 1.0 ms) and Achilles tendon (averaged T₂ = 15 ms). Scalp structures from the bone enhanced image show high similarity with CT.

CONCLUSION

The UTE-MRF with sinusoidal TEs can simultaneously quantify T₁ , T₂ , proton density, and B₀ in long, short, even ultrashort T₂ musculoskeletal structures. Bone enhanced images can be achieved in the brain with UTE-MRF.

Three-dimensional ultrashort echo time imaging with tricomponent analysis for human cortical bone

PURPOSE

To investigate tricomponent analysis of human cortical bone using a multipeak fat signal model with 3D ultrashort TE Cones sequences on a clinical 3T scanner.

METHODS

Tricomponent fitting of bound water, pore water, and fat content using a multipeak fat spectra model was proposed for 3D ultrashort TE imaging of cortical bone. Three-dimensional ultrashort TE Cones acquisitions combined with tricomponent analysis were used to investigate bound and pore water T 2 ∗ and fractions, as well as fat T 2 ∗ and fraction in cortical bone. Feasibility studies were performed on 9 human cortical bone specimens with regions of interest selected from the endosteum to the periosteum in 4 circumferential regions. Microcomputed tomography studies were performed to measure bone porosity and bone mineral density for comparison and validation of the bound and pore water analyses.

RESULTS

The oscillation of the signal decay was well-fitted with the proposed tricomponent model. The sum of the pore water and fat fractions from tricomponent analysis showed a high correlation with microcomputed tomography porosity (R = 0.74, P < 0.01). Estimated bound-water fraction also demonstrated a high correlation with bone mineral density (R = 0.70, P < 0.01).

CONCLUSION

Tricomponent analysis significantly improves the estimation of bound-water and pore-water fractions in human cortical bone.

Hybrid PET-MRI in Alzheimer's Disease Research

Multiple factors, namely amyloid, tau, inflammation, metabolic, and perfusion changes, contribute to the cascade of neurodegeneration and functional decline occurring in Alzheimer's disease (AD). These molecular and cellular processes and related functional and morphological changes can be visualized in vivo by two imaging modalities, namely positron emission tomography (PET) and magnetic resonance imaging (MRI). These imaging biomarkers are now part of the diagnostic algorithm and of particular interest for patient stratification and targeted drug development.In this field the availability of hybrid PET/MR systems not only offers a comprehensive evaluation in a single imaging session, but also opens new possibilities for the integration of the two imaging information. Here, we cover the clinical protocols and practical details of FDG, amyloid, and tau PET/MR imaging as applied in our institutions.

Generation of PET Attenuation Map for Whole-Body Time-of-Flight 18F-FDG PET/MRI Using a Deep Neural Network Trained with Simultaneously Reconstructed Activity and Attenuation Maps

We propose a new deep learning-based approach to provide more accurate whole-body PET/MRI attenuation correction than is possible with the Dixon-based 4-segment method. We use activity and attenuation maps estimated using the maximum-likelihood reconstruction of activity and attenuation (MLAA) algorithm as inputs to a convolutional neural network (CNN) to learn a CT-derived attenuation map. Methods: The whole-body ¹⁸F-FDG PET/CT scan data of 100 cancer patients (38 men and 62 women; age, 57.3 ± 14.1 y) were retrospectively used for training and testing the CNN. A modified U-net was trained to predict a CT-derived μ-map (μ-CT) from the MLAA-generated activity distribution (λ-MLAA) and μ-map (μ-MLAA). We used 1.3 million patches derived from 60 patients' data for training the CNN, data of 20 others were used as a validation set to prevent overfitting, and the data of the other 20 were used as a test set for the CNN performance analysis. The attenuation maps generated using the proposed method (μ-CNN), μ-MLAA, and 4-segment method (μ-segment) were compared with the μ-CT, a ground truth. We also compared the voxelwise correlation between the activity images reconstructed using ordered-subset expectation maximization with the μ-maps, and the SUVs of primary and metastatic bone lesions obtained by drawing regions of interest on the activity images. Results: The CNN generates less noisy attenuation maps and achieves better bone identification than MLAA. The average Dice similarity coefficient for bone regions between μ-CNN and μ-CT was 0.77, which was significantly higher than that between μ-MLAA and μ-CT (0.36). Also, the CNN result showed the best pixel-by-pixel correlation with the CT-based results and remarkably reduced differences in activity maps in comparison to CT-based attenuation correction. Conclusion: The proposed deep neural network produced a more reliable attenuation map for 511-keV photons than the 4-segment method currently used in whole-body PET/MRI studies.

Deep Learning Based Attenuation Correction of PET/MRI in Pediatric Brain Tumor Patients: Evaluation in a Clinical Setting

Aim: Positron emission tomography (PET) imaging is a useful tool for assisting in correct differentiation of tumor progression from reactive changes. O-(2-18F-fluoroethyl)-L-tyrosine (FET)-PET in combination with MRI can add valuable information for clinical decision making. Acquiring FET-PET/MRI simultaneously allows for a one-stop-shop that limits the need for a second sedation or anesthesia as with PET and MRI in sequence. PET/MRI is challenged by lack of a direct measure of photon attenuation. Accepted solutions for attenuation correction (AC) might not be applicable to pediatrics. The aim of this study was to evaluate the use of the subject-specific MR-derived AC method RESOLUTE, modified to a pediatric cohort, against the performance of an MR-AC technique based on deep learning in a pediatric brain tumor cohort.

Methods: The modifications to RESOLUTE and the implementation of a deep learning method were performed using 79 pediatric patient examinations. We analyzed the 36 of these with active brain tumor area above 1 mL. We measured background (B), tumor mean and maximal activity (T_MEAN, T_MAX), biological tumor volume (BTV), and calculated the clinical metrics T_MEAN/B and T_MAX/B.

Results: Overall, we found both RESOLUTE and our DeepUTE methodologies to accurately reproduce the CT-AC clinical metrics. Regardless of age, both methods were able to obtain AC maps similar to the CT-AC, albeit with DeepUTE producing the most similar based on both quantitative metrics and visual inspection. In the patient-by-patient analysis DeepUTE was the only technique with all patients inside the predefined acceptable clinical limits. It also had a higher precision with relative %-difference to the reference CT-AC (T_MAX/B mean: -0.1%, CI: [-0.8%, 0.5%], p = 0.54) compared to RESOLUTE (T_MAX/B mean: 0.3%, CI: [-0.6%, 1.2%], p = 0.67) and DIXON-AC (T_MAX/B mean: 5.9%, CI: [4.5%, 7.4%], p < 0.0001).

Conclusion: Overall, we found DeepUTE to be the AC method that most robustly reproduced the CT-AC clinical metrics per se, closely followed by RESOLUTE modified to pediatric cohorts. The added accuracy due to better noise handling of DeepUTE, ease of use, as well as the improved runtime makes DeepUTE the method of choice for PET/MRI attenuation correction.

Accurate hybrid template-based and MR-based attenuation correction using UTE images for simultaneous PET/MR brain imaging applications

Background

Attenuation correction is one of the most crucial correction factors for accurate PET data quantitation in hybrid PET/MR scanners, and computing accurate attenuation coefficient maps from MR brain acquisitions is challenging. Here, we develop a method for accurate bone and air segmentation using MR ultrashort echo time (UTE) images.

Methods

MR UTE images from simultaneous MR and PET imaging of five healthy volunteers was used to generate a whole head, bone and air template image for inclusion into an improved MR derived attenuation correction map, and applied to PET image data for quantitative analysis. Bone, air and soft tissue were segmented based on Gaussian Mixture Models with probabilistic tissue maps as a priori information. We present results for two approaches for bone attenuation coefficient assignments: one using a constant attenuation correction value; and another using an estimated continuous attenuation value based on a calibration fit. Quantitative comparisons were performed to evaluate the accuracy of the reconstructed PET images, with respect to a reference image reconstructed with manually segmented attenuation maps.

Results

The DICE coefficient analysis for the air and bone regions in the images demonstrated improvements compared to the UTE approach, and other state-of-the-art techniques. The most accurate whole brain and regional brain analyses were obtained using constant bone attenuation coefficient values.

Conclusions

A novel attenuation correction method for PET data reconstruction is proposed. Analyses show improvements in the quantitative accuracy of the reconstructed PET images compared to other state-of-the-art AC methods for simultaneous PET/MR scanners. Further evaluation is needed with radiopharmaceuticals other than FDG, and in larger cohorts of participants.

Evaluation of zero-echo-time attenuation correction for integrated PET/MR brain imaging-comparison to head atlas and 68Ge-transmission-based attenuation correction

BACKGROUND

MRI does not offer a direct method to obtain attenuation correction maps as its predecessors (stand-alone PET and PET/CT), and bone visualisation is particularly challenging. Recently, zero-echo-time (ZTE) was suggested for MR-based attenuation correction (AC). The aim of this work was to evaluate ZTE- and atlas-AC by comparison to 68Ge-transmission scan-based AC. Nine patients underwent brain PET/MR and stand-alone PET scanning using the dopamine transporter ligand 11C-PE2I. For each of them, two AC maps were obtained from the MR images: an atlas-based, obtained from T1-weighted LAVA-FLEX imaging with cortical bone inserted using a CT-based atlas, and an AC map generated from proton-density-weighted ZTE images. Stand-alone PET 68Ge-transmission AC map was used as gold standard. PET images were reconstructed using the three AC methods and standardised uptake value (SUV) values for the striatal, limbic and cortical regions, as well as the cerebellum (VOIs) were compared. SUV ratio (SUVR) values normalised for the cerebellum were also assessed. Bias, precision and agreement were calculated; statistical significance was evaluated using Wilcoxon matched-pairs signed-rank test.

RESULTS

Both ZTE- and atlas-AC showed a similar bias of 6-8% in SUV values across the regions. Correlation coefficients with 68Ge-AC were consistently high for ZTE-AC (r 0.99 for all regions), whereas they were lower for atlas-AC, varying from 0.99 in the striatum to 0.88 in the posterior cortical regions. SUVR showed an overall bias of 2.9 and 0.5% for atlas-AC and ZTE-AC, respectively. Correlations with 68Ge-AC were higher for ZTE-AC, varying from 0.99 in the striatum to 0.96 in the limbic regions, compared to atlas-AC (0.99 striatum to 0.77 posterior cortex).

CONCLUSIONS

Absolute SUV values showed less variability for ZTE-AC than for atlas-AC when compared to 68Ge-AC, but bias was similar for both methods. This bias is largely caused by higher linear attenuation coefficients in atlas- and ZTE-AC image compared to 68Ge-images. For SUVR, bias was lower when using ZTE-AC than for atlas-AC. ZTE-AC shows to be a more robust technique than atlas-AC in terms of both intra- and inter-patient variability.

From simultaneous to synergistic MR-PET brain imaging: A review of hybrid MR-PET imaging methodologies

Simultaneous Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) scanning is a recent major development in biomedical imaging. The full integration of the PET detector ring and electronics within the MR system has been a technologically challenging design to develop but provides capacity for simultaneous imaging and the potential for new diagnostic and research capability. This article reviews state-of-the-art MR-PET hardware and software, and discusses future developments focusing on neuroimaging methodologies for MR-PET scanning. We particularly focus on the methodologies that lead to an improved synergy between MRI and PET, including optimal data acquisition, PET attenuation and motion correction, and joint image reconstruction and processing methods based on the underlying complementary and mutual information. We further review the current and potential future applications of simultaneous MR-PET in both systems neuroscience and clinical neuroimaging research. We demonstrate a simultaneous data acquisition protocol to highlight new applications of MR-PET neuroimaging research studies.

Capturing Bone Signal in MRI of Pelvis, as a Large FOV Region, Using TWIST Sequence and Generating a 5-Class Attenuation Map for Prostate PET/MRI Imaging

Purpose:

Prostate imaging is a major application of hybrid positron emission tomography/magnetic resonance imaging (PET/MRI). Currently, MRI-based attenuation correction (MRAC) for whole-body PET/MRI in which the bony structures are ignored is the main obstacle to successful implementation of the hybrid modality in the clinical work flow. Ultrashort echo time sequence captures bone signal but needs specific hardware–software and is challenging in large field of view (FOV) regions, such as pelvis. The main aims of the work are (1) to capture a part of the bone signal in pelvis using short echo time (STE) imaging based on time-resolved angiography with interleaved stochastic trajectories (TWIST) sequence and (2) to consider the bone in pelvis attenuation map (µ-map) to MRAC for PET/MRI systems.

Procedures:

Time-resolved angiography with interleaved stochastic trajectories, which is routinely used for MR angiography with high temporal and spatial resolution, was employed for fast/STE MR imaging. Data acquisition was performed in a TE of 0.88 milliseconds (STE) and 4.86 milliseconds (long echo time [LTE]) in pelvis region. Region of interest (ROI)-based analysis was used for comparing the signal-to-noise ratio (SNR) of cortical bone in STE and LTE images. A hybrid segmentation protocol, which is comprised of image subtraction, a Fuzzy-based segmentation, and a dedicated morphologic operation, was used for generating a 5-class µ-map consisting of cortical bone, air cavity, fat, soft tissue, and background (µ-map_MR-5c). A MR-based 4-class µ-map (µ-map_MR-4c) that considered soft tissue rather than bone was generated. As such, a bilinear (µ-map_CT-ref), 5 (µ-map_CT-5c), and 4 class µ-map (µ-map_CT-4c) based on computed tomography (CT) images were generated. Finally, simulated PET data were corrected using µ-map_MR-5c (PET-MRAC5c), µ-map_MR-4c (PET-MRAC4c), µ-map_CT-5c (PET-CTAC5c), and µ-map_CT-ref (PET-CTAC).

Results:

The ratio of SNR_bone to SNR_{air cavity} in LTE images was 0.8, this factor was increased to 4.4 in STE images. The Dice, Sensitivity, and Accuracy metrics for bone segmentation in proposed method were 72.4% ± 5.5%, 69.6% ± 7.5%, and 96.5% ± 3.5%, respectively, where the segmented CT served as reference. The mean relative error in bone regions in the simulated PET images were −13.98% ± 15%, −35.59% ± 15.41%, and 1.81% ± 12.2%, respectively, in PET-MRAC5c, PET-MRAC4c, and PET-CTAC5c where PET-CTAC served as the reference. Despite poor correlation in the joint histogram of µ-map_MR-4c versus µ-map_CT-5c (R² > 0.78) and PET-MRAC4c versus PET-CTAC5c (R² = 0.83), high correlations were observed in µ-map_MR-5c versus µ-map_CT-5c (R² > 0.94) and PET-MRAC5c versus PET-CTAC5c (R² > 0.96).

Conclusions:

According to the SNR_{STE, pelvic bone}, the cortical bone can be separate from air cavity in STE imaging based on TWIST sequence. The proposed method generated an MRI-based µ-map containing bone and air cavity that led to more accurate tracer uptake estimation than MRAC4c. Uptake estimation in hybrid PET/MRI can be improved by employing the proposed method.

Improving PET/MR brain quantitation with template-enhanced ZTE

PURPOSE

The impact of MR-based attenuation correction on PET quantitation accuracy is an ongoing cause of concern for advanced brain research with PET/MR. The purpose of this study was to evaluate a new, template-enhanced zero-echo-time attenuation correction method for PET/MR scanners.

METHODS

30 subjects underwent a clinically-indicated ¹⁸F-FDG-PET/CT, followed by PET/MR on a GE SIGNA PET/MR. For each patient, a 42-s zero echo time (ZTE) sequence was used to generate two attenuation maps: one with the standard ZTE segmentation-based method; and another with a modification of the method, wherein pre-registered anatomical templates and CT data were used to enhance the segmentation. CT data, was used as gold standard. Reconstructed PET images were qualified visually and quantified in 68 volumes-of-interest using a standardized brain atlas.

RESULTS

Attenuation maps were successfully generated in all cases, without manual intervention or parameter tuning. One patient was excluded from the quantitative analysis due to the presence of multiple brain metastases. The PET bias with template-enhanced ZTE attenuation correction was measured to be -0.9% ± 0.9%, compared with -1.4% ± 1.1% with regular ZTE attenuation correction. In terms of absolute bias, the new method yielded 1.1% ± 0.7%, compared with 1.6% ± 0.9% with regular ZTE. Statistically significant bias reduction was obtained in the frontal region (from -2.0% to -1.0%), temporal (from -1.2% to -0.2%), parietal (from -1.9% to -1.1%), occipital (from -2.0% to -1.1%) and insula (from -1.4% to -1.1%).

CONCLUSION

These results indicate that the co-registration of pre-recorded anatomical templates to ZTE data is feasible in clinical practice and can be effectively used to improve the performance of segmentation-based attenuation correction.

Rapid dual-echo ramped hybrid encoding MR-based attenuation correction (dRHE-MRAC) for PET/MR

PURPOSE

In this study, we propose a rapid acquisition for MR-based attenuation correction (MRAC) in positron emission tomography (PET)/MR imaging, in which an ultrashort echo time (UTE) image and an out-of-phase echo image are obtained within a single rapid scan (35 s) at high spatial resolution (1 mm3 ), which allows accurate estimation of a pseudo CT image using 4-class tissue classification (discrete bone, discrete air, continuous fat, and continuous water).

METHODS

In dual-echo ramped hybrid encoding (dRHE), a UTE echo is directly followed by a second out-of-phase echo, in which hybrid spatial encoding combining single-point imaging and 3-dimensional radial frequency encoding is used to improve the quality of both images. Two-point Dixon reconstruction is used to estimate fat- and water-separated images, and UTE images are used to estimate bone. Air and bone segmentation is improved by using multiple UTE images with an advanced hybrid-encoding scheme that allows reconstruction of multiple UTE images. To evaluate the proposed method, dRHE-MRAC PET/MR brain imaging was performed in 10 subjects. Dice coefficients and PET reconstruction errors relative to CT-based attenuation correction were compared with existing system MRAC approaches.

RESULTS

In dRHE-MRAC, the Dice coefficients for soft tissue, air, and bone were respectively 0.95 ± 0.01, 0.62 ± 0.06, and 0.78 ± 0.05, which was a significantly improved result compared with existing approaches. In most brain regions, dRHE-MRAC showed significantly reduced PET error (less than 1%) with P values less than 0.05.

CONCLUSIONS

Dual-echo ramped hybrid encoding enables rapid and robust imaging for MRAC with a very rapid acquisition. Magn Reson Med 79:2912-2922, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Technical Note: Deep learning based MRAC using rapid ultrashort echo time imaging

PURPOSE

In this study, we explore the feasibility of a novel framework for MR-based attenuation correction for PET/MR imaging based on deep learning via convolutional neural networks, which enables fully automated and robust estimation of a pseudo CT image based on ultrashort echo time (UTE), fat, and water images obtained by a rapid MR acquisition.

METHODS

MR images for MRAC are acquired using dual echo ramped hybrid encoding (dRHE), where both UTE and out-of-phase echo images are obtained within a short single acquisition (35 s). Tissue labeling of air, soft tissue, and bone in the UTE image is accomplished via a deep learning network that was pre-trained with T1-weighted MR images. UTE images are used as input to the network, which was trained using labels derived from co-registered CT images. The tissue labels estimated by deep learning are refined by a conditional random field based correction. The soft tissue labels are further separated into fat and water components using the two-point Dixon method. The estimated bone, air, fat, and water images are then assigned appropriate Hounsfield units, resulting in a pseudo CT image for PET attenuation correction. To evaluate the proposed MRAC method, PET/MR imaging of the head was performed on eight human subjects, where Dice similarity coefficients of the estimated tissue labels and relative PET errors were evaluated through comparison to a registered CT image.

RESULT

Dice coefficients for air (within the head), soft tissue, and bone labels were 0.76 ± 0.03, 0.96 ± 0.006, and 0.88 ± 0.01. In PET quantitation, the proposed MRAC method produced relative PET errors less than 1% within most brain regions.

CONCLUSION

The proposed MRAC method utilizing deep learning with transfer learning and an efficient dRHE acquisition enables reliable PET quantitation with accurate and rapid pseudo CT generation.

PET/MRI Hybrid Systems

Over the last decade, the combination of PET and MRI in one system has proven to be highly successful in basic preclinical research, as well as in clinical research. Nowadays, PET/MRI systems are well established in preclinical imaging and are progressing into clinical applications to provide further insights into specific diseases, therapeutic assessments, and biological pathways. Certain challenges in terms of hardware had to be resolved concurrently with the development of new techniques to be able to reach the full potential of both combined techniques. This review provides an overview of these challenges and describes the opportunities that simultaneous PET/MRI systems can exploit in comparison with stand-alone or other combined hybrid systems. New approaches were developed for simultaneous PET/MRI systems to correct for attenuation of 511 keV photons because MRI does not provide direct information on gamma photon attenuation properties. Furthermore, new algorithms to correct for motion were developed, because MRI can accurately detect motion with high temporal resolution. The additional information gained by the MRI can be employed to correct for partial volume effects as well. The development of new detector designs in combination with fast-decaying scintillator crystal materials enabled time-of-flight detection and incorporation in the reconstruction algorithms. Furthermore, this review lists the currently commercially available systems both for preclinical and clinical imaging and provides an overview of applications in both fields. In this regard, special emphasis has been placed on data analysis and the potential for both modalities to evolve with advanced image analysis tools, such as cluster analysis and machine learning.

Predicting CT Image From MRI Data Through Feature Matching With Learned Nonlinear Local Descriptors

Attenuation correction for positron-emission tomography (PET)/magnetic resonance (MR) hybrid imaging systems and dose planning for MR-based radiation therapy remain challenging due to insufficient high-energy photon attenuation information. We present a novel approach that uses the learned nonlinear local descriptors and feature matching to predict pseudo computed tomography (pCT) images from T1-weighted and T2-weighted magnetic resonance imaging (MRI) data. The nonlinear local descriptors are obtained by projecting the linear descriptors into the nonlinear high-dimensional space using an explicit feature map and low-rank approximation with supervised manifold regularization. The nearest neighbors of each local descriptor in the input MR images are searched in a constrained spatial range of the MR images among the training dataset. Then the pCT patches are estimated through k-nearest neighbor regression. The proposed method for pCT prediction is quantitatively analyzed on a dataset consisting of paired brain MRI and CT images from 13 subjects. Our method generates pCT images with a mean absolute error (MAE) of 75.25 ± 18.05 Hounsfield units, a peak signal-to-noise ratio of 30.87 ± 1.15 dB, a relative MAE of 1.56 ± 0.5% in PET attenuation correction, and a dose relative structure volume difference of 0.055 ± 0.107% in , as compared with true CT. The experimental results also show that our method outperforms four state-of-the-art methods.

Zero TE-based pseudo-CT image conversion in the head and its application in PET/MR attenuation correction and MR-guided radiation therapy planning

PURPOSE

To describe a method for converting Zero TE (ZTE) MR images into X-ray attenuation information in the form of pseudo-CT images and demonstrate its performance for (1) attenuation correction (AC) in PET/MR and (2) dose planning in MR-guided radiation therapy planning (RTP).

METHODS

Proton density-weighted ZTE images were acquired as input for MR-based pseudo-CT conversion, providing (1) efficient capture of short-lived bone signals, (2) flat soft-tissue contrast, and (3) fast and robust 3D MR imaging. After bias correction and normalization, the images were segmented into bone, soft-tissue, and air by means of thresholding and morphological refinements. Fixed Hounsfield replacement values were assigned for air (-1000 HU) and soft-tissue (+42 HU), whereas continuous linear mapping was used for bone.

RESULTS

The obtained ZTE-derived pseudo-CT images accurately resembled the true CT images (i.e., Dice coefficient for bone overlap of 0.73 ± 0.08 and mean absolute error of 123 ± 25 HU evaluated over the whole head, including errors from residual registration mismatches in the neck and mouth regions). The linear bone mapping accounted for bone density variations. Averaged across five patients, ZTE-based AC demonstrated a PET error of -0.04 ± 1.68% relative to CT-based AC. Similarly, for RTP assessed in eight patients, the absolute dose difference over the target volume was found to be 0.23 ± 0.42%.

CONCLUSION

The described method enables MR to pseudo-CT image conversion for the head in an accurate, robust, and fast manner without relying on anatomical prior knowledge. Potential applications include PET/MR-AC, and MR-guided RTP.

PET/MRI: Where might it replace PET/CT?

Simultaneous positron emission tomography and MRI (PET/MRI) is a technology that combines the anatomic and quantitative strengths of MR imaging with physiologic information obtained from PET. PET and computed tomography (PET/CT) performed in a single scanning session is an established technology already in widespread and accepted use worldwide. Given the higher cost and complexity of operating and interpreting the studies obtained on a PET/MRI system, there has been question as to which patients would benefit most from imaging with PET/MRI versus PET/CT. In this article, we compare PET/MRI with PET/CT, detail the applications for which PET/MRI has shown promise and discuss impediments to future adoption. It is our hope that future work will prove the benefit of PET/MRI to specific groups of patients, initially those in which PET/CT and MRI are already performed, leveraging simultaneity and allowing for greater degrees of multiparametric evaluation.

LEVEL OF EVIDENCE

5 Technical Efficacy: Stage 5 J. Magn. Reson. Imaging 2017;46:1247-1262.

Zero-Echo-Time and Dixon Deep Pseudo-CT (ZeDD CT): Direct Generation of Pseudo-CT Images for Pelvic PET/MRI Attenuation Correction Using Deep Convolutional Neural Networks with Multiparametric MRI

Accurate quantification of uptake on PET images depends on accurate attenuation correction in reconstruction. Current MR-based attenuation correction methods for body PET use a fat and water map derived from a 2-echo Dixon MRI sequence in which bone is neglected. Ultrashort-echo-time or zero-echo-time (ZTE) pulse sequences can capture bone information. We propose the use of patient-specific multiparametric MRI consisting of Dixon MRI and proton-density-weighted ZTE MRI to directly synthesize pseudo-CT images with a deep learning model: we call this method ZTE and Dixon deep pseudo-CT (ZeDD CT). Methods: Twenty-six patients were scanned using an integrated 3-T time-of-flight PET/MRI system. Helical CT images of the patients were acquired separately. A deep convolutional neural network was trained to transform ZTE and Dixon MR images into pseudo-CT images. Ten patients were used for model training, and 16 patients were used for evaluation. Bone and soft-tissue lesions were identified, and the SUV_max was measured. The root-mean-squared error (RMSE) was used to compare the MR-based attenuation correction with the ground-truth CT attenuation correction. Results: In total, 30 bone lesions and 60 soft-tissue lesions were evaluated. The RMSE in PET quantification was reduced by a factor of 4 for bone lesions (10.24% for Dixon PET and 2.68% for ZeDD PET) and by a factor of 1.5 for soft-tissue lesions (6.24% for Dixon PET and 4.07% for ZeDD PET). Conclusion: ZeDD CT produces natural-looking and quantitatively accurate pseudo-CT images and reduces error in pelvic PET/MRI attenuation correction compared with standard methods.

Subject-specific bone attenuation correction for brain PET/MR: can ZTE-MRI substitute CT scan accurately?

In brain PET/MR applications, accurate attenuation maps are required for accurate PET image quantification. An implemented attenuation correction (AC) method for brain imaging is the single-atlas approach that estimates an AC map from an averaged CT template. As an alternative, we propose to use a zero echo time (ZTE) pulse sequence to segment bone, air and soft tissue. A linear relationship between histogram normalized ZTE intensity and measured CT density in Hounsfield units ([Formula: see text]) in bone has been established thanks to a CT-MR database of 16 patients. Continuous AC maps were computed based on the segmented ZTE by setting a fixed linear attenuation coefficient (LAC) to air and soft tissue and by using the linear relationship to generate continuous μ values for the bone. Additionally, for the purpose of comparison, four other AC maps were generated: a ZTE derived AC map with a fixed LAC for the bone, an AC map based on the single-atlas approach as provided by the PET/MR manufacturer, a soft-tissue only AC map and, finally, the CT derived attenuation map used as the gold standard (CTAC). All these AC maps were used with different levels of smoothing for PET image reconstruction with and without time-of-flight (TOF). The subject-specific AC map generated by combining ZTE-based segmentation and linear scaling of the normalized ZTE signal into [Formula: see text] was found to be a good substitute for the measured CTAC map in brain PET/MR when used with a Gaussian smoothing kernel of [Formula: see text] corresponding to the PET scanner intrinsic resolution. As expected TOF reduces AC error regardless of the AC method. The continuous ZTE-AC performed better than the other alternative MR derived AC methods, reducing the quantification error between the MRAC corrected PET image and the reference CTAC corrected PET image.

Quantitative hemodynamic PET imaging using image-derived arterial input function and a PET/MR hybrid scanner

Positron emission tomography (PET) with ¹⁵O-tracers is commonly used to measure brain hemodynamic parameters such as cerebral blood flow, cerebral blood volume, and cerebral metabolic rate of oxygen. Conventionally, the absolute quantification of these parameters requires an arterial input function that is obtained invasively by sampling blood from an artery. In this work, we developed and validated an image-derived arterial input function technique that avoids the unreliable and burdensome arterial sampling procedure for full quantitative ¹⁵O-PET imaging. We then compared hemodynamic PET imaging performed on a PET/MR hybrid scanner against a conventional PET only scanner. We demonstrated the proposed imaging-based technique was able to generate brain hemodynamic parameter measurements in strong agreement with the traditional arterial sampling based approach. We also demonstrated that quantitative ¹⁵O-PET imaging can be successfully implemented on a PET/MR hybrid scanner.

Female pelvic synthetic CT generation based on joint intensity and shape analysis

Using MRI for radiotherapy treatment planning and image guidance is appealing as it provides superior soft tissue information over CT scans and avoids possible systematic errors introduced by aligning MR to CT images. This study presents a method that generates Synthetic CT (MRCT) volumes by performing probabilistic tissue classification of voxels from MRI data using a single imaging sequence (T1 Dixon). The intensity overlap between different tissues on MR images, a major challenge for voxel-based MRCT generation methods, is addressed by adding bone shape information to an intensity-based classification scheme. A simple pelvic bone shape model, built from principal component analysis of pelvis shape from 30 CT image volumes, is fitted to the MR volumes. The shape model generates a rough bone mask that excludes air and covers bone along with some surrounding soft tissues. Air regions are identified and masked out from the tissue classification process by intensity thresholding outside the bone mask. A regularization term is added to the fuzzy c-means classification scheme that constrains voxels outside the bone mask from being assigned memberships in the bone class. MRCT image volumes are generated by multiplying the probability of each voxel being represented in each class with assigned attenuation values of the corresponding class and summing the result across all classes. The MRCT images presented intensity distributions similar to CT images with a mean absolute error of 13.7 HU for muscle, 15.9 HU for fat, 49.1 HU for intra-pelvic soft tissues, 129.1 HU for marrow and 274.4 HU for bony tissues across 9 patients. Volumetric modulated arc therapy (VMAT) plans were optimized using MRCT-derived electron densities, and doses were recalculated using corresponding CT-derived density grids. Dose differences to planning target volumes were small with mean/standard deviation of 0.21/0.42 Gy for D0.5cc and 0.29/0.33 Gy for D99%. The results demonstrate the accuracy of the method and its potential in supporting MRI only radiotherapy treatment planning.

A review of substitute CT generation for MRI-only radiation therapy

Radiotherapy based on magnetic resonance imaging as the sole modality (MRI-only RT) is an area of growing scientific interest due to the increasing use of MRI for both target and normal tissue delineation and the development of MR based delivery systems. One major issue in MRI-only RT is the assignment of electron densities (ED) to MRI scans for dose calculation and a similar need for attenuation correction can be found for hybrid PET/MR systems. The ED assigned MRI scan is here named a substitute CT (sCT). In this review, we report on a collection of typical performance values for a number of main approaches encountered in the literature for sCT generation as compared to CT. A literature search in the Scopus database resulted in 254 papers which were included in this investigation. A final number of 50 contributions which fulfilled all inclusion criteria were categorized according to applied method, MRI sequence/contrast involved, number of subjects included and anatomical site investigated. The latter included brain, torso, prostate and phantoms. The contributions geometric and/or dosimetric performance metrics were also noted. The majority of studies are carried out on the brain for 5–10 patients with PET/MR applications in mind using a voxel based method. T1 weighted images are most commonly applied. The overall dosimetric agreement is in the order of 0.3–2.5%. A strict gamma criterion of 1% and 1mm has a range of passing rates from 68 to 94% while less strict criteria show pass rates > 98%. The mean absolute error (MAE) is between 80 and 200 HU for the brain and around 40 HU for the prostate. The Dice score for bone is between 0.5 and 0.95. The specificity and sensitivity is reported in the upper 80s% for both quantities and correctly classified voxels average around 84%. The review shows that a variety of promising approaches exist that seem clinical acceptable even with standard clinical MRI sequences. A consistent reference frame for method benchmarking is probably necessary to move the field further towards a widespread clinical implementation.

Attenuation Correction of PET/MR Imaging

PET/MR is a promising multimodality imaging approach. Attenuation is by far the largest correction required for quantitative PET imaging. MR-based attenuation correction have been extensively pursued, especially for brain imaging, in the past several years. In this article, we review atlas and direct imaging MR-based PET attenuation correction methods. The technical principles behind these methods are detailed and the advantages and disadvantages of these methods are discussed.

A multi-centre evaluation of eleven clinically feasible brain PET/MRI attenuation correction techniques using a large cohort of patients

AIM

To accurately quantify the radioactivity concentration measured by PET, emission data need to be corrected for photon attenuation; however, the MRI signal cannot easily be converted into attenuation values, making attenuation correction (AC) in PET/MRI challenging. In order to further improve the current vendor-implemented MR-AC methods for absolute quantification, a number of prototype methods have been proposed in the literature. These can be categorized into three types: template/atlas-based, segmentation-based, and reconstruction-based. These proposed methods in general demonstrated improvements compared to vendor-implemented AC, and many studies report deviations in PET uptake after AC of only a few percent from a gold standard CT-AC. Using a unified quantitative evaluation with identical metrics, subject cohort, and common CT-based reference, the aims of this study were to evaluate a selection of novel methods proposed in the literature, and identify the ones suitable for clinical use.

METHODS

In total, 11 AC methods were evaluated: two vendor-implemented (MR-AC_DIXON and MR-AC_UTE), five based on template/atlas information (MR-AC_SEGBONE (Koesters et al., 2016), MR-AC_ONTARIO (Anazodo et al., 2014), MR-AC_BOSTON (Izquierdo-Garcia et al., 2014), MR-AC_UCL (Burgos et al., 2014), and MR-AC_MAXPROB (Merida et al., 2015)), one based on simultaneous reconstruction of attenuation and emission (MR-AC_MLAA (Benoit et al., 2015)), and three based on image-segmentation (MR-AC_MUNICH (Cabello et al., 2015), MR-AC_CAR-RiDR (Juttukonda et al., 2015), and MR-AC_RESOLUTE (Ladefoged et al., 2015)). We selected 359 subjects who were scanned using one of the following radiotracers: [¹⁸F]FDG (210), [¹¹C]PiB (51), and [¹⁸F]florbetapir (98). The comparison to AC with a gold standard CT was performed both globally and regionally, with a special focus on robustness and outlier analysis.

RESULTS

The average performance in PET tracer uptake was within ±5% of CT for all of the proposed methods, with the average±SD global percentage bias in PET FDG uptake for each method being: MR-AC_DIXON (-11.3±3.5)%, MR-AC_UTE (-5.7±2.0)%, MR-AC_ONTARIO (-4.3±3.6)%, MR-AC_MUNICH (3.7±2.1)%, MR-AC_MLAA (-1.9±2.6)%, MR-AC_SEGBONE (-1.7±3.6)%, MR-AC_UCL (0.8±1.2)%, MR-AC_CAR-RiDR (-0.4±1.9)%, MR-AC_MAXPROB (-0.4±1.6)%, MR-AC_BOSTON (-0.3±1.8)%, and MR-AC_RESOLUTE (0.3±1.7)%, ordered by average bias. The overall best performing methods (MR-AC_BOSTON, MR-AC_MAXPROB, MR-AC_RESOLUTE and MR-AC_UCL, ordered alphabetically) showed regional average errors within ±3% of PET with CT-AC in all regions of the brain with FDG, and the same four methods, as well as MR-AC_CAR-RiDR, showed that for 95% of the patients, 95% of brain voxels had an uptake that deviated by less than 15% from the reference. Comparable performance was obtained with PiB and florbetapir.

CONCLUSIONS

All of the proposed novel methods have an average global performance within likely acceptable limits (±5% of CT-based reference), and the main difference among the methods was found in the robustness, outlier analysis, and clinical feasibility. Overall, the best performing methods were MR-ACBOSTON, MR-ACMAXPROB, MR-ACRESOLUTE and MR-ACUCL, ordered alphabetically. These methods all minimized the number of outliers, standard deviation, and average global and local error. The methods MR-ACMUNICH and MR-ACCAR-RiDR were both within acceptable quantitative limits, so these methods should be considered if processing time is a factor. The method MR-ACSEGBONE also demonstrates promising results, and performs well within the likely acceptable quantitative limits. For clinical routine scans where processing time can be a key factor, this vendor-provided solution currently outperforms most methods. With the performance of the methods presented here, it may be concluded that the challenge of improving the accuracy of MR-AC in adult brains with normal anatomy has been solved to a quantitatively acceptable degree, which is smaller than the quantification reproducibility in PET imaging.

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

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) atlas-registration 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.

MR-guided attenuation map for prostate PET-MRI: an intensity and morphologic-based segmentation approach for generating a five-class attenuation map in pelvic region

PURPOSE

Prostate imaging is one of the major application of hybrid PET/MRI systems. Inaccurate attenuation maps (µ-maps) derived by direct segmentation (SEG) in which the cortical bone is ignored and the volume of the air in cavities is underestimated is the main challenge of commercial PET/MRI systems for the quantitative analysis of the pelvic region. The present study considered the cortical bone and air cavity along with soft tissue, fat, and background air in the µ-map of the pelvic region using a method based on SEG. The proposed method uses a dedicated imaging technique that increases the contrast between regions and a hybrid segmentation method to classify MR images based on intensity and morphologic characteristics of tissues, such as symmetry and similarity of bony structures.

PROCEDURES

Ten healthy volunteers underwent MRI and ultra-low dose CT imaging. The dedicated MR imaging technique uses the short echo time (STE) based on the conventional sequencing implemented on a clinical 1.5T MRI scanner. The generation of a µ-map comprises the following steps: (1) bias field correction; (2) hybrid segmentation (HSEG), including segmenting images into clusters of cortical bone-air, soft tissue, and fat using spatial fuzzy c-means (SFCM), and separation of cortical bone and internal air cavities using morphologic characteristics; (3) the active contour approach for the separation of background air; and (4) the generation of a five-class μ-map for cortical bone, internal air cavity, soft tissue, fat tissue, and background air. Validation was done by comparison with segmented CT images.

RESULTS

The Dice and sensitivity metrics of cortical bone structures and internal air cavities were 72 ± 11 and 66 ± 13 and 73 ± 10 and 68 ± 20 %, respectively. High correlation was observed between CT and HSEG-based µ-maps (R ² > 0.99) and the corresponding sinograms (R ² > 0.98).

CONCLUSIONS

Currently, pelvis µ-maps provided by the current PET/MRI systems and the ultra-short echo time and atlas-based methods tend to be inaccurate. The proposed method acceptably generated a five-class μ-map using only one image.