PET Attenuation Correction Research Articles

Brain MR images segmentation using 3D CNN with features recalibration mechanism for segmented CT generation

The segmentation of MR (magnetic resonance) images is a simple approach to create Pseudo CT images which are useful for many medical imaging analysis applications. One of the main challenges of this process is the bone segmentation of brain MR images. Deep convolutional neural networks (CNNs) have been widely and efficiently applied to perform MR images segmentation. The aim of this work is to propose a novel excitation-based CNN by recalibrating the network features adaptively to enhance the bone segmentation by segmenting the brain MR images into three tissue classes: bone, soft tissue, and air. The proposed method combines two types of features excitation mechanisms namely: (1) spatial squeeze and channel excitation block (cSE) and (2) channel squeeze and spatial excitation block (sSE). The two blocks are combined sequentially and integrated seamlessly into a 3D convolutional encoder decoder network. The novelty of this work emerges in the combination of the two excitation blocks sequentially to improve the segmentation performance and reduce the model complexity. The proposed approach is evaluated through a comparison with computed tomography (CT) images as ground truth and validated with other methods in the literature that applied deep CNN approaches to perform MR image segmentation for PET attenuation correction. Brain MR and CT datasets which consist of 50 patients are used to evaluate the proposed method. The segmentation performance of the three brain classes is evaluated using precision, recall, dice similarity coefficient (DSC), and Jaccard index. The presented method improves the bone tissue segmentation compared to the baseline model and other methods in the literature where the DSC is improved from 0.6278 ± 0.0006 to 0.6437 ± 0.0006 with an improvement percentage of 2.53% for bone class. The proposed excitation-based segmentation network architecture demonstrates promising and competitive results compared with other methods in the literature and reduces the model complexity thanks to the sequential combination of the two excitation blocks.

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

BackgroundQuantitative 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.MethodsEleven 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 (PETDeep). (2) Dixon MRI using the vendor-provided atlas-based method (PETAtlas). (3) CT, serving as reference (PETCT). 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.ResultsThe average (± standard deviation) PET voxel error was 0.0 ± 11.4% for PETDeep and −1.3 ± 21.8% for PETAtlas. The error in mean PET uptake in bone/air was much lower for PETDeep (−4%/12%) than for PETAtlas (−15%/84%) and PETDeep 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 PETDeep and −3.5 ± 4.6% for PETAtlas.ConclusionThe 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.

Deep learning-based attenuation correction for whole-body PET - a multi-tracer study with 18F-FDG, 68Ga-DOTATATE, and 18F-Fluciclovine.

Clinical whole-body PET/CT datasets of 18F-FDG (N = 113), 68Ga-DOTATATE (N = 76), and 18F-Fluciclovine (N = 90) were used to train and test tracer-specific neural networks. For each tracer, forty subjects were used to train the neural network to predict attenuation maps (µ-DL). µ-DL and µ-MLAA were compared to the gold-standard µ-CT. PET images reconstructed using the OSEM algorithm with µ-DL (OSEMDL) and µ-MLAA (OSEMMLAA) were compared to the CT-based reconstruction (OSEMCT). Tumor regions of interest were segmented by two radiologists and tumor SUV and volume measures were reported, as well as evaluation using conventional image analysis metrics. µ-DL yielded high resolution and fine detail recovery of the attenuation map, which was superior in quality as compared to µ-MLAA in all metrics for all tracers. Using OSEMCT as the gold-standard, OSEMDL provided more accurate tumor quantification than OSEMMLAA for all three tracers, e.g., error in SUVmax for OSEMMLAA vs. OSEMDL: - 3.6 ± 4.4% vs. - 1.7 ± 4.5% for 18F-FDG (N = 152), - 4.3 ± 5.1% vs. 0.4 ± 2.8% for 68Ga-DOTATATE (N = 70), and - 7.3 ± 2.9% vs. - 2.8 ± 2.3% for 18F-Fluciclovine (N = 44). OSEMDL also yielded more accurate tumor volume measures than OSEMMLAA, i.e., - 8.4 ± 14.5% (OSEMMLAA) vs. - 3.0 ± 15.0% for 18F-FDG, - 14.1 ± 19.7% vs. 1.8 ± 11.6% for 68Ga-DOTATATE, and - 15.9 ± 9.1% vs. - 6.4 ± 6.4% for 18F-Fluciclovine. The proposed framework provides accurate and robust attenuation correction for whole-body 18F-FDG, 68Ga-DOTATATE and 18F-Fluciclovine in tumor SUV measures as well as tumor volume estimation. The proposed method provides clinically equivalent quality as compared to CT in attenuation correction for the three tracers.

Marker-less and calibration-less motion correction method for brain PET.

Marker-less head motion correction methods have been well-studied; however, no reports discussing potential issues in positional calibration between a PET system and an external sensor remain limited. In this study, we develop a method for positional calibration between the PET system and an external range sensor to achieve practical head motion correction. The basic concept of the developed method involves using the subject's face model as a marker not only for head motion detection but also for the system positional calibration. The face model of the subject, which can be obtained easily using the range sensor, can also be calculated from a computed tomography (CT) image of the same subject. The CT image, which is acquired separately for attenuation correction in PET, has the same coordinates as the PET image because of the appropriate matching algorithm between CT and PET images. The proposed method was implemented in the helmet-type PET and the motion correction accuracy was assessed quantitatively using a mannequin head. The phantom experiments demonstrated the performance of the developed motion correction method; high-resolution images with no trace of the applied motion were obtained as if no motion was provided. Statistical analysis supported the visual assessment results in terms of the spatial resolution, contrast recovery; uniformity, and the results implied that motion with correction slightly improved image quality compared with the motionless case. The tolerance of the developed method against potential tracking errors had a minimum 10% difference in the amplitude of the rotation angle.

Impact of Different Metal Artifact Reduction Techniques on Attenuation Correction of Normal Organs in 18F-FDG-PET/CT.

Purpose: To evaluate the impact of different metal artifact reduction algorithms on Hounsfield units (HU) and the standardized uptake value (SUV) in normal organs in patients with different metal implants. Methods: This study prospectively included 66 patients (mean age of 66.02 ± 13.1 years) with 87 different metal implants. CT image reconstructions were performed using weighted filtered back projection (WFBP) as the standard method, metal artifact reduction in image space (MARIS), and an iterative metal artifacts reduction (iMAR) algorithm for large implants. These datasets were used for PET attenuation correction. HU and SUV measurements were performed in nine predefined anatomical locations: liver, lower lung lobes, descending aorta, thoracic vertebral body, autochthonous back muscles, pectoral muscles, and internal jugular vein. Differences between HU and SUV measurements were compared using paired t-tests. The significance level was determined as p = 0.017 using Bonferroni correction. Results: No significant differences were observed between reconstructed images using iMAR and WFBP concerning HU and SUV measurements in liver (HU: p = 0.055; SUVmax: p = 0.586), lung (HU: p = 0.276; SUVmax: p = 1.0 for the right side and HU: p = 0.630; SUVmax: p = 0.109 for the left side), descending aorta (HU: p = 0.333; SUVmax: p = 0.083), thoracic vertebral body (HU: p = 0.725; SUVmax: p = 0.392), autochthonous back muscles (HU: p = 0.281; SUVmax: p = 0.839), pectoral muscles (HU: p = 0.481; SUVmax: p = 0.277 for the right side and HU: p = 0.313; SUVmax: p = 0.859 for the left side), or the internal jugular vein (HU: p = 0.343; SUVmax: p = 0.194). Conclusion: Metal artifact reduction algorithms such as iMAR do not alter the data information of normal organs not affected by artifacts.

11C-methionine PET/MRI in postoperative patients after craniotomy: zero echo time and head atlas versus CT-based attenuation correction.

Attenuation correction (AC) is an important topic in PET/MRI and particularly challenging after brain tumor surgery, near metal implants, adjacent bone and burr holes. In this study, we evaluated the performance of two MR-driven AC methods, zero-echo-time AC (ZTE-AC) and atlas-AC, in comparison to reference standard CT-AC in patients with surgically treated brain tumors at 11C-methionine PET/MRI. This retrospective study investigated seven postoperative patients with neuropathologically confirmed brain tumor at 11C-methionine PET/MRI. Three AC maps - ZTE-AC, atlas-AC and reference standard CT-AC - were generated for each patient. Standardized uptake values (SUV) were obtained at the metal implant, adjacent bone and burr hole. Standard uptake ratio (SUR) SURmetal/mirror, SURbone/mirror and SURburrhole/mirror were then calculated and analyzed with Bland-Altman, Pearson correlation and intraclass correlation reliability. Smaller mean percent bias range (Bland-Altman) was found for ZTE-AC than atlas-AC in all analyses (metal ZTE -0.46 to -0.02, metal atlas -3.57 to -3.26; bone ZTE -4.60 to -2.16, bone atlas -5.25 to -3.81; burr hole ZTE -0.95 to -0.52, burr hole atlas 7.86 to 8.87). Percent SD range (Bland-Altman) was large for both methods in all analyses, with lower absolute values for ZTE-AC (ZTE 7.02-8.49; atlas 11.47-14.83). A very strong correlation (Pearson correlation) was demonstrated for both methods compared to CT-AC (ZTE ρ 0.97-0.99, P<0.001; atlas ρ 0.88-0.91, P≤0.009) with higher absolute values for ZTE. An excellent intraclass correlation coefficient was found across all analyses for ZTE, atlas and CT maps (ICC ≥0.88). ZTE for MR-driven PET attenuation correction presented a more comparable performance to reference standard CT-AC at the postoperative site. ZTE-AC may serve as a useful diagnostic tool for MR-driven AC in patients with surgically treated brain tumors.

Coronary artery calcium burden, carotid atherosclerotic plaque burden, and myocardial blood flow in patients with end-stage renal disease: A non-invasive imaging study combining PET/CT and 3D ultrasound

BackgroundImaging-based measures of atherosclerosis such as coronary artery calcium score (CACS) and coronary flow reserve (CFR) as well as carotid atherosclerotic plaque burden (cPB) are predictors of cardiovascular events in the general population. The objective of this study was to correlate CACS, cPB, myocardial blood flow (MBF), and CFR in patients with end-stage renal disease (ESRD). Methods and results39 patients (mean age 53 ± 12 years) with ESRD prior to kidney transplantation were enrolled. MBF and CFR were quantified at baseline and under hyperemia by 13N-NH3-PET/CT. CACS was calculated from low-dose CT scans acquired for PET attenuation correction. cPB was assessed by 3D ultrasound. Uni- and multivariate regression analyses between these and clinical parameters were performed. Median follow-up time for clinical events was 4.4 years. Kaplan–Meier survival estimates with log-rank test were performed with regards to cardiovascular (CV) events and death of any cause. CACS and cPB were associated in ESRD patients (r = 0.48; p ≤ 0.01). While cPB correlated with age (r = 0.43; p < 0.01), CACS did not. MBFstress was negatively associated with age (r = 0.44; p < 0.01) and time on dialysis (r = 0.42; p < 0.01). There were negative correlations between MBFstress and CACS (r = − 0.62; p < 0.001) and between MBFstress and cPB (r = − 0.43; p < 0.01). Age and CACS were the strongest predictors for MBFstress. CFR was impaired (< 2.0) in eight patients who also presented with higher cPB and higher CACS compared to those with a CFR > 2.0 (p = 0.06 and p = 0.4). In contrast to MBFstress, there was neither a significant correlation between CFR and CACS (r = − 0.2; p = 0.91) nor between CFR and cPB (r = − 0.1; p = 0.55). CV event-free survival was associated with reduced CFR and MBFstress (p = 0.001 and p < 0.001) but not with cPB or CACS. ConclusionsCACS, cPB, and MBFstress are associated in patients with ESRD. Atherosclerosis is earlier detected by MBFstress than by CFR. CV event-free survival is associated with impaired CFR and MBFstress.

Virtual magnetic resonance lumbar spine images generated from computed tomography images using conditional generative adversarial networks

IntroductionThe aim of this study was to generate virtual Magnetic resonance (MR) from computed tomography (CT) using conditional generative adversarial networks (cGAN). MethodsWe selected examinations from 22 adults who obtained their CT and MR lumbar spine examinations. Overall, 4 examinations were used as test data, and 18 examinations were used as training data. A cGAN was trained to generate virtual MR images from the CT images using the corresponding MR images as targets. After training, the generated virtual MR images from test data in epochs 1, 10, 50, 100, 500, and 1000 were compared with the original ones using the mean square error (MSE) and structural similarity index (SSIM). Additionally, two radiologists also performed qualitative assessments. ResultsThe MSE of the virtual MR images decreased as the epoch of the cGANs increased from the original CT images: 8876.7 ± 1192.9 (original CT), 1567.5 ± 433.9 (Epoch 1), 1242.4 ± 442.0 (Epoch 10), 1065.8 ± 478.1 (Epoch 50), 1276.1 ± 718.9 (Epoch 100), 1046.7 ± 488.2 (Epoch 500), and 1031.7 ± 400.0 (Epoch 1000). No considerable differences were observed in the qualitative evaluation between the virtual MR images and the original ones, except in the structure of the spinal canal. ConclusionVirtual MR lumbar spine images using cGANs could be a feasible technique to generate near-MR images from CT without MR examinations for evaluation of the vertebral body and intervertebral disc. Implications for practiceVirtual MR lumbar spine images using cGANs can offer virtual CT images with sufficient quality for attenuation correction for PET or dose planning in radiotherapy.

Lesion detection in 18F-sodium fluoride bone imaging: a comparison of attenuation-corrected versus nonattenuation-corrected PET reconstructions from modern PET-CT systems.

An earlier study demonstrated comparable lesion detection between attenuation-corrected (AC) and nonattenuation-corrected (NAC) 18F-sodium fluoride (NaF) PET images, which is relevant for computed tomography (CT) radiation dose-saving. However, this finding may not be applicable to newer systems. The aim was to compare lesion detection between AC and NAC NaF PET images on modern PET-CT systems. One expert and one nonexpert observer retrospectively surveyed NaF PET data in 25 breast cancer patients. At both lesion and patient level, each observer classified bone abnormalities as malignant, equivocal or benign, from NAC and AC PET images in the absence of CT. Expert interpretation of NaF PET-CT with the review of all diagnostic imaging/pathology reports for at least the subsequent 12 months provided reference standard metastases status at the patient level. Two-tailed Wilcoxon signed-rank tests measured statistically significant differences in total lesion detection between AC and NAC PET. Quadratic-weighted kappa score measured agreement in patient metastases status between observers. On a lesion-basis, AC PET images showed significantly more lesions than NAC for both the expert (122 versus 96; P = 0.002) and nonexpert (146 versus 132; P = 0.036) observers, with a large number of patients demonstrating disparity between AC and NAC images. For metastases status at the patient level without CT, NAC PET showed slightly better diagnostic accuracy than AC due to fewer false-positive results, as fewer lesions were identified. AC PET data provided superior lesion detection to NAC in NaF bone examinations and are thus required for clinical interpretation.

Assessment of deep learning-based PET attenuation correction frameworks in the sinogram domain

This study set out to investigate various deep learning frameworks for PET attenuation correction in the sinogram domain. Different models for both time-of-flight (TOF) and non-TOF PET emission data were implemented, including direct estimation of the attenuation corrected (AC) emission sinograms from the nonAC sinograms, estimation of the attenuation correction factors (ACFs) from PET emission data, correction of scattered photons prior to training of the models, and separate training of the models for each segment of the emission sinograms. A segmentation-based 2-class AC map was included as a bottom-line technique for comparison of the different models considering PET/CT AC as reference. Fifty clinical TOF PET/CT brain scans were employed for training whereas 20 were used for evaluation of the models. Quantitative analysis of the resulting PET images was carried out through region-wise standardized uptake value (SUV) bias calculation. The models relying on TOF information significantly outperformed the nonTOF models as well as the segmentation-based AC map resulting in maximum SUV bias of 6.5%, 9.5%, and 14.0%, respectively. Estimation of ACFs from either TOF or nonTOF PET emission data was very sensitive to prior scatter correction. However, direct estimation of AC sinograms from nonAC sinograms revealed no sensitivity to scatter correction, thus obviating the need for prior scatter estimation. For TOF PET data, though direct prediction of the AC sinograms does not require prior estimation of scattered photons, it requires input/output channels equal to the number of TOF bins which might be computationally or memory-wise expensive. Prediction of the ACF matrices from TOF emission data is less demanding in terms of memory as it requires only a single channel for output. AC in the sinogram domain of TOF PET data exhibited superior performance compared to both nonTOF and segmentation-based methods. However, such models require multiple input/output channels.

Deep learning for whole-body medical image generation.

Artificial intelligence (AI) algorithms based on deep convolutional networks have demonstrated remarkable success for image transformation tasks. State-of-the-art results have been achieved by generative adversarial networks (GANs) and training approaches which do not require paired data. Recently, these techniques have been applied in the medical field for cross-domain image translation. This study investigated deep learning transformation in medical imaging. It was motivated to identify generalizable methods which would satisfy the simultaneous requirements of quality and anatomical accuracy across the entire human body. Specifically, whole-body MR patient data acquired on a PET/MR system were used to generate synthetic CT image volumes. The capacity of these synthetic CT data for use in PET attenuation correction (AC) was evaluated and compared to current MR-based attenuation correction (MR-AC) methods, which typically use multiphase Dixon sequences to segment various tissue types. This work aimed to investigate the technical performance of a GAN system for general MR-to-CT volumetric transformation and to evaluate the performance of the generated images for PET AC. A dataset comprising matched, same-day PET/MR and PET/CT patient scans was used for validation. A combination of training techniques was used to produce synthetic images which were of high-quality and anatomically accurate. Higher correlation was found between the values of mu maps calculated directly from CT data and those derived from the synthetic CT images than those from the default segmented Dixon approach. Over the entire body, the total amounts of reconstructed PET activities were similar between the two MR-AC methods, but the synthetic CT method yielded higher accuracy for quantifying the tracer uptake in specific regions. The findings reported here demonstrate the feasibility of this technique and its potential to improve certain aspects of attenuation correction for PET/MR systems. Moreover, this work may have larger implications for establishing generalized methods for inter-modality, whole-body transformation in medical imaging. Unsupervised deep learning techniques can produce high-quality synthetic images, but additional constraints may be needed to maintain medical integrity in the generated data.

Data-driven respiratory phase-matched PET attenuation correction without CT

We propose a deep learning-based data-driven respiratory phase-matched gated-PET attenuation correction (AC) method that does not need a gated-CT. The proposed method is a multi-step process that consists of data-driven respiratory gating, gated attenuation map estimation using maximum-likelihood reconstruction of attenuation and activity (MLAA) algorithm, and enhancement of the gated attenuation maps using convolutional neural network (CNN). The gated MLAA attenuation maps enhanced by the CNN allowed for the phase-matched AC of gated-PET images. We conducted a non-rigid registration of the gated-PET images to generate motion-free PET images. We trained the CNN by conducting a 3D patch-based learning with 80 oncologic whole-body 18F-fluorodeoxyglucose (18F-FDG) PET/CT scan data and applied it to seven regional PET/CT scans that cover the lower lung and upper liver. We investigated the impact of the proposed respiratory phase-matched AC of PET without utilizing CT on tumor size and standard uptake value (SUV) assessment, and PET image quality (%STD). The attenuation corrected gated and motion-free PET images generated using the proposed method yielded sharper organ boundaries and better noise characteristics than conventional gated and ungated PET images. A banana artifact observed in a phase-mismatched CT-based AC was not observed in the proposed approach. By employing the proposed method, the size of tumor was reduced by 12.3% and SUV90% was increased by 13.3% in tumors with larger movements than 5 mm. %STD of liver uptake was reduced by 11.1%. The deep learning-based data-driven respiratory phase-matched AC method improved the PET image quality and reduced the motion artifacts.

Feasibility of Deep Learning-Guided Attenuation and Scatter Correction of Whole-Body 68Ga-PSMA PET Studies in the Image Domain.

This study evaluates the feasibility of direct scatter and attenuation correction of whole-body 68Ga-PSMA PET images in the image domain using deep learning. Whole-body 68Ga-PSMA PET images of 399 subjects were used to train a residual deep learning model, taking PET non-attenuation-corrected images (PET-nonAC) as input and CT-based attenuation-corrected PET images (PET-CTAC) as target (reference). Forty-six whole-body 68Ga-PSMA PET images were used as an independent validation dataset. For validation, synthetic deep learning-based attenuation-corrected PET images were assessed considering the corresponding PET-CTAC images as reference. The evaluation metrics included the mean absolute error (MAE) of the SUV, peak signal-to-noise ratio, and structural similarity index (SSIM) in the whole body, as well as in different regions of the body, namely, head and neck, chest, and abdomen and pelvis. The deep learning-guided direct attenuation and scatter correction produced images of comparable visual quality to PET-CTAC images. It achieved an MAE, relative error (RE%), SSIM, and peak signal-to-noise ratio of 0.91 ± 0.29 (SUV), -2.46% ± 10.10%, 0.973 ± 0.034, and 48.171 ± 2.964, respectively, within whole-body images of the independent external validation dataset. The largest RE% was observed in the head and neck region (-5.62% ± 11.73%), although this region exhibited the highest value of SSIM metric (0.982 ± 0.024). The MAE (SUV) and RE% within the different regions of the body were less than 2.0% and 6%, respectively, indicating acceptable performance of the deep learning model. This work demonstrated the feasibility of direct attenuation and scatter correction of whole-body 68Ga-PSMA PET images in the image domain using deep learning with clinically tolerable errors. The technique has the potential of performing attenuation correction on stand-alone PET or PET/MRI systems.

Deep learning\u2013based metal artefact reduction in PET/CT imaging

ObjectivesThe susceptibility of CT imaging to metallic objects gives rise to strong streak artefacts and skewed information about the attenuation medium around the metallic implants. This metal-induced artefact in CT images leads to inaccurate attenuation correction in PET/CT imaging. This study investigates the potential of deep learning–based metal artefact reduction (MAR) in quantitative PET/CT imaging.MethodsDeep learning–based metal artefact reduction approaches were implemented in the image (DLI-MAR) and projection (DLP-MAR) domains. The proposed algorithms were quantitatively compared to the normalized MAR (NMAR) method using simulated and clinical studies. Eighty metal-free CT images were employed for simulation of metal artefact as well as training and evaluation of the aforementioned MAR approaches. Thirty 18F-FDG PET/CT images affected by the presence of metallic implants were retrospectively employed for clinical assessment of the MAR techniques.ResultsThe evaluation of MAR techniques on the simulation dataset demonstrated the superior performance of the DLI-MAR approach (structural similarity (SSIM) = 0.95 ± 0.2 compared to 0.94 ± 0.2 and 0.93 ± 0.3 obtained using DLP-MAR and NMAR, respectively) in minimizing metal artefacts in CT images. The presence of metallic artefacts in CT images or PET attenuation correction maps led to quantitative bias, image artefacts and under- and overestimation of scatter correction of PET images. The DLI-MAR technique led to a quantitative PET bias of 1.3 ± 3% compared to 10.5 ± 6% without MAR and 3.2 ± 0.5% achieved by NMAR.ConclusionThe DLI-MAR technique was able to reduce the adverse effects of metal artefacts on PET images through the generation of accurate attenuation maps from corrupted CT images.Key Points• The presence of metallic objects, such as dental implants, gives rise to severe photon starvation, beam hardening and scattering, thus leading to adverse artefacts in reconstructed CT images.• The aim of this work is to develop and evaluate a deep learning–based MAR to improve CT-based attenuation and scatter correction in PET/CT imaging.• Deep learning–based MAR in the image (DLI-MAR) domain outperformed its counterpart implemented in the projection (DLP-MAR) domain. The DLI-MAR approach minimized the adverse impact of metal artefacts on whole-body PET images through generating accurate attenuation maps from corrupted CT images.

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

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

Evaluating the image quality of combined positron emission tomography-magnetic resonance images acquired in the pelvic radiotherapy position

Positron emission tomography-magnetic resonance (PET-MR) scanners could improve radiotherapy planning through combining PET and MR functional imaging. This depends on acquiring high quality and quantitatively accurate images in the radiotherapy position. This study evaluated PET-MR image quality using a flat couch and coil bridge for pelvic radiotherapy. MR and PET image quality phantoms were imaged in three setups: phantom on the PET-MR couch with anterior coil on top (diagnostic), phantom on a flat couch with coil on top (couch), and phantom on the flat couch with coil on a coil bridge (radiotherapy). PET images were also acquired in each setup without the anterior coil. PET attenuation correction of the flat couch and coil bridge were generated using kilovoltage computed tomography (CT) images and of the anterior coil using megavoltage CT images. MR image quality was substantially affected, with MR signal to noise ratio (SNR) relative to the diagnostic setup of 89% ± 2% (mean ± standard error of the mean, couch) and 54% ± 1% (radiotherapy), likely due to the increased distance between the patient and receive coils. The reduction impacted the low-contrast detectability score: 23 ± 1 (diagnostic), 19.7 ± 0.3 (couch) and 15 ± 1 (radiotherapy). All other MR metrics agreed within one standard error. PET quantitative accuracy was also affected, with measured activity with anterior coil being different to diagnostic without anterior coil by −16.7% ± 0.2% (couch) and −17.7 ± 0.1% (radiotherapy), without attenuation correction modification. Including the couch and coil bridge attenuation correction reduced this difference to −7.5% ± 0.1%, and including the anterior coil reduced this to −2.7% ± 0.1%. This was better than the diagnostic setup with anterior coil (difference −8.3% ± 0.2%). This translated into greater PET SNR performance for the fully corrected radiotherapy setup compared to diagnostic with coil. However contrast recovery was unchanged by the modified attenuation correction, with the diagnostic setup remaining ∼2% better. Quantitative PET in the radiotherapy setup is possible if appropriate attenuation correction is used. Pelvic radiotherapy PET-MR imaging protocols will need to consider the impact on PET-MR image quality.

Integrated PET/MRI scanner with oxygen-15 labeled gases for quantification of cerebral blood flow, cerebral blood volume, cerebral oxygen extraction fraction and cerebral metabolic rate of oxygen.

Measurement of cerebral blood flow (CBF), cerebral blood volume (CBV), cerebral oxygen extraction fraction (OEF) and cerebral metabolic rate of oxygen (CMRO2) by PET with oxygen-15 labeled gases is useful for diagnosis and treatment planning in cases of chronic occlusive cerebrovascular disease. In the present study, CBF, CBV, OEF and CMRO2 were measured using the integrated design of PET/MRI scanner system. This is a first attempt to measure cerebral perfusion and oxygen metabolism using PET/MRI with oxygen-15 labeled gases. PET/MRI measurements with the steady-state method of oxygen-15 labeled gases, carbon monoxide (C15O), oxygen (15O2), and carbon dioxide (C15O2) were performed on nine healthy men. Two kinds of attenuation correction for PET were performed using MRI with Dixon sequence (DIXON) and Dixon sequence with model-based bone segmentation (DIXONbone). A real-time motion correction of PET images was also performed using simultaneously measured MR images to detect head motion. Mean and SD values of CBF, CBV, OEF, and CMRO2 in the cerebral cortices with attenuation correction by DIXON were 31 ± 4mL/100mL/min, 2.7 ± 0.2mL/mL, 0.40 ± 0.07, and 2.5 ± 0.3mL/100mL/min without real-time motion correction, and 33 ± 4mL/100mL/min, 2.7 ± 0.2mL/mL, 0.40 ± 0.07, and 2.6 ± 0.3mL/100mL/min with real-time motion correction, respectively. Values with of CBF, CBV, OEF, and CMRO2 with attenuation correction by DIXONbone were 35 ± 5mL/100mL/min, 2.8 ± 0.2mL/mL, 0.40 ± 0.07, and 2.8 ± 0.3mL/100mL/min without real-time motion correction, and 38 ± 5mL/100mL/min, 2.8 ± 0.2mL/mL, 0.40 ± 0.07, and 3.0 ± 0.4mL/100mL/min with real-time motion correction, respectively. Using PET/MRI with oxygen-15 labeled gases, CBF, CBV, OEF, and CMRO2 could be measured. Values of CBF, CBV, and CMRO2 measured with attenuation correction by DIXON were significantly lower than those measured with correction by DIXONbone. One of the reasons for this is that attenuation correction of DIXON does not take into consideration of the photon absorption by bone. OEF values, corresponding to ratios of CMRO2 to CBF, were not affected by attenuation correction methods. Values of CBF and CMRO2 with a real-time motion correction were significantly higher than those without correction. Using PET/MRI with adequate corrections, similar values of CBF, CBV, OEF, and CMRO2 as PET alone scanner system reported previously were obtained. The UMIN clinical trial number: UMIN000033382.

Transforming UTE-mDixon MR Abdomen-Pelvis Images Into CT by Jointly Leveraging Prior Knowledge and Partial Supervision.

Computed tomography (CT) provides information for diagnosis, PET attenuation correction (AC), and radiation treatment planning (RTP). Disadvantages of CT include poor soft tissue contrast and exposure to ionizing radiation. While MRI can overcome these disadvantages, it lacks the photon absorption information needed for PET AC and RTP. Thus, an intelligent transformation from MR to CT, i.e., the MR-based synthetic CT generation, is of great interest as it would support PET/MR AC and MR-only RTP. Using an MR pulse sequence that combines ultra-short echo time (UTE) and modified Dixon (mDixon), we propose a novel method for synthetic CT generation jointly leveraging prior knowledge as well as partial supervision (SCT-PK-PS for short) on large-field-of-view images that span abdomen and pelvis. Two key machine learning techniques, i.e., the knowledge-leveraged transfer fuzzy c-means (KL-TFCM) and the Laplacian support vector machine (LapSVM), are used in SCT-PK-PS. The significance of our effort is threefold: 1) Using the prior knowledge-referenced KL-TFCM clustering, SCT-PK-PS is able to group the feature data of MR images into five initial clusters of fat, soft tissue, air, bone, and bone marrow. Via these initial partitions, clusters needing to be refined are observed and for each of them a few additionally labeled examples are given as the partial supervision for the subsequent semi-supervised classification using LapSVM; 2) Partial supervision is usually insufficient for conventional algorithms to learn the insightful classifier. Instead, exploiting not only the given supervision but also the manifold structure embedded primarily in numerous unlabeled data, LapSVM is capable of training multiple desired tissue-recognizers; 3) Benefiting from the joint use of KL-TFCM and LapSVM, and assisted by the edge detector filter based feature extraction, the proposed SCT-PK-PS method features good recognition accuracy of tissue types, which ultimately facilitates the good transformation from MR images to CT images of the abdomen-pelvis. Applying the method on twenty subjects' feature data of UTE-mDixon MR images, the average score of the mean absolute prediction deviation (MAPD) of all subjects is 140.72 ± 30.60 HU which is statistically significantly better than the 241.36 ± 21.79 HU obtained using the all-water method, the 262.77 ± 42.22 HU obtained using the four-cluster-partitioning (FCP, i.e., external-air, internal-air, fat, and soft tissue) method, and the 197.05 ± 76.53 HU obtained via the conventional SVM method. These results demonstrate the effectiveness of our method for the intelligent transformation from MR to CT on the body section of abdomen-pelvis.

Abstract 14657: Intermuscular Fat is Independently Associated With Coronary Microvascular Dysfunction and Adverse Cardiovascular Outcomes in Patients With No Obstructive CAD

Introduction: Intermuscular adipose tissue (IMAT) reflects skeletal muscle quality and is associated with inflammation, a key determinant in cardiometabolic disease. Coronary flow reserve (CFR), a marker of coronary microvascular dysfunction (CMD), is independently associated with BMI, inflammation and risk of heart failure, myocardial infarction and death. Using a novel neural networks algorithm for body composition analysis, we sought to investigate the relationship between skeletal muscle quality, CMD and cardiovascular outcomes. Methods: Consecutive patients (N=669) undergoing evaluation for CAD with cardiac stress PET demonstrating normal perfusion and preserved LVEF were followed over median 6 years for events (MACE), including death and hospitalization for myocardial infarction or heart failure. CFR was calculated as the stress/rest myocardial blood flow ratio. Subcutaneous fat (SAT) and IMAT areas (cm 2 ) were obtained from PET attenuation correction CTs using a fully automated algorithm validated by manual segmentation at T10 and T12 levels, respectively. Results: Median age was 63 years, 70% were female and 46% nonwhite. Nearly half of patients were obese (46% with BMI 30-61) and BMI correlated closely with SAT and IMAT (r=0.8 and 0.7, respectively, p&lt;0.001). Independent of age, sex, race, BMI, LVEF and hypertension, CFR was associated with IMAT (β -0.10, p=0.047). In adjusted analyses, both CFR and increased IMAT were independently associated with MACE [HR 1.6 (1.1-2.4) per -1U CFR and 1.5 (1.2-1.7) per +10 cm 2 IMAT, p≤0.01]. Patients with both CMD and poor muscle quality demonstrated highest risk of MACE (adjusted p&lt;0.02). Conclusion: Increased intermuscular fat is associated with CMD and adverse cardiovascular outcomes independently of BMI and traditional risk factors. Combined assessment of skeletal muscle quality and coronary microvascular function may help to redefine at-risk cardiometabolic phenotypes.

PET Image Quality Improvement for Simultaneous PET/MRI with a Lightweight MRI Surface Coil.

Background During simultaneous PET/MRI, flexible MRI surface coils that lay on the patient are often omitted from PET attenuation correction processing, leading to quantification bias in PET images. Purpose To identify potential PET image quality improvement by using a recently developed lightweight MRI coil technology for the anterior array (AA) surface coil in both a phantom and in vivo study. Materials and Methods A phantom study and a prospective in vivo study were performed with a PET/CT scanner under three conditions: (a) no MRI surface coil (standard of reference), (b) traditional AA coil, and (c) lightweight AA coil. AA coils were not used in attenuation correction processing to emulate clinical PET/MRI. For the phantom study, PET images were reconstructed with and without time of flight (TOF) to assess quantification accuracy and uniformity. The in vivo study consisted of 10 participants (mean age, 66 years ± 10 [standard deviation]; six men) referred for a PET/CT oncologic examination who had undergone imaging between October 2019 and February 2020. Assessment of image quantification bias (defined as the standard error of the mean values) was conducted by comparing mean liver region of interest standardized uptake values with the no-coil standard of reference. A Wilcoxon signed-rank test was used to establish significance. Results For TOF and non-TOF, respectively, the phantom study revealed a mean PET quantification bias of -9.0% and -8.6% with the traditional AA coil and a mean PET quantification bias of -4.3% and -4.0% with the lightweight AA coil. The coefficients of variation reduced from 4.3% and 6.2% with the traditional AA coil to 2.1% and 2.7% with the lightweight AA coil, which demonstrated a homogeneity benefit from the lightweight coil that was greater with, versus without, TOF reconstruction. For the in vivo study, the mean liver standardized uptake value error was -5.9% with the traditional AA coil (P = .002 vs no coil) and -2.4% with the lightweight AA coil (P = .004 vs no coil). Conclusion The lightweight anterior array coil reduced PET image quantification bias by more than 50% compared with the traditional coil. Using the lightweight coil and performing time of flight-based reconstruction each reduced the variation of error. © RSNA, 2020 Online supplemental material is available for this article.