Attenuation correction for brain PET imaging using deep neural network based on Dixon and ZTE MR images

Deep learning applications for quantitative and qualitative PET in PET/MR: technical and clinical unmet needs

We aim to provide an overview of technical and clinical unmet needs in deep learning (DL) applications for quantitative and qualitative PET in PET/MR, with a focus on attenuation correction, image enhancement, motion correction, kinetic modeling, and simulated data generation. (1) DL-based attenuation correction (DLAC) remains an area of limited exploration for pediatric whole-body PET/MR and lung-specific DLAC due to data shortages and technical limitations. (2) DL-based image enhancement approximating MR-guided regularized reconstruction with a high-resolution MR prior has shown promise in enhancing PET image quality. However, its clinical value has not been thoroughly evaluated across various radiotracers, and applications outside the head may pose challenges due to motion artifacts. (3) Robust training for DL-based motion correction requires pairs of motion-corrupted and motion-corrected PET/MR data. However, these pairs are rare. (4) DL-based approaches can address the limitations of dynamic PET, such as long scan durations that may cause patient discomfort and motion, providing new research opportunities. (5) Monte-Carlo simulations using anthropomorphic digital phantoms can provide extensive datasets to address the shortage of clinical data. This summary of technical/clinical challenges and potential solutions may provide research opportunities for the research community towards the clinical translation of DL solutions.

A systematic literature review: deep learning techniques for synthetic medical image generation and their applications in radiotherapy

The aim of this systematic review is to determine whether Deep Learning (DL) algorithms can provide a clinically feasible alternative to classic algorithms for synthetic Computer Tomography (sCT). The following categories are presented in this study: ∙ MR-based treatment planning and synthetic CT generation techniques. ∙ Generation of synthetic CT images based on Cone Beam CT images. ∙ Low-dose CT to High-dose CT generation. ∙ Attenuation correction for PET images. To perform appropriate database searches, we reviewed journal articles published between January 2018 and June 2023. Current methodology, study strategies, and results with relevant clinical applications were analyzed as we outlined the state-of-the-art of deep learning based approaches to inter-modality and intra-modality image synthesis. This was accomplished by contrasting the provided methodologies with traditional research approaches. The key contributions of each category were highlighted, specific challenges were identified, and accomplishments were summarized. As a final step, the statistics of all the cited works from various aspects were analyzed, which revealed that DL-based sCTs have achieved considerable popularity, while also showing the potential of this technology. In order to assess the clinical readiness of the presented methods, we examined the current status of DL-based sCT generation.

Contrast-enhanced to non-contrast-enhanced image translation to exploit a clinical data warehouse of T1-weighted brain MRI

BACKGROUND

Clinical data warehouses provide access to massive amounts of medical images, but these images are often heterogeneous. They can for instance include images acquired both with or without the injection of a gadolinium-based contrast agent. Harmonizing such data sets is thus fundamental to guarantee unbiased results, for example when performing differential diagnosis. Furthermore, classical neuroimaging software tools for feature extraction are typically applied only to images without gadolinium. The objective of this work is to evaluate how image translation can be useful to exploit a highly heterogeneous data set containing both contrast-enhanced and non-contrast-enhanced images from a clinical data warehouse.

METHODS

We propose and compare different 3D U-Net and conditional GAN models to convert contrast-enhanced T1-weighted (T1ce) into non-contrast-enhanced (T1nce) brain MRI. These models were trained using 230 image pairs and tested on 77 image pairs from the clinical data warehouse of the Greater Paris area.

RESULTS

Validation using standard image similarity measures demonstrated that the similarity between real and synthetic T1nce images was higher than between real T1nce and T1ce images for all the models compared. The best performing models were further validated on a segmentation task. We showed that tissue volumes extracted from synthetic T1nce images were closer to those of real T1nce images than volumes extracted from T1ce images.

CONCLUSION

We showed that deep learning models initially developed with research quality data could synthesize T1nce from T1ce images of clinical quality and that reliable features could be extracted from the synthetic images, thus demonstrating the ability of such methods to help exploit a data set coming from a clinical data warehouse.

Enhanced direct joint attenuation and scatter correction of whole-body PET images via context-aware deep networks

In positron emission tomography (PET), attenuation and scatter corrections are necessary steps toward accurate quantitative reconstruction of the radiopharmaceutical distribution. Inspired by recent advances in deep learning, many algorithms based on convolutional neural networks have been proposed for automatic attenuation and scatter correction, enabling applications to CT-less or MR-less PET scanners to improve performance in the presence of CT-related artifacts. A known characteristic of PET imaging is to have varying tracer uptakes for various patients and/or anatomical regions. However, existing deep learning-based algorithms utilize a fixed model across different subjects and/or anatomical regions during inference, which could result in spurious outputs. In this work, we present a novel deep learning-based framework for the direct reconstruction of attenuation and scatter-corrected PET from non-attenuation-corrected images in the absence of structural information in the inference. To deal with inter-subject and intra-subject uptake variations in PET imaging, we propose a novel model to perform subject- and region-specific filtering through modulating the convolution kernels in accordance to the contextual coherency within the neighboring slices. This way, the context-aware convolution can guide the composition of intermediate features in favor of regressing input-conditioned and/or region-specific tracer uptakes. We also utilized a large cohort of 910 whole-body studies for training and evaluation purposes, which is more than one order of magnitude larger than previous works. In our experimental studies, qualitative assessments showed that our proposed CT-free method is capable of producing corrected PET images that accurately resemble ground truth images corrected with the aid of CT scans. For quantitative assessments, we evaluated our proposed method over 112 held-out subjects and achieved an absolute relative error of 14.30±3.88% and a relative error of -2.11%±2.73% in whole-body.

Attenuation Correction of Long Axial Field-of-View Positron Emission Tomography Using Synthetic Computed Tomography Derived from the Emission Data: Application to Low-Count Studies and Multiple Tracers

Recent advancements in PET/CT, including the emergence of long axial field-of-view (LAFOV) PET/CT scanners, have increased PET sensitivity substantially. Consequently, there has been a significant reduction in the required tracer activity, shifting the primary source of patient radiation dose exposure to the attenuation correction (AC) CT scan during PET imaging. This study proposes a parameter-transferred conditional generative adversarial network (PT-cGAN) architecture to generate synthetic CT (sCT) images from non-attenuation corrected (NAC) PET images, with separate networks for [18F]FDG and [15O]H2O tracers. The study includes a total of 1018 subjects (n = 972 [18F]FDG, n = 46 [15O]H2O). Testing was performed on the LAFOV scanner for both datasets. Qualitative analysis found no differences in image quality in 30 out of 36 cases in FDG patients, with minor insignificant differences in the remaining 6 cases. Reduced artifacts due to motion between NAC PET and CT were found. For the selected organs, a mean average error of 0.45% was found for the FDG cohort, and that of 3.12% was found for the H2O cohort. Simulated low-count images were included in testing, which demonstrated good performance down to 45 s scans. These findings show that the AC of total-body PET is feasible across tracers and in low-count studies and might reduce the artifacts due to motion and metal implants.

Deep-learning-based methods of attenuation correction for SPECT and PET

Attenuation correction (AC) is essential for quantitative analysis and clinical diagnosis of single-photon emission computed tomography (SPECT) and positron emission tomography (PET). In clinical practice, computed tomography (CT) is utilized to generate attenuation maps (μ-maps) for AC of hybrid SPECT/CT and PET/CT scanners. However, CT-based AC methods frequently produce artifacts due to CT artifacts and misregistration of SPECT-CT and PET-CT scans. Segmentation-based AC methods using magnetic resonance imaging (MRI) for PET/MRI scanners are inaccurate and complicated since MRI does not contain direct information of photon attenuation. Computational AC methods for SPECT and PET estimate attenuation coefficients directly from raw emission data, but suffer from low accuracy, cross-talk artifacts, high computational complexity, and high noise level. The recently evolving deep-learning-based methods have shown promising results in AC of SPECT and PET, which can be generally divided into two categories: indirect and direct strategies. Indirect AC strategies apply neural networks to transform emission, transmission, or MR images into synthetic μ-maps or CT images which are then incorporated into AC reconstruction. Direct AC strategies skip the intermediate steps of generating μ-maps or CT images and predict AC SPECT or PET images from non-attenuation-correction (NAC) SPECT or PET images directly. These deep-learning-based AC methods show comparable and even superior performance to non-deep-learning methods. In this article, we first discussed the principles and limitations of non-deep-learning AC methods, and then reviewed the status and prospects of deep-learning-based methods for AC of SPECT and PET.

Transfer learning-based attenuation correction for static and dynamic cardiac PET using a generative adversarial network

PURPOSE

The goal of this work is to demonstrate the feasibility of directly generating attenuation-corrected PET images from non-attenuation-corrected (NAC) PET images for both rest and stress-state static or dynamic [13N]ammonia MP PET based on a generative adversarial network.

METHODS

We recruited 60 subjects for rest-only scans and 14 subjects for rest-stress scans, all of whom underwent [13N]ammonia cardiac PET/CT examinations to acquire static and dynamic frames with both 3D NAC and CT-based AC (CTAC) PET images. We developed a 3D pix2pix deep learning AC (DLAC) framework via a U-net + ResNet-based generator and a convolutional neural network-based discriminator. Paired static or dynamic NAC and CTAC PET images from 60 rest-only subjects were used as network inputs and labels for static (S-DLAC) and dynamic (D-DLAC) training, respectively. The pre-trained S-DLAC network was then fine-tuned by paired dynamic NAC and CTAC PET frames of 60 rest-only subjects to derive an improved D-DLAC-FT for dynamic PET images. The 14 rest-stress subjects were used as an internal testing dataset and separately tested on different network models without training. The proposed methods were evaluated using visual quality and quantitative metrics.

RESULTS

The proposed S-DLAC, D-DLAC, and D-DLAC-FT methods were consistent with clinical CTAC in terms of various images and quantitative metrics. The S-DLAC (slope = 0.9423, R2 = 0.947) showed a higher correlation with the reference static CTAC as compared to static NAC (slope = 0.0992, R2 = 0.654). D-DLAC-FT yielded lower myocardial blood flow (MBF) errors in the whole left ventricular myocardium than D-DLAC, but with no significant difference, both for the 60 rest-state subjects (6.63 ± 5.05% vs. 7.00 ± 6.84%, p = 0.7593) and the 14 stress-state subjects (1.97 ± 2.28% vs. 3.21 ± 3.89%, p = 0.8595).

CONCLUSION

The proposed S-DLAC, D-DLAC, and D-DLAC-FT methods achieve comparable performance with clinical CTAC. Transfer learning shows promising potential for dynamic MP PET.

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.

Quantitative Assessment of Myocardial Ischemia With Positron Emission Tomography

Recent advances in positron emission tomography (PET) technology and reconstruction techniques have now made quantitative assessment using cardiac PET readily available in most cardiac PET imaging centers. Multiple PET myocardial perfusion imaging (MPI) radiopharmaceuticals are available for quantitative examination of myocardial ischemia, with each having distinct convenience and accuracy profile. Important properties of these radiopharmaceuticals ( 15 O-water, 13 N-ammonia, 82 Rb, 11 C-acetate, and 18 F-flurpiridaz) including radionuclide half-life, mean positron range in tissue, and the relationship between kinetic parameters and myocardial blood flow (MBF) are presented. Absolute quantification of MBF requires PET MPI to be performed with protocols that allow the generation of dynamic multiframes of reconstructed data. Using a tissue compartment model, the rate constant that governs the rate of PET MPI radiopharmaceutical extraction from the blood plasma to myocardial tissue is calculated. Then, this rate constant ( K1 ) is converted to MBF using an established extraction formula for each radiopharmaceutical. As most of the modern PET scanners acquire the data only in list mode, techniques of processing the list-mode data into dynamic multiframes are also reviewed. Finally, the impact of modern PET technologies such as PET/CT, PET/MR, total-body PET, machine learning/deep learning on comprehensive and quantitative assessment of myocardial ischemia is briefly described in this review.

Synthetic Attenuation Correction Maps for SPECT Imaging Using Deep Learning: A Study on Myocardial Perfusion Imaging

(1) Background: The CT-based attenuation correction of SPECT images is essential for obtaining accurate quantitative images in cardiovascular imaging. However, there are still many SPECT cameras without associated CT scanners throughout the world, especially in developing countries. Performing additional CT scans implies troublesome planning logistics and larger radiation doses for patients, making it a suboptimal solution. Deep learning (DL) offers a revolutionary way to generate complementary images for individual patients at a large scale. Hence, we aimed to generate linear attenuation coefficient maps from SPECT emission images reconstructed without attenuation correction using deep learning. (2) Methods: A total of 384 SPECT myocardial perfusion studies that used ^99mTc-sestamibi were included. A DL model based on a 2D U-Net architecture was trained using information from 312 patients. The quality of the generated synthetic attenuation correction maps (ACMs) and reconstructed emission values were evaluated using three metrics and compared to standard-of-care data using Bland-Altman plots. Finally, a quantitative evaluation of myocardial uptake was performed, followed by a semi-quantitative evaluation of myocardial perfusion. (3) Results: In a test set of 66 test patients, the ACM quality metrics were MSSIM = 0.97 ± 0.001 and NMAE = 3.08 ± 1.26 (%), and the reconstructed emission quality metrics were MSSIM = 0.99 ± 0.003 and NMAE = 0.23 ± 0.13 (%). The 95% limits of agreement (LoAs) at the voxel level for reconstructed SPECT images were: [-9.04; 9.00]%, and for the segment level, they were [-11; 10]%. The 95% LoAs for the Summed Stress Score values between the images reconstructed were [-2.8, 3.0]. When global perfusion scores were assessed, only 2 out of 66 patients showed changes in perfusion categories. (4) Conclusion: Deep learning can generate accurate attenuation correction maps from non-attenuation-corrected cardiac SPECT images. These high-quality attenuation maps are suitable for attenuation correction in myocardial perfusion SPECT imaging and could obviate the need for additional imaging in standalone SPECT scanners.

Generation of Whole-Body FDG Parametric Ki Images from Static PET Images Using Deep Learning

FDG parametric Ki images show great advantage over static SUV images, due to the higher contrast and better accuracy in tracer uptake rate estimation. In this study, we explored the feasibility of generating synthetic Ki images from static SUV ratio (SUVR) images using three configurations of U-Nets with different sets of input and output image patches, which were the U-Nets with single input and single output (SISO), multiple inputs and single output (MISO), and single input and multiple outputs (SIMO). SUVR images were generated by averaging three 5-min dynamic SUV frames starting at 60 minutes post-injection, and then normalized by the mean SUV values in the blood pool. The corresponding ground truth Ki images were derived using Patlak graphical analysis with input functions from measurement of arterial blood samples. Even though the synthetic Ki values were not quantitatively accurate compared with ground truth, the linear regression analysis of joint histograms in the voxels of body regions showed that the mean R2 values were higher between U-Net prediction and ground truth (0.596, 0.580, 0.576 in SISO, MISO and SIMO), than that between SUVR and ground truth Ki (0.571). In terms of similarity metrics, the synthetic Ki images were closer to the ground truth Ki images (mean SSIM = 0.729, 0.704, 0.704 in SISO, MISO and MISO) than the input SUVR images (mean SSIM = 0.691). Therefore, it is feasible to use deep learning networks to estimate surrogate map of parametric Ki images from static SUVR images.

Synthetic PET via Domain Translation of 3-D MRI

Historically, patient datasets have been used to develop and validate various reconstruction algorithms for PET/MRI and PET/CT. To enable such algorithm development, without the need for acquiring hundreds of patient exams, in this article we demonstrate a deep learning technique to generate synthetic but realistic whole-body PET sinograms from abundantly available whole-body MRI. Specifically, we use a dataset of 56 18F-FDG-PET/MRI exams to train a 3-D residual UNet to predict physiologic PET uptake from whole-body T1-weighted MRI. In training, we implemented a balanced loss function to generate realistic uptake across a large dynamic range and computed losses along tomographic lines of response to mimic the PET acquisition. The predicted PET images are forward projected to produce synthetic PET (sPET) time-of-flight (ToF) sinograms that can be used with vendor-provided PET reconstruction algorithms, including using CT-based attenuation correction (CTAC) and MR-based attenuation correction (MRAC). The resulting synthetic data recapitulates physiologic 18F-FDG uptake, e.g., high uptake localized to the brain and bladder, as well as uptake in liver, kidneys, heart, and muscle. To simulate abnormalities with high uptake, we also insert synthetic lesions. We demonstrate that this sPET data can be used interchangeably with real PET data for the PET quantification task of comparing CTAC and MRAC methods, achieving ≤ 7.6% error in mean-SUV compared to using real data. These results together show that the proposed sPET data pipeline can be reasonably used for development, evaluation, and validation of PET/MRI reconstruction methods.

The performance of machine learning approaches for attenuation correction of PET in neuroimaging: A meta-analysis

PURPOSE

This systematic review provides a consensus on the clinical feasibility of machine learning (ML) methods for brain PET attenuation correction (AC). Performance of ML-AC were compared to clinical standards.

METHODS

Two hundred and eighty studies were identified through electronic searches of brain PET studies published between January 1, 2008, and August 1, 2022. Reported outcomes for image quality, tissue classification performance, regional and global bias were extracted to evaluate ML-AC performance. Methodological quality of included studies and the quality of evidence of analysed outcomes were assessed using QUADAS-2 and GRADE, respectively.

RESULTS

A total of 19 studies (2371 participants) met the inclusion criteria. Overall, the global bias of ML methods was 0.76 ± 1.2%. For image quality, the relative mean square error (RMSE) was 0.20 ± 0.4 while for tissues classification, the Dice similarity coefficient (DSC) for bone/soft tissue/air were 0.82 ± 0.1 / 0.95 ± 0.03 / 0.85 ± 0.14.

CONCLUSIONS

In general, ML-AC performance is within acceptable limits for clinical PET imaging. The sparse information on ML-AC robustness and its limited qualitative clinical evaluation may hinder clinical implementation in neuroimaging, especially for PET/MRI or emerging brain PET systems where standard AC approaches are not readily available.

A deep neural network for parametric image reconstruction on a large axial field-of-view PET

PURPOSE

The PET scanners with long axial field of view (AFOV) having ~ 20 times higher sensitivity than conventional scanners provide new opportunities for enhanced parametric imaging but suffer from the dramatically increased volume and complexity of dynamic data. This study reconstructed a high-quality direct Patlak Ki image from five-frame sinograms without input function by a deep learning framework based on DeepPET to explore the potential of artificial intelligence reducing the acquisition time and the dependence of input function in parametric imaging.

METHODS

This study was implemented on a large AFOV PET/CT scanner (Biograph Vision Quadra) and twenty patients were recruited with ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) dynamic scans. During training and testing of the proposed deep learning framework, the last five-frame (25 min, 40-65 min post-injection) sinograms were set as input and the reconstructed Patlak Ki images by a nested EM algorithm on the vendor were set as ground truth. To evaluate the image quality of predicted Ki images, mean square error (MSE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM) were calculated. Meanwhile, a linear regression process was applied between predicted and true Ki means on avid malignant lesions and tumor volume of interests (VOIs).

RESULTS

In the testing phase, the proposed method achieved excellent MSE of less than 0.03%, high SSIM, and PSNR of ~ 0.98 and ~ 38 dB, respectively. Moreover, there was a high correlation (DeepPET: [Formula: see text]= 0.73, self-attention DeepPET: [Formula: see text]=0.82) between predicted Ki and traditionally reconstructed Patlak Ki means over eleven lesions.

CONCLUSIONS

The results show that the deep learning-based method produced high-quality parametric images from small frames of projection data without input function. It has much potential to address the dilemma of the long scan time and dependency on input function that still hamper the clinical translation of dynamic PET.

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.

Artificial intelligence in multi-parametric magnetic resonance imaging: A review

Multi-parametric magnetic resonance imaging (mpMRI) is an indispensable tool in the clinical workflow for the diagnosis and treatment planning of various diseases. Machine learning-based artificial intelligence (AI) methods, especially those adopting the deep learning technique, have been extensively employed to perform mpMRI image classification, segmentation, registration, detection, reconstruction, and super-resolution. The current availability of increasing computational power and fast-improving AI algorithms have empowered numerous computer-based systems for applying mpMRI to disease diagnosis, imaging-guided radiotherapy, patient risk and overall survival time prediction, and the development of advanced quantitative imaging technology for magnetic resonance fingerprinting. However, the wide application of these developed systems in the clinic is still limited by a number of factors, including robustness, reliability, and interpretability. This survey aims to provide an overview for new researchers in the field as well as radiologists with the hope that they can understand the general concepts, main application scenarios, and remaining challenges of AI in mpMRI. This article is protected by copyright. All rights reserved.

Deep Learning for Synthetic CT from Bone MRI in the Head and Neck

BACKGROUND AND PURPOSE

Bone MR imaging techniques enable visualization of cortical bone without the need for ionizing radiation. Automated conversion of bone MR imaging to synthetic CT is highly desirable for downstream image processing and eventual clinical adoption. Given the complex anatomy and pathology of the head and neck, deep learning models are ideally suited for learning such mapping.

MATERIALS AND METHODS

This was a retrospective study of 39 pediatric and adult patients with bone MR imaging and CT examinations of the head and neck. For each patient, MR imaging and CT data sets were spatially coregistered using multiple-point affine transformation. Paired MR imaging and CT slices were generated for model training, using 4-fold cross-validation. We trained 3 different encoder-decoder models: Light_U-Net (2 million parameters) and VGG-16 U-Net (29 million parameters) without and with transfer learning. Loss functions included mean absolute error, mean squared error, and a weighted average. Performance metrics included Pearson R, mean absolute error, mean squared error, bone precision, and bone recall. We investigated model generalizability by training and validating across different conditions.

RESULTS

The Light_U-Net architecture quantitatively outperformed VGG-16 models. Mean absolute error loss resulted in higher bone precision, while mean squared error yielded higher bone recall. Performance metrics decreased when using training data captured only in a different environment but increased when local training data were augmented with those from different hospitals, vendors, or MR imaging techniques.

CONCLUSIONS

We have optimized a robust deep learning model for conversion of bone MR imaging to synthetic CT, which shows good performance and generalizability when trained on different hospitals, vendors, and MR imaging techniques. This approach shows promise for facilitating downstream image processing and adoption into clinical practice.

Zero-TE MRI: principles and applications in the head and neck

Zero echo-time (ZTE) MRI is a novel imaging technique that utilizes ultrafast readouts to capture signal from short-T2 tissues. Additional sequence advantages include rapid imaging times, silent scanning, and artifact resistance. A robust application of this technology is imaging of cortical bone without the use of ionizing radiation, thus representing a viable alternative to CT for both rapid screening and "one-stop-shop" MRI. Although ZTE is increasingly used in musculoskeletal and body imaging, neuroimaging applications have historically been limited by complex anatomy and pathology. In this article, we review the imaging physics of ZTE including pulse sequence options, practical limitations, and image reconstruction. We then discuss optimization of settings for ZTE bone neuroimaging including acquisition, processing, segmentation, synthetic CT generation, and artifacts. Finally, we examine clinical utility of ZTE in the head and neck with imaging examples including malformations, trauma, tumors, and interventional procedures.

Quantitative evaluation of a deep learning-based framework to generate whole-body attenuation maps using LSO background radiation in long axial FOV PET scanners

Purpose

Attenuation correction is a critically important step in data correction in positron emission tomography (PET) image formation. The current standard method involves conversion of Hounsfield units from a computed tomography (CT) image to construct attenuation maps (µ-maps) at 511 keV. In this work, the increased sensitivity of long axial field-of-view (LAFOV) PET scanners was exploited to develop and evaluate a deep learning (DL) and joint reconstruction-based method to generate µ-maps utilizing background radiation from lutetium-based (LSO) scintillators.

Methods

Data from 18 subjects were used to train convolutional neural networks to enhance initial µ-maps generated using joint activity and attenuation reconstruction algorithm (MLACF) with transmission data from LSO background radiation acquired before and after the administration of ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) (µ-map_MLACF-PRE and µ-map_MLACF-POST respectively). The deep learning-enhanced µ-maps (µ-map_DL-MLACF-PRE and µ-map_{DL-MLACF-POST}) were compared against MLACF-derived and CT-based maps (µ-map_CT). The performance of the method was also evaluated by assessing PET images reconstructed using each µ-map and computing volume-of-interest based standard uptake value measurements and percentage relative mean error (rME) and relative mean absolute error (rMAE) relative to CT-based method.

Results

No statistically significant difference was observed in rME values for µ-map_DL-MLACF-PRE and µ-map_{DL-MLACF-POST} both in fat-based and water-based soft tissue as well as bones, suggesting that presence of the radiopharmaceutical activity in the body had negligible effects on the resulting µ-maps. The rMAE values µ-map_{DL-MLACF-POST} were reduced by a factor of 3.3 in average compared to the rMAE of µ-map_MLACF-POST. Similarly, the average rMAE values of PET images reconstructed using µ-map_{DL-MLACF-POST} (PET_{DL-MLACF-POST}) were 2.6 times smaller than the average rMAE values of PET images reconstructed using µ-map_MLACF-POST. The mean absolute errors in SUV values of PET_{DL-MLACF-POST} compared to PET_CT were less than 5% in healthy organs, less than 7% in brain grey matter and 4.3% for all tumours combined.

Conclusion

We describe a deep learning-based method to accurately generate µ-maps from PET emission data and LSO background radiation, enabling CT-free attenuation and scatter correction in LAFOV PET scanners.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00259-022-05909-3.

Evaluation of a 2D UNet-Based Attenuation Correction Methodology for PET/MR Brain Studies

Deep learning (DL) strategies applied to magnetic resonance (MR) images in positron emission tomography (PET)/MR can provide synthetic attenuation correction (AC) maps, and consequently PET images, more accurate than segmentation or atlas-registration strategies. As first objective, we aim to investigate the best MR image to be used and the best point of the AC pipeline to insert the synthetic map in. Sixteen patients underwent a 18F-fluorodeoxyglucose (FDG) PET/computed tomography (CT) and a PET/MR brain study in the same day. PET/CT images were reconstructed with attenuation maps obtained: (1) from CT (reference), (2) from MR with an atlas-based and a segmentation-based method and (3) with a 2D UNet trained on MR image/attenuation map pairs. As for MR, T1-weighted and Zero Time Echo (ZTE) images were considered; as for attenuation maps, CTs and 511 keV low-resolution attenuation maps were assessed. As second objective, we assessed the ability of DL strategies to provide proper AC maps in presence of cranial anatomy alterations due to surgery. Three 11C-methionine (METH) PET/MR studies were considered. PET images were reconstructed with attenuation maps obtained: (1) from diagnostic coregistered CT (reference), (2) from MR with an atlas-based and a segmentation-based method and (3) with 2D UNets trained on the sixteen FDG anatomically normal patients. Only UNets taking ZTE images in input were considered. FDG and METH PET images were quantitatively evaluated. As for anatomically normal FDG patients, UNet AC models generally provide an uptake estimate with lower bias than atlas-based or segmentation-based methods. The intersubject average bias on images corrected with UNet AC maps is always smaller than 1.5%, except for AC maps generated on too coarse grids. The intersubject bias variability is the lowest (always lower than 2%) for UNet AC maps coming from ZTE images, larger for other methods. UNet models working on MR ZTE images and generating synthetic CT or 511 keV low-resolution attenuation maps therefore provide the best results in terms of both accuracy and variability. As for METH anatomically altered patients, DL properly reconstructs anatomical alterations. Quantitative results on PET images confirm those found on anatomically normal FDG patients.

Composite attenuation correction method using a 68Ge-transmission multi-atlas for quantitative brain PET/MR

In positron emission tomography (PET), ⁶⁸Ge-transmission scanning is considered the gold standard in attenuation correction (AC) though not available in current dual imaging systems. In this experimental study we evaluated a novel AC method for PET/magnetic resonance (MR) imaging which is essentially based on a composite database of multiple ⁶⁸Ge-transmission maps and T1-weighted (T1w) MR image-pairs (composite transmission, CTR-AC). This proof-of-concept study used retrospectively a database with 125 pairs of co-registered ⁶⁸Ge-AC maps and T1w MR images from anatomical normal subjects and a validation dataset comprising dynamic [¹¹C]PE2I PET data from nine patients with Parkinsonism. CTR-AC maps were generated by non-rigid image registration of all database T1w MRI to each subject's T1w, applying the same transformation to every ⁶⁸Ge-AC map, and averaging the resulting ⁶⁸Ge-AC maps. [¹¹C]PE2I PET images were reconstructed using CTR-AC and a patient-specific ⁶⁸Ge-AC map as the reference standard. Standardized uptake values (SUV) and quantitative parameters of kinetic analysis were compared, i.e., relative delivery (R₁) and non-displaceable binding potential (BP_ND). CTR-AC showed high accuracy for whole-brain SUV (mean %bias ± SD: 0.5 ± 3.5%), whole-brain R₁ (-0.1 ± 3.2%), and putamen BP_ND (3.7 ± 8.1%). SUV and R₁ precision (SD of %bias) were modest and lowest in the anterior cortex, with an R₁ %bias of -1.1 ± 6.4%). The prototype CTR-AC is capable of providing accurate MRAC-maps with continuous linear attenuation coefficients though still experimental. The method's accuracy is comparable to the best MRAC methods published so far, both in SUV and as found for ZTE-AC in quantitative parameters of kinetic modelling.

Ultrashort Echo-Time MR Imaging of the Pediatric Head and Neck

Bone MR imaging techniques use extremely rapid echo times to maximize detection of short-T2 tissues with low water concentrations. The major approaches used in clinical practice are ultrashort echo-time and zero echo-time. Synthetic CT generation is feasible using atlas-based, voxel-based, and deep learning approaches. Major clinical applications in the pediatric head and neck include evaluation for craniosynostosis, sinonasal and jaw imaging, trauma, interventional planning, and postoperative follow-up. In this article, we review the technical background and practical usefulness of bone MR imaging with key imaging examples.

Direct Attenuation Correction Using Deep Learning for Cardiac SPECT: A Feasibility Study

Dedicated cardiac SPECT scanners with cadmium-zinc-telluride cameras have shown capabilities for shortened scan times or reduced radiation doses, as well as improved image quality. Since most dedicated scanners do not have integrated CT, image quantification with attenuation correction (AC) is challenging and artifacts are routinely encountered in daily clinical practice. In this work, we demonstrated a direct AC technique using deep learning (DL) for myocardial perfusion imaging (MPI). Methods: In an institutional review board-approved retrospective study, 100 cardiac SPECT/CT datasets with 99mTc-tetrofosmin, obtained using a scanner specifically with a small field of view, were collected at the Yale New Haven Hospital. A convolutional neural network was used for generating DL-based attenuation-corrected SPECT (SPECTDL) directly from noncorrected SPECT (SPECTNC) without undergoing an additional image reconstruction step. The accuracy of SPECTDL was evaluated by voxelwise and segmentwise analyses against the reference, CT-based AC (SPECTCTAC), using the 17-segment myocardial model of the American Heart Association. Polar maps of representative (best, median, and worst) cases were visually compared to illustrate potential benefits and pitfalls of the DL approach. Results: The voxelwise correlations with SPECTCTAC were 92.2% ± 3.7% (slope, 0.87; R2 = 0.81) and 97.7% ± 1.8% (slope, 0.94; R2 = 0.91) for SPECTNC and SPECTDL, respectively. The segmental errors of SPECTNC scattered from -35% to 21% (P < 0.001), whereas the errors of SPECTDL stayed mostly within ±10% (P < 0.001). The average segmental errors (mean ± SD) were -6.11% ± 8.06% and 0.49% ± 4.35% for SPECTNC and SPECTDL, respectively. The average absolute segmental errors were 7.96% ± 6.23% and 3.31% ± 2.87% for SPECTNC and SPECTDL, respectively. Review of polar maps revealed successful reduction of attenuation artifacts; however, the performance of SPECTDL was not consistent for all subjects, likely because of different amounts of attenuation and different uptake patterns. Conclusion: We demonstrated the feasibility of direct AC using DL for SPECT MPI. Overall, our DL approach reduced attenuation artifacts substantially compared with SPECTNC, justifying further studies to establish safety and consistency for clinical applications in stand-alone SPECT systems suffering from attenuation artifacts.

MRI-based synthetic CT of the lumbar spine: Geometric measurements for surgery planning in comparison with CT

PURPOSE

MRI is the imaging modality of choice for soft tissue-related spine disease. However, CT is superior to MRI in providing clear visualization of bony morphology. The purpose of this study is to test equivalency of MRI-based synthetic CT to conventional CT in quantitatively assessing bony morphology of the lumbar spine.

METHOD

A prospective study with an equivalency design was performed. Adult patients who had undergone MRI and CT of the lumbar spine were included. Synthetic CT images were generated from MRI using a deep learning-based image synthesis method. Two readers independently measured pedicle width, spinal canal width, neuroforamen length, anterior and posterior vertebral body height, superior and inferior vertebral body length, superior and inferior vertebral body width, maximal disc height, lumbar curvature and spinous process length on synthetic CT and CT. The agreement among CT and synthetic CT was evaluated using equivalency statistical testing.

RESULTS

Thirty participants were included (14 men and 16 women, range 20-60 years). The measurements performed on synthetic CT of pedicle width, spinal canal width, vertebral body height, vertebral body width, vertebral body length and spinous process length were statistically equivalent to CT measurements at the considered margins. Excellent inter- and intra-reader reliability was found for both synthetic CT and CT.

CONCLUSIONS

Equivalency of MRI-based synthetic CT to CT was demonstrated on geometrical measurements in the lumbar spine. In combination with the soft tissue information of the conventional MRI, this provides new possibilities in diagnosis and surgical planning without ionizing radiation.

Comparison of different deep learning architectures for synthetic CT generation from MR images

PURPOSE

Among the different available methods for synthetic CT generation from MR images for the task of MR-guided radiation planning, the deep learning algorithms have and do outperform their conventional counterparts. In this study, we investigated the performance of some most popular deep learning architectures including eCNN, U-Net, GAN, V-Net, and Res-Net for the task of sCT generation. As a baseline, an atlas-based method is implemented to which the results of the deep learning-based model are compared.

METHODS

A dataset consisting of 20 co-registered MR-CT pairs of the male pelvis is applied to assess the different sCT production methods' performance. The mean error (ME), mean absolute error (MAE), Pearson correlation coefficient (PCC), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) metrics were computed between the estimated sCT and the ground truth (reference) CT images.

RESULTS

The visual inspection revealed that the sCTs produced by eCNN, V-Net, and ResNet, unlike the other methods, were less noisy and greatly resemble the ground truth CT image. In the whole pelvis region, the eCNN yielded the lowest MAE (26.03 ± 8.85 HU) and ME (0.82 ± 7.06 HU), and the highest PCC metrics were yielded by the eCNN (0.93 ± 0.05) and ResNet (0.91 ± 0.02) methods. The ResNet model had the highest PSNR of 29.38 ± 1.75 among all models. In terms of the Dice similarity coefficient, the eCNN method revealed superior performance in major tissue identification (air, bone, and soft tissue).

CONCLUSIONS

All in all, the eCNN and ResNet deep learning methods revealed acceptable performance with clinically tolerable quantification errors.

Artificial Intelligence-Based Data Corrections for Attenuation and Scatter in Position Emission Tomography and Single-Photon Emission Computed Tomography

Recent developments in artificial intelligence (AI) technology have enabled new developments that can improve attenuation and scatter correction in PET and single-photon emission computed tomography (SPECT). These technologies will enable the use of accurate and quantitative imaging without the need to acquire a computed tomography image, greatly expanding the capability of PET/MR imaging, PET-only, and SPECT-only scanners. The use of AI to aid in scatter correction will lead to improvements in image reconstruction speed, and improve patient throughput. This article outlines the use of these new tools, surveys contemporary implementation, and discusses their limitations.

Evaluation of attenuation correction in PET/MRI with synthetic lesion insertion

Purpose: One major challenge facing simultaneous positron emission tomography (PET)/ magnetic resonance imaging (MRI) is PET attenuation correction (AC) measurement and evaluation of its accuracy. There is a crucial need for the evaluation of current and emergent PET AC methodologies in terms of absolute quantitative accuracy in the reconstructed PET images. Approach: To address this need, we developed and evaluated a lesion insertion tool for PET/MRI that will facilitate this evaluation process. This tool was developed for the Biograph mMR and evaluated using phantom and patient data. Contrast recovery coefficients (CRC) from the NEMA IEC phantom of synthesized lesions were compared to measurements. In addition, SUV biases of lesions inserted in human brain and pelvis images were assessed from PET images reconstructed with MRI-based AC (MRAC) and CT-based AC (CTAC). Results: For cross-comparison PET/MRI scanners AC evaluation, we demonstrated that the developed lesion insertion tool can be harmonized with the GE-SIGNA lesion insertion tool. About < 3 % CRC curves difference between simulation and measurement was achieved. An average of 1.6% between harmonized simulated CRC curves obtained with mMR and SIGNA lesion insertion tools was achieved. A range of - 5 % to 12% MRAC to CTAC SUV bias was respectively achieved in the vicinity and inside bone tissues in patient images in two anatomical regions, the brain, and pelvis. Conclusions: A lesion insertion tool was developed for the Biograph mMR PET/MRI scanner and harmonized with the SIGNA PET/MRI lesion insertion tool. These tools will allow for an accurate evaluation of different PET/MRI AC approaches and permit exploration of subtle attenuation correction differences across systems.

Deep learning based synthetic-CT generation in radiotherapy and PET: A review

Recently,deep learning (DL)-based methods for the generation of synthetic computed tomography (sCT) have received significant research attention as an alternative to classical ones. We present here a systematic review of these methods by grouping them into three categories, according to their clinical applications: (i) to replace computed tomography in magnetic resonance (MR) based treatment planning, (ii) facilitate cone-beam computed tomography based image-guided adaptive radiotherapy, and (iii) derive attenuation maps for the correction of positron emission tomography. Appropriate database searching was performed on journal articles published between January 2014 and December 2020. The DL methods' key characteristics were extracted from each eligible study, and a comprehensive comparison among network architectures and metrics was reported. A detailed review of each category was given, highlighting essential contributions, identifying specific challenges, and summarizing the achievements. Lastly, the statistics of all the cited works from various aspects were analyzed, revealing the popularity and future trends and the potential of DL-based sCT generation. The current status of DL-based sCT generation was evaluated, assessing the clinical readiness of the presented methods.

Application of PET Imaging in the Brain Regions of the Emotional Control Loop in Patients with Generalized Anxiety Disorder

Objective

This study uses PET imaging to observe the uptake and metabolism of 18F-fluorodeoxyglucose (¹⁸F-FDG) in the multibrain areas of the emotional control loop in patients with generalized anxiety disorder (GAD) and investigate the brain of GAD patient's functional abnormality mechanism.

Methods

The thesis clinically collected 20 GAD patients and 20 healthy subjects. Dynamic PET-CT scans were used. At the same time, ¹⁸F-FDG whole-brain uptake and metabolism data were collected. Image fusion and semiquantitative analysis were used to measure emotional control loops. The maximum standard uptake value (SUVmax) and dynamic uptake and metabolic changes of 11 time points in the brain area at 150 min were measured.

Results

Compared with the healthy control group, the peak uptake of the bilateral prefrontal cortex and the average uptake rate before the peak in GAD patients were significantly reduced (P < 0.05), and the average metabolic rate after the peak was significantly increased (P < 0.05). The peak uptake of the left striatum and the left hippocampus, the average uptake rate before the peak, and the average metabolic rate after the peak were all significantly reduced (P < 0.05); There were no obvious changes in the three indexes of the right striatum and the right hippocampus.

Conclusion

There are ¹⁸F-FDG uptake and metabolic disorders in multiple brain areas of the affective control loop of GAD patients. The abnormal peak and rate of uptake may be related to the pathogenesis of GAD.

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-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.

Development of attenuation correction methods using deep learning in brain-perfusion single-photon emission computed tomography

PURPOSE

Computed tomography (CT)-based attenuation correction (CTAC) in single-photon emission computed tomography (SPECT) is highly accurate, but it requires hybrid SPECT/CT instruments and additional radiation exposure. To obtain attenuation correction (AC) without the need for additional CT images, a deep learning method was used to generate pseudo-CT images has previously been reported, but it is limited because of cross-modality transformation, resulting in misalignment and modality-specific artifacts. This study aimed to develop a deep learning-based approach using non-attenuation-corrected (NAC) images and CTAC-based images for training to yield AC images in brain-perfusion SPECT. This study also investigated whether the proposed approach is superior to conventional Chang's AC (ChangAC).

METHODS

In total, 236 patients who underwent brain-perfusion SPECT were randomly divided into two groups: the training group (189 patients; 80%) and the test group (47 patients; 20%). Two models were constructed using Autoencoder (AutoencoderAC) and U-Net (U-NetAC), respectively. ChangAC, AutoencoderAC, and U-NetAC approaches were compared with CTAC using qualitative analysis (visual evaluation) and quantitative analysis (normalized mean squared error [NMSE] and the percentage error in each brain region). Statistical analyses were performed using the Wilcoxon signed-rank sum test and Bland-Altman analysis.

RESULTS

U-NetAC had the highest visual evaluation score. The NMSE results for the U-NetAC were the lowest, followed by AutoencoderAC and ChangAC (P < 0.001). Bland-Altman analysis showed a fixed bias for ChangAC and AutoencoderAC and a proportional bias for ChangAC. ChangAC underestimated counts by 30-40% in all brain regions. AutoencoderAC and U-NetAC produced mean errors of <1% and maximum errors of 3%, respectively.

CONCLUSION

New deep learning-based AC methods for AutoencoderAC and U-NetAC were developed. Their accuracy was higher than that obtained by ChangAC. U-NetAC exhibited higher qualitative and quantitative accuracy than AutoencoderAC. We generated highly accurate AC images directly from NAC images without the need for intermediate pseudo-CT images. To verify our models' generalizability, external validation is required.

Applications of artificial intelligence in nuclear medicine image generation

Recently, the application of artificial intelligence (AI) in medical imaging (including nuclear medicine imaging) has rapidly developed. Most AI applications in nuclear medicine imaging have focused on the diagnosis, treatment monitoring, and correlation analyses with pathology or specific gene mutation. It can also be used for image generation to shorten the time of image acquisition, reduce the dose of injected tracer, and enhance image quality. This work provides an overview of the application of AI in image generation for single-photon emission computed tomography (SPECT) and positron emission tomography (PET) either without or with anatomical information [CT or magnetic resonance imaging (MRI)]. This review focused on four aspects, including imaging physics, image reconstruction, image postprocessing, and internal dosimetry. AI application in generating attenuation map, estimating scatter events, boosting image quality, and predicting internal dose map is summarized and discussed.

Pitfalls on PET/MRI

A decade of PET/MRI clinical imaging has passed and many of the pitfalls are similar to those on earlier studies. However, techniques to overcome them have emerged and continue to develop. Although clinically significant lung nodules are demonstrable, smaller nodules may be detected using ultrashort/zero echo-time (TE) lung MRI. Fast reconstruction ultrashort TE sequences have also been used to achieve high-resolution lung MRI even with free-breathing. The introduction and improvement of time-of-flight scanners and increasing the axial length of the PET detector arrays have more than doubled the sensitivity of the PET part of the system. MRI for attenuation correction has provided many potential pitfalls, including misclassification of tissue classes based on MRI information for attenuation correction. Although the use of short echo times have helped to address these pitfalls, one of the most exciting developments has been the use of deep learning algorithms and computational neural networks to rapidly provide soft tissue, fat, bone and air information for the attenuation correction as a supplement to the attenuation correction information from fat-water imaging. Challenges with motion correction, particularly respiratory and cardiac remain but are being addressed with respiratory monitors and using PET data. In order to address truncation artefacts, the system manufacturers have developed methods to extend the MR field-of-view for the purpose of the attenuation and scatter corrections. General pitfalls like stitching of body sections for individual studies, optimum delivery of images for viewing and reporting, and resource implications for the sheer volume of data generated remain Methods to overcome these pitfalls serve as a strong foundation for the future of PET/MRI. Advances in the underlying technology with significant evolution in hard-ware and software and the exiting developments in use of deep learning algorithms and computational neural networks will drive the next decade of PET/MRI imaging.

Attenuation correction using deep Learning and integrated UTE/multi-echo Dixon sequence: evaluation in amyloid and tau PET imaging

PURPOSE

PET measures of amyloid and tau pathologies are powerful biomarkers for the diagnosis and monitoring of Alzheimer's disease (AD). Because cortical regions are close to bone, quantitation accuracy of amyloid and tau PET imaging can be significantly influenced by errors of attenuation correction (AC). This work presents an MR-based AC method that combines deep learning with a novel ultrashort time-to-echo (UTE)/multi-echo Dixon (mUTE) sequence for amyloid and tau imaging.

METHODS

Thirty-five subjects that underwent both 11C-PiB and 18F-MK6240 scans were included in this study. The proposed method was compared with Dixon-based atlas method as well as magnetization-prepared rapid acquisition with gradient echo (MPRAGE)- or Dixon-based deep learning methods. The Dice coefficient and validation loss of the generated pseudo-CT images were used for comparison. PET error images regarding standardized uptake value ratio (SUVR) were quantified through regional and surface analysis to evaluate the final AC accuracy.

RESULTS

The Dice coefficients of the deep learning methods based on MPRAGE, Dixon, and mUTE images were 0.84 (0.91), 0.84 (0.92), and 0.87 (0.94) for the whole-brain (above-eye) bone regions, respectively, higher than the atlas method of 0.52 (0.64). The regional SUVR error for the atlas method was around 6%, higher than the regional SUV error. The regional SUV and SUVR errors for all deep learning methods were below 2%, with mUTE-based deep learning method performing the best. As for the surface analysis, the atlas method showed the largest error (> 10%) near vertices inside superior frontal, lateral occipital, superior parietal, and inferior temporal cortices. The mUTE-based deep learning method resulted in the least number of regions with error higher than 1%, with the largest error (> 5%) showing up near the inferior temporal and medial orbitofrontal cortices.

CONCLUSION

Deep learning with mUTE can generate accurate AC for amyloid and tau imaging in PET/MR.

Advanced Magnetic Resonance Imaging of the Skull Base

Magnetic resonance (MR) imaging is a crucial tool for evaluation of the skull base, enabling characterization of complex anatomy by utilizing multiple image contrasts. Recent technical MR advances have greatly enhanced radiologists' capability to diagnose skull base pathology and help direct management. In this paper, we will summarize cutting-edge clinical and emerging research MR techniques for the skull base, including high-resolution, phase-contrast, diffusion, perfusion, vascular, zero echo-time, elastography, spectroscopy, chemical exchange saturation transfer, PET/MR, ultra-high-field, and 3D visualization. For each imaging technique, we provide a high-level summary of underlying technical principles accompanied by relevant literature review and clinical imaging examples.

MR Imaging in the 21st Century: Technical Innovation over the First Two Decades

Clinical MRI systems have continually improved over the years since their introduction in the 1980s. In MRI technical development, the developments in each MRI system component, including data acquisition, image reconstruction, and hardware systems, have impacted the others. Progress in each component has induced new technology development opportunities in other components. New technologies outside of the MRI field, for example, computer science, data processing, and semiconductors, have been immediately incorporated into MRI development, which resulted in innovative applications. With high performance computing and MR technology innovations, MRI can now provide large volumes of functional and anatomical image datasets, which are important tools in various research fields. MRI systems are now combined with other modalities, such as positron emission tomography (PET) or therapeutic devices. These hybrid systems provide additional capabilities.

In this review, MRI advances in the last two decades will be considered. We will discuss the progress of MRI systems, the enabling technology, established applications, current trends, and the future outlook.

Comparison of Supervised and Unsupervised Deep Learning Methods for Medical Image Synthesis between Computed Tomography and Magnetic Resonance Images

Cross-modality medical image synthesis between magnetic resonance (MR) images and computed tomography (CT) images has attracted increasing attention in many medical imaging area. Many deep learning methods have been used to generate pseudo-MR/CT images from counterpart modality images. In this study, we used U-Net and Cycle-Consistent Adversarial Networks (CycleGAN), which were typical networks of supervised and unsupervised deep learning methods, respectively, to transform MR/CT images to their counterpart modality. Experimental results show that synthetic images predicted by the proposed U-Net method got lower mean absolute error (MAE), higher structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR) in both directions of CT/MR synthesis, especially in synthetic CT image generation. Though synthetic images by the U-Net method has less contrast information than those by the CycleGAN method, the pixel value profile tendency of the synthetic images by the U-Net method is closer to the ground truth images. This work demonstrated that supervised deep learning method outperforms unsupervised deep learning method in accuracy for medical tasks of MR/CT synthesis.

Automatic attenuation map estimation from SPECT data only for brain perfusion scans using convolutional neural networks

In clinical brain SPECT, correction for photon attenuation in the patient is essential to obtain images which provide quantitative information on the regional activity concentration per unit volume (kBq.[Formula: see text]). This correction generally requires an attenuation map ([Formula: see text] map) denoting the attenuation coefficient at each voxel which is often derived from a CT or MRI scan. However, such an additional scan is not always available and the method may suffer from registration errors. Therefore, we propose a SPECT-only-based strategy for [Formula: see text] map estimation that we apply to a stationary multi-pinhole clinical SPECT system (G-SPECT-I) for ^99mTc-HMPAO brain perfusion imaging. The method is based on the use of a convolutional neural network (CNN) and was validated with Monte Carlo simulated scans. Data acquired in list mode was used to employ the energy information of both primary and scattered photons to obtain information about the tissue attenuation as much as possible. Multiple SPECT reconstructions were performed from different energy windows over a large energy range. Locally extracted 4D SPECT patches (three spatial plus one energy dimension) were used as input for the CNN which was trained to predict the attenuation coefficient of the corresponding central voxel of the patch. Results show that Attenuation Correction using the Ground Truth [Formula: see text] maps (GT-AC) or using the CNN estimated [Formula: see text] maps (CNN-AC) achieve comparable accuracy. This was confirmed by a visual assessment as well as a quantitative comparison; the mean deviation from the GT-AC when using the CNN-AC is within 1.8% for the standardized uptake values in all brain regions. Therefore, our results indicate that a CNN-based method can be an automatic and accurate tool for SPECT attenuation correction that is independent of attenuation data from other imaging modalities or human interpretations about head contours.

CT-less Direct Correction of Attenuation and Scatter in the Image Space Using Deep Learning for Whole-Body FDG PET: Potential Benefits and Pitfalls

PURPOSE

To demonstrate the feasibility of CT-less attenuation and scatter correction (ASC) in the image space using deep learning for whole-body PET, with a focus on the potential benefits and pitfalls.

MATERIALS AND METHODS

In this retrospective study, 110 whole-body fluorodeoxyglucose (FDG) PET/CT studies acquired in 107 patients (mean age ± standard deviation, 58 years ± 18; age range, 11-92 years; 72 females) from February 2016 through January 2018 were randomly collected. A total of 37.3% (41 of 110) of the studies showed metastases, with diverse FDG PET findings throughout the whole body. A U-Net-based network was developed for directly transforming noncorrected PET (PETNC) into attenuation- and scatter-corrected PET (PETASC). Deep learning-corrected PET (PETDL) images were quantitatively evaluated by using the standardized uptake value (SUV) of the normalized root mean square error, the peak signal-to-noise ratio, and the structural similarity index, in addition to a joint histogram for statistical analysis. Qualitative reviews by radiologists revealed the potential benefits and pitfalls of this correction method.

RESULTS

The normalized root mean square error (0.21 ± 0.05 [mean SUV ± standard deviation]), mean peak signal-to-noise ratio (36.3 ± 3.0), mean structural similarity index (0.98 ± 0.01), and voxelwise correlation (97.62%) of PETDL demonstrated quantitatively high similarity with PETASC. Radiologist reviews revealed the overall quality of PETDL. The potential benefits of PETDL include a radiation dose reduction on follow-up scans and artifact removal in the regions with attenuation correction- and scatter correction-based artifacts. The pitfalls involve potential false-negative results due to blurring or missing lesions or false-positive results due to pseudo-low-uptake patterns.

CONCLUSION

Deep learning-based direct ASC at whole-body PET is feasible and potentially can be used to overcome the current limitations of CT-based approaches, benefiting patients who are sensitive to radiation from CT.Supplemental material is available for this article.© RSNA, 2020.

MR-based Attenuation Correction for Brain PET Using 3D Cycle-Consistent Adversarial Network

Attenuation correction (AC) is important for the quantitative merits of positron emission tomography (PET). However, attenuation coefficients cannot be derived from magnetic resonance (MR) images directly for PET/MR systems. In this work, we aimed to derive continuous AC maps from Dixon MR images without the requirement of MR and computed tomography (CT) image registration. To achieve this, a 3D generative adversarial network with both discriminative and cycle-consistency loss (Cycle-GAN) was developed. The modified 3D U-net was employed as the structure of the generative networks to generate the pseudo CT/MR images. The 3D patch-based discriminative networks were used to distinguish the generated pseudo CT/MR images from the true CT/MR images. To evaluate its performance, datasets from 32 patients were used in the experiment. The Dixon segmentation and atlas methods provided by the vendor and the convolutional neural network (CNN) method which utilized registered MR and CT images were employed as the reference methods. Dice coefficients of the pseudo-CT image and the regional quantification in the reconstructed PET images were compared. Results show that the Cycle-GAN framework can generate better AC compared to the Dixon segmentation and atlas methods, and shows comparable performance compared to the CNN method.

DiCyc: GAN-based deformation invariant cross-domain information fusion for medical image synthesis

Cycle-consistent generative adversarial network (CycleGAN) has been widely used for cross-domain medical image synthesis tasks particularly due to its ability to deal with unpaired data. However, most CycleGAN-based synthesis methods cannot achieve good alignment between the synthesized images and data from the source domain, even with additional image alignment losses. This is because the CycleGAN generator network can encode the relative deformations and noises associated to different domains. This can be detrimental for the downstream applications that rely on the synthesized images, such as generating pseudo-CT for PET-MR attenuation correction. In this paper, we present a deformation invariant cycle-consistency model that can filter out these domain-specific deformation. The deformation is globally parameterized by thin-plate-spline (TPS), and locally learned by modified deformable convolutional layers. Robustness to domain-specific deformations has been evaluated through experiments on multi-sequence brain MR data and multi-modality abdominal CT and MR data. Experiment results demonstrated that our method can achieve better alignment between the source and target data while maintaining superior image quality of signal compared to several state-of-the-art CycleGAN-based methods.

A review on medical imaging synthesis using deep learning and its clinical applications

This paper reviewed the deep learning-based studies for medical imaging synthesis and its clinical application. Specifically, we summarized the recent developments of deep learning-based methods in inter- and intra-modality image synthesis by listing and highlighting the proposed methods, study designs, and reported performances with related clinical applications on representative studies. The challenges among the reviewed studies were then summarized with discussion.

Accuracy and precision of zero-echo-time, single- and multi-atlas attenuation correction for dynamic [11C]PE2I PET-MR brain imaging

BACKGROUND

A valid photon attenuation correction (AC) method is instrumental for obtaining quantitatively correct PET images. Integrated PET/MR systems provide no direct information on attenuation, and novel methods for MR-based AC (MRAC) are still under investigation. Evaluations of various AC methods have mainly focused on static brain PET acquisitions. In this study, we determined the validity of three MRAC methods in a dynamic PET/MR study of the brain.

METHODS

Nine participants underwent dynamic brain PET/MR scanning using the dopamine transporter radioligand [11C]PE2I. Three MRAC methods were evaluated: single-atlas (Atlas), multi-atlas (MaxProb) and zero-echo-time (ZTE). The 68Ge-transmission data from a previous stand-alone PET scan was used as reference method. Parametric relative delivery (R1) images and binding potential (BPND) maps were generated using cerebellar grey matter as reference region. Evaluation was based on bias in MRAC maps, accuracy and precision of [11C]PE2I BPND and R1 estimates, and [11C]PE2I time-activity curves. BPND was examined for striatal regions and R1 in clusters of regions across the brain.

RESULTS

For BPND, ZTE-MRAC showed the highest accuracy (bias < 2%) in striatal regions. Atlas-MRAC exhibited a significant bias in caudate nucleus (- 12%) while MaxProb-MRAC revealed a substantial, non-significant bias in the putamen (9%). R1 estimates had a marginal bias for all MRAC methods (- 1.0-3.2%). MaxProb-MRAC showed the largest intersubject variability for both R1 and BPND. Standardized uptake values (SUV) of striatal regions displayed the strongest average bias for ZTE-MRAC (~ 10%), although constant over time and with the smallest intersubject variability. Atlas-MRAC had highest variation in bias over time (+10 to - 10%), followed by MaxProb-MRAC (+5 to - 5%), but MaxProb showed the lowest mean bias. For the cerebellum, MaxProb-MRAC showed the highest variability while bias was constant over time for Atlas- and ZTE-MRAC.

CONCLUSIONS

Both Maxprob- and ZTE-MRAC performed better than Atlas-MRAC when using a 68Ge transmission scan as reference method. Overall, ZTE-MRAC showed the highest precision and accuracy in outcome parameters of dynamic [11C]PE2I PET analysis with use of kinetic modelling.

Comparison of deep learning synthesis of synthetic CTs using clinical MRI inputs

There has been substantial interest in developing techniques for synthesizing CT-like images from MRI inputs, with important applications in simultaneous PET/MR and radiotherapy planning. Deep learning has recently shown great potential for solving this problem. The goal of this research was to investigate the capability of four common clinical MRI sequences (T1-weighted gradient-echo [T1], T2-weighted fat-suppressed fast spin-echo [T2-FatSat], post-contrast T1-weighted gradient-echo [T1-Post], and fast spin-echo T2-weighted fluid-attenuated inversion recovery [CUBE-FLAIR]) as inputs into a deep CT synthesis pipeline. Data were obtained retrospectively in 92 subjects who had undergone an MRI and CT scan on the same day. The patient's MR and CT scans were registered to one another using affine registration. The deep learning model was a convolutional neural network encoder-decoder with skip connections similar to the U-net architecture and Inception V3 inspired blocks instead of sequential convolution blocks. After training with 150 epochs and a batch size of 6, the model was evaluated using structural similarity index (SSIM), peak SNR (PSNR), mean absolute error (MAE), and dice coefficient. We found that feasible results were attainable for each image type, and no single image type was superior for all analyses. The MAE (in HU) of the resulting synthesized CT in the whole brain was 51.236 ± 4.504 for CUBE-FLAIR, 45.432 ± 8.517 for T1, 44.558 ± 7.478 for T1-Post, and 45.721 ± 8.7767 for T2, showing not only feasible, but also very compelling results on clinical images. Deep learning-based synthesis of CT images from MRI is possible with a wide range of inputs, suggesting that viable images can be created from a wide range of clinical input types.

Deep learning for brain disorders: from data processing to disease treatment

In order to reach precision medicine and improve patients' quality of life, machine learning is increasingly used in medicine. Brain disorders are often complex and heterogeneous, and several modalities such as demographic, clinical, imaging, genetics and environmental data have been studied to improve their understanding. Deep learning, a subpart of machine learning, provides complex algorithms that can learn from such various data. It has become state of the art in numerous fields, including computer vision and natural language processing, and is also growingly applied in medicine. In this article, we review the use of deep learning for brain disorders. More specifically, we identify the main applications, the concerned disorders and the types of architectures and data used. Finally, we provide guidelines to bridge the gap between research studies and clinical routine.

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.

Potentials and caveats of AI in hybrid imaging

State-of-the-art patient management frequently mandates the investigation of both anatomy and physiology of the patients. Hybrid imaging modalities such as the PET/MRI, PET/CT and SPECT/CT have the ability to provide both structural and functional information of the investigated tissues in a single examination. With the introduction of such advanced hardware fusion, new problems arise such as the exceedingly large amount of multi-modality data that requires novel approaches of how to extract a maximum of clinical information from large sets of multi-dimensional imaging data. Artificial intelligence (AI) has emerged as one of the leading technologies that has shown promise in facilitating highly integrative analysis of multi-parametric data. Specifically, the usefulness of AI algorithms in the medical imaging field has been heavily investigated in the realms of (1) image acquisition and reconstruction, (2) post-processing and (3) data mining and modelling. Here, we aim to provide an overview of the challenges encountered in hybrid imaging and discuss how AI algorithms can facilitate potential solutions. In addition, we highlight the pitfalls and challenges in using advanced AI algorithms in the context of hybrid imaging and provide suggestions for building robust AI solutions that enable reproducible and transparent research.