Deep learning-based attenuation map generation for myocardial perfusion SPECT

Artificial Intelligence in Nuclear Cardiology: An Update and Future Trends

Myocardial perfusion imaging (MPI), using either single photon emission computed tomography (SPECT) or positron emission tomography (PET), is one of the most commonly ordered cardiac imaging tests, with prominent clinical roles for disease diagnosis and risk prediction. Artificial intelligence (AI) could potentially play a role in many steps along the typical MPI workflow, from image acquisition through to clinical reporting and risk estimation. AI can be utilized to improve image quality, reducing radiation exposure and image acquisition times. Once images are acquired, AI can help optimize motion correction and image registration during image reconstruction or provide direct image attenuation correction. Utilizing these image sets, AI can segment a number of anatomic features from associated computed tomographic imaging or even generate synthetic attenuation imaging. Lastly, AI may play an important role in disease diagnosis or risk prediction by combining the large number of potentially important clinical, stress, and imaging-related variables. This review will focus on the most recent developments in the field, providing clinicians and researchers with a timely update on the field. Additionally, it will discuss future trends including applications of AI during multiple points of the typical MPI workflow to maximize clinical utility and methods to maximize the information that can be obtained from hybrid imaging.

TAI-GAN: A Temporally and Anatomically Informed Generative Adversarial Network for early-to-late frame conversion in dynamic cardiac PET inter-frame motion correction

Inter-frame motion in dynamic cardiac positron emission tomography (PET) using rubidium-82 (82Rb) myocardial perfusion imaging impacts myocardial blood flow (MBF) quantification and the diagnosis accuracy of coronary artery diseases. However, the high cross-frame distribution variation due to rapid tracer kinetics poses a considerable challenge for inter-frame motion correction, especially for early frames where intensity-based image registration techniques often fail. To address this issue, we propose a novel method called Temporally and Anatomically Informed Generative Adversarial Network (TAI-GAN) that utilizes an all-to-one mapping to convert early frames into those with tracer distribution similar to the last reference frame. The TAI-GAN consists of a feature-wise linear modulation layer that encodes channel-wise parameters generated from temporal information and rough cardiac segmentation masks with local shifts that serve as anatomical information. Our proposed method was evaluated on a clinical 82Rb PET dataset, and the results show that our TAI-GAN can produce converted early frames with high image quality, comparable to the real reference frames. After TAI-GAN conversion, the motion estimation accuracy and subsequent myocardial blood flow (MBF) quantification with both conventional and deep learning-based motion correction methods were improved compared to using the original frames. The code is available at https://github.com/gxq1998/TAI-GAN.

Gated SPECT-Derived Myocardial Strain Estimated From Deep-Learning Image Translation Validated From N-13 Ammonia PET

RATIONALE AND OBJECTIVES

This study investigated the use of deep learning-generated virtual positron emission tomography (PET)-like gated single-photon emission tomography (SPECTVP) for assessing myocardial strain, overcoming limitations of conventional SPECT.

MATERIALS AND METHODS

SPECT-to-PET translation models for short-axis, horizontal, and vertical long-axis planes were trained using image pairs from the same patients in stress (720 image pairs from 18 patients) and resting states (920 image pairs from 23 patients). Patients without ejection-fraction changes during SPECT and PET were selected for training. We independently analyzed circumferential strains from short-axis-gated SPECT, PET, and model-generated SPECTVP images using a feature-tracking algorithm. Longitudinal strains were similarly measured from horizontal and vertical long-axis images. Intraclass correlation coefficients (ICCs) were calculated with two-way random single-measure SPECT and SPECTVP (PET). ICCs (95% confidence intervals) were defined as excellent (≥0.75), good (0.60-0.74), moderate (0.40-0.59), or poor (≤0.39).

RESULTS

Moderate ICCs were observed for SPECT-derived stressed circumferential strains (0.56 [0.41-0.69]). Excellent ICCs were observed for SPECTVP-derived stressed circumferential strains (0.78 [0.68-0.85]). Excellent ICCs of stressed longitudinal strains from horizontal and vertical long axes, derived from SPECT and SPECTVP, were observed (0.83 [0.73-0.90], 0.91 [0.85-0.94]).

CONCLUSION

Deep-learning SPECT-to-PET transformation improves circumferential strain measurement accuracy using standard-gated SPECT. Furthermore, the possibility of applying longitudinal strain measurements via both PET and SPECTVP was demonstrated. This study provides preliminary evidence that SPECTVP obtained from standard-gated SPECT with postprocessing potentially adds clinical value through PET-equivalent myocardial strain analysis without increasing the patient burden.

Current status and future directions in artificial intelligence for nuclear cardiology

INTRODUCTION

Myocardial perfusion imaging (MPI) is one of the most commonly ordered cardiac imaging tests. Accurate motion correction, image registration, and reconstruction are critical for high-quality imaging, but this can be technically challenging and has traditionally relied on expert manual processing. With accurate processing, there is a rich variety of clinical, stress, functional, and anatomic data that can be integrated to guide patient management.

AREAS COVERED

PubMed and Google Scholar were reviewed for articles related to artificial intelligence in nuclear cardiology published between 2020 and 2024. We will outline the prominent roles for artificial intelligence (AI) solutions to provide motion correction, image registration, and reconstruction. We will review the role for AI in extracting anatomic data for hybrid MPI which is otherwise neglected. Lastly, we will discuss AI methods to integrate the wealth of data to improve disease diagnosis or risk stratification.

EXPERT OPINION

There is growing evidence that AI will transform the performance of MPI by automating and improving on aspects of image acquisition and reconstruction. Physicians and researchers will need to understand the potential strengths of AI in order to benefit from the full clinical utility of MPI.

Clinical Feasibility of Deep Learning-Based Attenuation Correction Models for Tl-201 Myocardial Perfusion SPECT

PURPOSE

We aimed to develop deep learning (DL)-based attenuation correction models for Tl-201 myocardial perfusion SPECT (MPS) images and evaluate their clinical feasibility.

PATIENTS AND METHODS

We conducted a retrospective study of patients with suspected or known coronary artery disease. We proposed a DL-based image-to-image translation technique to transform non-attenuation-corrected images into CT-based attenuation-corrected (CT AC ) images. The model was trained using a modified U-Net with structural similarity index (SSIM) loss and mean squared error (MSE) loss and compared with other models. Segment-wise analysis using a polar map and visual assessment for the generated attenuation-corrected (GEN AC ) images were also performed to evaluate clinical feasibility.

RESULTS

This study comprised 657 men and 328 women (age, 65 ± 11 years). Among the various models, the modified U-Net achieved the highest performance with an average mean absolute error of 0.003, an SSIM of 0.990, and a peak signal-to-noise ratio of 33.658. The performance of the model was not different between the stress and rest datasets. In the segment-wise analysis, the myocardial perfusion of the inferior wall was significantly higher in GEN AC images than in the non-attenuation-corrected images in both the rest and stress test sets ( P < 0.05). In the visual assessment of patients with diaphragmatic attenuation, scores of 4 (similar to CT AC images) or 5 (indistinguishable from CT AC images) were assigned to most GEN AC images (65/68).

CONCLUSIONS

Our clinically feasible DL-based attenuation correction models can replace the CT-based method in Tl-201 MPS, and it would be useful in case SPECT/CT is unavailable for MPS.

Deep learning approach using SPECT-to-PET translation for attenuation correction in CT-less myocardial perfusion SPECT imaging

OBJECTIVE

Deep learning approaches have attracted attention for improving the scoring accuracy in computed tomography-less single photon emission computed tomography (SPECT). In this study, we proposed a novel deep learning approach referring to positron emission tomography (PET). The aims of this study were to analyze the agreement of representative voxel values and perfusion scores of SPECT-to-PET translation model-generated SPECT (SPECTSPT) against PET in 17 segments according to the American Heart Association (AHA).

METHODS

This retrospective study evaluated the patient-to-patient stress, resting SPECT, and PET datasets of 71 patients. The SPECTSPT generation model was trained (stress: 979 image pairs, rest: 987 image pairs) and validated (stress: 421 image pairs, rest: 425 image pairs) using 31 cases of SPECT and PET image pairs using an image-to-image translation network. Forty of 71 cases of left ventricular base-to-apex short-axis images were translated to SPECTSPT in the stress and resting state (stress: 1830 images, rest: 1856 images). Representative voxel values of SPECT and SPECTSPT in the 17 AHA segments against PET were compared. The stress, resting, and difference scores of 40 cases of SPECT and SPECTSPT were also compared in each of the 17 segments.

RESULTS

For AHA 17-segment-wise analysis, stressed SPECT but not SPECTSPT voxel values showed significant error from PET at basal anterior regions (segments #1, #6), and at mid inferoseptal regions (segments #8, #9, and #10). SPECT, but not SPECTSPT, voxel values at resting state showed significant error at basal anterior regions (segments #1, #2, and #6), and at mid inferior regions (segments #8, #9, and #11). Significant SPECT overscoring was observed against PET in basal-to-apical inferior regions (segments #4, #10, and #15) during stress. No significant overscoring was observed in SPECTSPT at stress, and only moderate over and underscoring in the basal inferior region (segment #4) was found in the resting and difference states.

CONCLUSIONS

Our PET-supervised deep learning model is a new approach to correct well-known inferior wall attenuation in SPECT myocardial perfusion imaging. As standalone SPECT systems are used worldwide, the SPECTSPT generation model may be applied as a low-cost and practical clinical tool that provides powerful auxiliary information for the diagnosis of myocardial blood flow.

Short-axis PET image quality improvement based on a uEXPLORER total-body PET system through deep learning

PURPOSE

The axial field of view (AFOV) of a positron emission tomography (PET) scanner greatly affects the quality of PET images. Although a total-body PET scanner (uEXPLORER) with a large AFOV is more sensitive, it is more expensive and difficult to widely use. Therefore, we attempt to utilize high-quality images generated by uEXPLORER to optimize the quality of images from short-axis PET scanners through deep learning technology while controlling costs.

METHODS

The experiments were conducted using PET images of three anatomical locations (brain, lung, and abdomen) from 335 patients. To simulate PET images from different axes, two protocols were used to obtain PET image pairs (each patient was scanned once). For low-quality PET (LQ-PET) images with a 320-mm AFOV, we applied a 300-mm FOV for brain reconstruction and a 500-mm FOV for lung and abdomen reconstruction. For high-quality PET (HQ-PET) images, we applied a 1940-mm AFOV during the reconstruction process. A 3D Unet was utilized to learn the mapping relationship between LQ-PET and HQ-PET images. In addition, the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) were employed to evaluate the model performance. Furthermore, two nuclear medicine doctors evaluated the image quality based on clinical readings.

RESULTS

The generated PET images of the brain, lung, and abdomen were quantitatively and qualitatively compatible with the HQ-PET images. In particular, our method achieved PSNR values of 35.41 ± 5.45 dB (p < 0.05), 33.77 ± 6.18 dB (p < 0.05), and 38.58 ± 7.28 dB (p < 0.05) for the three beds. The overall mean SSIM was greater than 0.94 for all patients who underwent testing. Moreover, the total subjective quality levels of the generated PET images for three beds were 3.74 ± 0.74, 3.69 ± 0.81, and 3.42 ± 0.99 (the highest possible score was 5, and the minimum score was 1) from two experienced nuclear medicine experts. Additionally, we evaluated the distribution of quantitative standard uptake values (SUV) in the region of interest (ROI). Both the SUV distribution and the peaks of the profile show that our results are consistent with the HQ-PET images, proving the superiority of our approach.

CONCLUSION

The findings demonstrate the potential of the proposed technique for improving the image quality of a PET scanner with a 320 mm or even shorter AFOV. Furthermore, this study explored the potential of utilizing uEXPLORER to achieve improved short-axis PET image quality at a limited economic cost, and computer-aided diagnosis systems that are related can help patients and radiologists.

Artificial Intelligence Empowered Nuclear Medicine and Molecular Imaging in Cardiology: A State-of-the-Art Review

Nuclear medicine and molecular imaging plays a significant role in the detection and management of cardiovascular disease (CVD). With recent advancements in computer power and the availability of digital archives, artificial intelligence (AI) is rapidly gaining traction in the field of medical imaging, including nuclear medicine and molecular imaging. However, the complex and time-consuming workflow and interpretation involved in nuclear medicine and molecular imaging, limit their extensive utilization in clinical practice. To address this challenge, AI has emerged as a fundamental tool for enhancing the role of nuclear medicine and molecular imaging. It has shown promising applications in various crucial aspects of nuclear cardiology, such as optimizing imaging protocols, facilitating data processing, aiding in CVD diagnosis, risk classification and prognosis. In this review paper, we will introduce the key concepts of AI and provide an overview of its current progress in the field of nuclear cardiology. In addition, we will discuss future perspectives for AI in this domain.

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.

Deep learning approximation of attenuation maps for myocardial perfusion SPECT with an IQ[Formula: see text]SPECT collimator

BACKGROUND

The use of CT images for attenuation correction of myocardial perfusion imaging with single photon emission computer tomography (SPECT) increases diagnostic confidence. However, acquiring a CT image registered to a SPECT image is often not possible because most scanners are SPECT-only. It is possible to approximate attenuation maps using deep learning, but this has not yet been shown to work for a SPECT scanner with an IQ[Formula: see text]SPECT collimator. This study investigates whether it is possible to approximate attenuation maps from non-attenuation-corrected (nAC) reconstructions acquired with a SPECT scanner equipped with an IQ[Formula: see text]SPECT collimator.

METHODS

Attenuation maps and reconstructions were acquired retrospectively for 150 studies. A U-Net was trained to predict attenuation maps from nAC reconstructions using the conditional generative adversarial network framework. Predicted attenuation maps are compared to real attenuation maps using the normalized mean absolute error (NMAE). Attenuation-corrected reconstructions were computed, and the resulting polar maps were compared by pixel and by average perfusion per segment using the absolute percent error (APE). The training and evaluation code is available at https://gitlab.ub.uni-bielefeld.de/thuxohl/mu-map .

RESULTS

Predicted attenuation maps are similar to real attenuation maps, achieving an NMAE of 0.020±0.007. The same is true for polar maps generated from reconstructions with predicted attenuation maps compared to CT-based attenuation maps. Their pixel-wise absolute distance is 3.095±3.199, and the segment-wise APE is 1.155±0.769.

CONCLUSIONS

It is feasible to approximate attenuation maps from nAC reconstructions acquired by a scanner with an IQ[Formula: see text]SPECT collimator using deep learning.

Generative adversarial network-based attenuation correction for 99mTc-TRODAT-1 brain SPECT

Background

Attenuation correction (AC) is an important correction method to improve the quantification accuracy of dopamine transporter (DAT) single photon emission computed tomography (SPECT). Chang's method was developed for AC (Chang-AC) when CT-based AC was not available, assuming uniform attenuation coefficients inside the body contour. This study aims to evaluate Chang-AC and different deep learning (DL)-based AC approaches on 99mTc-TRODAT-1 brain SPECT using clinical patient data on two different scanners.

Methods

Two hundred and sixty patients who underwent 99mTc-TRODAT-1 SPECT/CT scans from two different scanners (scanner A and scanner B) were retrospectively recruited. The ordered-subset expectation-maximization (OS-EM) method reconstructed 120 projections with dual-energy scatter correction, with or without CT-AC. We implemented a 3D conditional generative adversarial network (cGAN) for the indirect deep learning-based attenuation correction (DL-ACμ) and direct deep learning-based attenuation correction (DL-AC) methods, estimating attenuation maps (μ-maps) and attenuation-corrected SPECT images from non-attenuation-corrected (NAC) SPECT, respectively. We further applied cross-scanner training (cross-scanner indirect deep learning-based attenuation correction [cull-ACμ] and cross-scanner direct deep learning-based attenuation correction [call-AC]) and merged the datasets from two scanners for ensemble training (ensemble indirect deep learning-based attenuation correction [eDL-ACμ] and ensemble direct deep learning-based attenuation correction [eDL-AC]). The estimated μ-maps from (c/e)DL-ACμ were then used in reconstruction for AC purposes. Chang's method was also implemented for comparison. Normalized mean square error (NMSE), structural similarity index (SSIM), specific uptake ratio (SUR), and asymmetry index (%ASI) of the striatum were calculated for different AC methods.

Results

The NMSE for Chang's method, DL-ACμ, DL-AC, cDL-ACμ, cDL-AC, eDL-ACμ, and eDL-AC is 0.0406 ± 0.0445, 0.0059 ± 0.0035, 0.0099 ± 0.0066, 0.0253 ± 0.0102, 0.0369 ± 0.0124, 0.0098 ± 0.0035, and 0.0162 ± 0.0118 for scanner A and 0.0579 ± 0.0146, 0.0055 ± 0.0034, 0.0063 ± 0.0028, 0.0235 ± 0.0085, 0.0349 ± 0.0086, 0.0115 ± 0.0062, and 0.0117 ± 0.0038 for scanner B, respectively. The SUR and %ASI results for DL-ACμ are closer to CT-AC, Followed by DL-AC, eDL-ACμ, cDL-ACμ, cDL-AC, eDL-AC, Chang's method, and NAC.

Conclusion

All DL-based AC methods are superior to Chang-AC. DL-ACμ is superior to DL-AC. Scanner-specific training is superior to cross-scanner and ensemble training. DL-based AC methods are feasible and robust for 99mTc-TRODAT-1 brain SPECT.

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.

Deep-learning-based estimation of attenuation map improves attenuation correction performance over direct attenuation estimation for myocardial perfusion SPECT

BACKGROUND

Deep learning (DL)-based attenuation correction (AC) is promising to improve myocardial perfusion (MP) SPECT. We aimed to optimize and compare the DL-based direct and indirect AC methods, with and without SPECT and CT mismatch.

METHODS

One hundred patients with different 99mTc-sestamibi activity distributions and anatomical variations were simulated by a population of XCAT phantoms. Additionally, 34 patients 99mTc-sestamibi stress/rest SPECT/CT scans were retrospectively recruited. Projections were reconstructed by OS-EM method with or without AC. Mismatch between SPECT and CT images was modeled. A 3D conditional generative adversarial network (cGAN) was optimized for two DL-based AC methods: (i) indirect approach, i.e., non-attenuation corrected (NAC) SPECT paired with the corresponding attenuation map for training. The projections were reconstructed with the DL-generated attenuation map for AC; (ii) direct approach, i.e., NAC SPECT paired with the corresponding AC SPECT for training to perform direct AC.

RESULTS

Mismatch between SPECT and CT degraded DL-based AC performance. The indirect approach is superior to direct approach for various physical and clinical indices, even with mismatch modeled.

CONCLUSION

DL-based estimation of attenuation map for AC is superior and more robust to direct generation of AC SPECT.

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.

Automated nonlinear registration of coronary PET to CT angiography using pseudo-CT generated from PET with generative adversarial networks

BACKGROUND

Coronary 18F-sodium-fluoride (18F-NaF) positron emission tomography (PET) showed promise in imaging coronary artery disease activity. Currently image processing remains subjective due to the need for manual registration of PET and computed tomography (CT) angiography data. We aimed to develop a novel fully automated method to register coronary 18F-NaF PET to CT angiography using pseudo-CT generated by generative adversarial networks (GAN).

METHODS

A total of 169 patients, 139 in the training and 30 in the testing sets were considered for generation of pseudo-CT from non-attenuation corrected (NAC) PET using GAN. Non-rigid registration was used to register pseudo-CT to CT angiography and the resulting transformation was used to align PET with CT angiography. We compared translations, maximal standard uptake value (SUVmax) and target to background ratio (TBRmax) at the location of plaques, obtained after observer and automated alignment.

RESULTS

Automatic end-to-end registration was performed for 30 patients with 88 coronary vessels and took 27.5 seconds per patient. Difference in displacement motion vectors between GAN-based and observer-based registration in the x-, y-, and z-directions was 0.8 ± 3.0, 0.7 ± 3.0, and 1.7 ± 3.9 mm, respectively. TBRmax had a coefficient of repeatability (CR) of 0.31, mean bias of 0.03 and narrow limits of agreement (LOA) (95% LOA: - 0.29 to 0.33). SUVmax had CR of 0.26, mean bias of 0 and narrow LOA (95% LOA: - 0.26 to 0.26).

CONCLUSION

Pseudo-CT generated by GAN are perfectly registered to PET can be used to facilitate quick and fully automated registration of PET and CT angiography.

Preparing for the Artificial Intelligence Revolution in Nuclear Cardiology

A major opportunity in nuclear cardiology is the many significant artificial intelligence (AI) applications that have recently been reported. These developments include using deep learning (DL) for reducing the needed injected dose and acquisition time in perfusion acquisitions also due to DL improvements in image reconstruction and filtering, SPECT attenuation correction using DL without need for transmission images, DL and machine learning (ML) use for feature extraction to define myocardial left ventricular (LV) borders for functional measurements and improved detection of the LV valve plane and AI, ML, and DL implementations for MPI diagnosis, prognosis, and structured reporting. Although some have, most of these applications have yet to make it to widespread commercial distribution due to the recency of their developments, most reported in 2020. We must be prepared both technically and socio-economically to fully benefit from these and a tsunami of other AI applications that are coming.

CT-free quantitative SPECT for automatic evaluation of %thyroid uptake based on deep-learning

PURPOSE

Quantitative thyroid single-photon emission computed tomography/computed tomography (SPECT/CT) requires computed tomography (CT)-based attenuation correction and manual thyroid segmentation on CT for %thyroid uptake measurements. Here, we aimed to develop a deep-learning-based CT-free quantitative thyroid SPECT that can generate an attenuation map (μ-map) and automatically segment the thyroid.

METHODS

Quantitative thyroid SPECT/CT data (n = 650) were retrospectively analyzed. Typical 3D U-Nets were used for the μ-map generation and automatic thyroid segmentation. Primary emission and scattering SPECTs were inputted to generate a μ-map, and the original μ-map from CT was labeled (268 and 30 for training and validation, respectively). The generated μ-map and primary emission SPECT were inputted for the automatic thyroid segmentation, and the manual thyroid segmentation was labeled (280 and 36 for training and validation, respectively). Other thyroid SPECT/CT (n = 36) and salivary SPECT/CT (n = 29) were employed for verification.

RESULTS

The synthetic μ-map demonstrated a strong correlation (R² = 0.972) and minimum error (mean square error = 0.936 × 10^-4, %normalized mean absolute error = 0.999%) of attenuation coefficients when compared to the ground truth (n = 30). Compared to manual segmentation, the automatic thyroid segmentation was excellent with a Dice similarity coefficient of 0.767, minimal thyroid volume difference of - 0.72 mL, and a short 95% Hausdorff distance of 9.416 mm (n = 36). Additionally, %thyroid uptake by synthetic μ-map and automatic thyroid segmentation (CT-free SPECT) was similar to that by the original μ-map and manual thyroid segmentation (SPECT/CT) (3.772 ± 5.735% vs. 3.682 ± 5.516%, p = 0.1090) (n = 36). Furthermore, the synthetic μ-map generation and automatic thyroid segmentation were successfully performed in the salivary SPECT/CT using the deep-learning algorithms trained by thyroid SPECT/CT (n = 29).

CONCLUSION

CT-free quantitative SPECT for automatic evaluation of %thyroid uptake can be realized by deep-learning.

Artificial Intelligence in Nuclear Cardiology

Artificial intelligence (AI) encompasses a variety of computer algorithms that have a wide range of potential clinical applications in nuclear cardiology. This article will introduce core terminology and concepts for AI including classifications of AI as well as training and testing regimens. We will then highlight the potential role for AI to improve image registration and image quality. Next, we will discuss methods for AI-driven image attenuation correction. Finally, we will review advancements in machine learning and deep-learning applications for disease diagnosis and risk stratification, including efforts to improve clinical translation of this valuable technology with explainable AI models.

Attenuation Correction for Dedicated Cardiac SPECT Imaging Without Using Transmission Data

Objectives

Attenuation correction (AC) using transmission scanning-like computed tomography (CT) is the standard method to increase the accuracy of cardiac single-photon emission computed tomography (SPECT) images. Recently developed dedicated cardiac SPECT do not support CT, and thus, scans on these systems are vulnerable to attenuation artifacts. This study presented a new method for generating an attenuation map directly from emission data by segmentation of precisely non-rigid registration extended cardiac-torso (XCAT)-digital phantom with cardiac SPECT images.

Methods

In-house developed non-rigid registration algorithm automatically aligns the XCAT- phantom with cardiac SPECT image to precisely segment the contour of organs. Pre-defined attenuation coefficients for given photon energies were assigned to generate attenuation maps. The CT-based attenuation maps were used for validation with which cardiac SPECT/CT data of 38 patients were included. Segmental myocardial counts of a 17-segment model from these databases were compared based on the basis of the paired t-test.

Results

The mean, and standard deviation of the mean square error and structural similarity index measure of the female stress phase between the proposed attenuation maps and the CT attenuation maps were 6.99±1.23% and 92±2.0%, of the male stress were 6.87±3.8% and 96±1.0%. Proposed attenuation correction and computed tomography based attenuation correction average myocardial perfusion count was significantly higher than that in non-AC in the mid-inferior, mid-lateral, basal-inferior, and lateral regions (p<0.001).

Conclusion

The proposed attenuation maps showed good agreement with the CT-based attenuation map. Therefore, it is feasible to enable AC for a dedicated cardiac SPECT or SPECT standalone scanners.

Artificial Intelligence in Nuclear Medicine: Opportunities, Challenges, and Responsibilities Toward a Trustworthy Ecosystem

Trustworthiness is a core tenet of medicine. The patient-physician relationship is evolving from a dyad to a broader ecosystem of health care. With the emergence of artificial intelligence (AI) in medicine, the elements of trust must be revisited. We envision a road map for the establishment of trustworthy AI ecosystems in nuclear medicine. In this report, AI is contextualized in the history of technologic revolutions. Opportunities for AI applications in nuclear medicine related to diagnosis, therapy, and workflow efficiency, as well as emerging challenges and critical responsibilities, are discussed. Establishing and maintaining leadership in AI require a concerted effort to promote the rational and safe deployment of this innovative technology by engaging patients, nuclear medicine physicians, scientists, technologists, and referring providers, among other stakeholders, while protecting our patients and society. This strategic plan was prepared by the AI task force of the Society of Nuclear Medicine and Molecular Imaging.

Quantification of intramyocardial blood volume using 99mTc-RBC SPECT/CT: a pilot human study

BACKGROUND

Quantification of intramyocardial blood volume (IMBV), the fraction of myocardium that is occupied by blood, is a promising Index to measure microcirculatory functions. In previous large animal SPECT/CT studies injected with 99mTc-labeled Red Blood Cell (RBC) and validated by ex vivo microCT, we have demonstrated that accurate IMBV can be measured. In this study, we report the data processing methods and results of the first-in-human pilot study.

METHODS

Data from three subjects have been included to date. Each subject underwent rest and adenosine-induced stress 99mTc-RBC SPECT/CT on a dedicated cardiac system with both non-contrast and contrast-enhanced CT acquired. Corrections of attenuation (AC) and scatter (SC), respiratory and cardiac gating, and partial volume correction (PVC) were applied. We also performed automatic segmentation and registration approach based on the blood pool topology in both SPECT and CT images.

RESULTS

The quantified IMBV across all subjects under resting conditions were 35.0% ± 3.3% for the end-diastolic phase and 24.1% ± 2.7% for the end-systolic phase. The cycle-dependent change in IMBV (ΔIMBV) between diastolic and systolic phases was 31.5% ± 3.0%. Under stress, IMBV were 40.6% ± 4.2% for the end-diastolic phase and 26.5% ± 2.8% for the end-systolic phase, and ΔIMBV was 34.7% ± 7.4%.

CONCLUSIONS

It is feasible to quantify IMBV in resting and stress conditions in human studies using SPECT/CT with 99mTc-RBC.

Deep learning-enhanced nuclear medicine SPECT imaging applied to cardiac studies

Deep learning (DL) has a growing popularity and is a well-established method of artificial intelligence for data processing, especially for images and videos. Its applications in nuclear medicine are broad and include, among others, disease classification, image reconstruction, and image de-noising. Positron emission tomography (PET) and single-photon emission computerized tomography (SPECT) are major image acquisition technologies in nuclear medicine. Though several studies have been conducted to apply DL in many nuclear medicine domains, such as cancer detection and classification, few studies have employed such methods for cardiovascular disease applications. The present paper reviews recent DL approaches focused on cardiac SPECT imaging. Extensive research identified fifty-five related studies, which are discussed. The review distinguishes between major application domains, including cardiovascular disease diagnosis, SPECT attenuation correction, image denoising, full-count image estimation, and image reconstruction. In addition, major findings and dominant techniques employed for the mentioned task are revealed. Current limitations of DL approaches and future research directions are discussed.

Deep learning-based attenuation map generation with simultaneously reconstructed PET activity and attenuation and low-dose application

Objective. In PET/CT imaging, CT is used for positron emission tomography (PET) attenuation correction (AC). CT artifacts or misalignment between PET and CT can cause AC artifacts and quantification errors in PET. Simultaneous reconstruction (MLAA) of PET activity (λ-MLAA) and attenuation (μ-MLAA) maps was proposed to solve those issues using the time-of-flight PET raw data only. However,λ-MLAA still suffers from quantification error as compared to reconstruction using the gold-standard CT-based attenuation map (μ-CT). Recently, a deep learning (DL)-based framework was proposed to improve MLAA by predictingμ-DL fromλ-MLAA andμ-MLAA using an image domain loss function (IM-loss). However, IM-loss does not directly measure the AC errors according to the PET attenuation physics. Our preliminary studies showed that an additional physics-based loss function can lead to more accurate PET AC. The main objective of this study is to optimize the attenuation map generation framework for clinical full-dose¹⁸F-FDG studies. We also investigate the effectiveness of the optimized network on predicting attenuation maps for synthetic low-dose oncological PET studies.Approach. We optimized the proposed DL framework by applying different preprocessing steps and hyperparameter optimization, including patch size, weights of the loss terms and number of angles in the projection-domain loss term. The optimization was performed based on 100 skull-to-toe¹⁸F-FDG PET/CT scans with minimal misalignment. The optimized framework was further evaluated on 85 clinical full-dose neck-to-thigh¹⁸F-FDG cancer datasets as well as synthetic low-dose studies with only 10% of the full-dose raw data.Main results. Clinical evaluation of tumor quantification as well as physics-based figure-of-merit metric evaluation validated the promising performance of our proposed method. For both full-dose and low-dose studies, the proposed framework achieved <1% error in tumor standardized uptake value measures.Significance. It is of great clinical interest to achieve CT-less PET reconstruction, especially for low-dose PET studies.

Cross-vender, cross-tracer, and cross-protocol deep transfer learning for attenuation map generation of cardiac SPECT

It has been proved feasible to generate attenuation maps (μ-maps) from cardiac SPECT using deep learning. However, this assumed that the training and testing datasets were acquired using the same scanner, tracer, and protocol. We investigated a robust generation of CT-derived μ-maps from cardiac SPECT acquired by different scanners, tracers, and protocols from the training data. We first pre-trained a network using 120 studies injected with 99mTc-tetrofosmin acquired from a GE 850 SPECT/CT with 360-degree gantry rotation, which was then fine-tuned and tested using 80 studies injected with 99mTc-sestamibi acquired from a Philips BrightView SPECT/CT with 180-degree gantry rotation. The error between ground-truth and predicted μ-maps by transfer learning was 5.13 ± 7.02%, as compared to 8.24 ± 5.01% by direct transition without fine-tuning and 6.45 ± 5.75% by limited-sample training. The error between ground-truth and reconstructed images with predicted μ-maps by transfer learning was 1.11 ± 1.57%, as compared to 1.72 ± 1.63% by direct transition and 1.68 ± 1.21% by limited-sample training. It is feasible to apply a network pre-trained by a large amount of data from one scanner to data acquired by another scanner using different tracers and protocols, with proper transfer learning.

Post-reconstruction attenuation correction for SPECT myocardium perfusion imaging facilitated by deep learning-based attenuation map generation

BACKGROUND

Attenuation correction can improve the quantitative accuracy of single-photon emission computed tomography (SPECT) images. Existing SPECT-only systems normally can only provide non-attenuation corrected (NC) images which are susceptible to attenuation artifacts. In this work, we developed a post-reconstruction attenuation correction (PRAC) approach facilitated by a deep learning-based attenuation map for myocardial perfusion SPECT imaging.

METHODS

In the PRAC method, new projection data were estimated via forwardly projecting the scanner-generated NC image. Then an attenuation map, generated from NC image using a pretrained deep learning (DL) convolutional neural network, was incorporated into an offline reconstruction algorithm to obtain the attenuation-corrected images from the forwardly projected projections. We evaluated the PRAC method using 30 subjects with a DL network trained with 40 subjects, using the vendor-generated AC images and CT-based attenuation maps as the ground truth.

RESULTS

The PRAC methods using DL-generated and CT-based attenuation maps were both highly consistent with the scanner-generated AC image. The globally normalized mean absolute errors were 1.1% ± .6% and .7% ± .4% and the localized absolute percentage errors were 8.9% ± 13.4% and 7.8% ± 11.4% in the left ventricular (LV) blood pool, respectively, and - 1.3% ± 8.0% and - 3.8% ± 4.5% in the LV myocardium for PRAC methods using DL-generated and CT-based attenuation maps, respectively. The summed stress scores after PRAC using both attenuation maps were more consistent with the ground truth than those of the NC images.

CONCLUSION

We developed a PRAC approach facilitated by deep learning-based attenuation maps for SPECT myocardial perfusion imaging. It may be feasible for this approach to provide AC images for SPECT-only scanner data.

CT-free attenuation correction for dedicated cardiac SPECT using a 3D dual squeeze-and-excitation residual dense network

BACKGROUND

Attenuation correction (AC) using CT transmission scanning enables the accurate quantitative analysis of dedicated cardiac SPECT. However, AC is challenging for SPECT-only scanners. We developed a deep learning-based approach to generate synthetic AC images from SPECT images without AC.

METHODS

CT-free AC was implemented using our customized Dual Squeeze-and-Excitation Residual Dense Network (DuRDN). 172 anonymized clinical hybrid SPECT/CT stress/rest myocardial perfusion studies were used in training, validation, and testing. Additional body mass index (BMI), gender, and scatter-window information were encoded as channel-wise input to further improve the network performance.

RESULTS

Quantitative and qualitative analysis based on image voxels and 17-segment polar map showed the potential of our approach to generate consistent SPECT AC images. Our customized DuRDN showed superior performance to conventional network design such as U-Net. The averaged voxel-wise normalized mean square error (NMSE) between the predicted AC images by DuRDN and the ground-truth AC images was 2.01 ± 1.01%, as compared to 2.23 ± 1.20% by U-Net.

CONCLUSIONS

Our customized DuRDN facilitates dedicated cardiac SPECT AC without CT scanning. DuRDN can efficiently incorporate additional patient information and may achieve better performance compared to conventional U-Net.

Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines)

An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms for nuclear medicine. To address this need, we propose a 4-class framework to evaluate AI algorithms for promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices to evaluate AI algorithms for each of these classes. Each class of evaluation yields a claim that provides a descriptive performance of the AI algorithm. Key best practices are tabulated as the RELAINCE (Recommendations for EvaLuation of AI for NuClear medicinE) guidelines. The report was prepared by the Society of Nuclear Medicine and Molecular Imaging AI Task Force Evaluation team, which consisted of nuclear-medicine physicians, physicists, computational imaging scientists, and representatives from industry and regulatory agencies.

Generation of Digital Brain Phantom for Machine Learning Application of Dopamine Transporter Radionuclide Imaging

While machine learning (ML) methods may significantly improve image quality for SPECT imaging for the diagnosis and monitoring of Parkinson’s disease (PD), they require a large amount of data for training. It is often difficult to collect a large population of patient data to support the ML research, and the ground truth of lesion is also unknown. This paper leverages a generative adversarial network (GAN) to generate digital brain phantoms for training ML-based PD SPECT algorithms. A total of 594 PET 3D brain models from 155 patients (113 male and 42 female) were reviewed and 1597 2D slices containing the full or a portion of the striatum were selected. Corresponding attenuation maps were also generated based on these images. The data were then used to develop a GAN for generating 2D brain phantoms, where each phantom consisted of a radioactivity image and the corresponding attenuation map. Statistical methods including histogram, Fréchet distance, and structural similarity were used to evaluate the generator based on 10,000 generated phantoms. When the generated phantoms and training dataset were both passed to the discriminator, similar normal distributions were obtained, which indicated the discriminator was unable to distinguish the generated phantoms from the training datasets. The generated digital phantoms can be used for 2D SPECT simulation and serve as the ground truth to develop ML-based reconstruction algorithms. The cumulated experience from this work also laid the foundation for building a 3D GAN for the same application.

Direct and indirect strategies of deep-learning-based attenuation correction for general purpose and dedicated cardiac SPECT

PURPOSE

Deep-learning-based attenuation correction (AC) for SPECT includes both indirect and direct approaches. Indirect approaches generate attenuation maps (μ-maps) from emission images, while direct approaches predict AC images directly from non-attenuation-corrected (NAC) images without μ-maps. For dedicated cardiac SPECT scanners with CZT detectors, indirect approaches are challenging due to the limited field-of-view (FOV). In this work, we aim to 1) first develop novel indirect approaches to improve the AC performance for dedicated SPECT; and 2) compare the AC performance between direct and indirect approaches for both general purpose and dedicated SPECT.

METHODS

For dedicated SPECT, we developed strategies to predict truncated μ-maps from NAC images reconstructed with a small matrix, or full μ-maps from NAC images reconstructed with a large matrix using 270 anonymized clinical studies scanned on a GE Discovery NM/CT 570c SPECT/CT. For general purpose SPECT, we implemented direct and indirect approaches using 400 anonymized clinical studies scanned on a GE NM/CT 850c SPECT/CT. NAC images in both photopeak and scatter windows were input to predict μ-maps or AC images.

RESULTS

For dedicated SPECT, the averaged normalized mean square error (NMSE) using our proposed strategies with full μ-maps was 1.20 ± 0.72% as compared to 2.21 ± 1.17% using the previous direct approaches. The polar map absolute percent error (APE) using our approaches was 3.24 ± 2.79% (R² = 0.9499) as compared to 4.77 ± 3.96% (R² = 0.9213) using direct approaches. For general purpose SPECT, the averaged NMSE of the predicted AC images using the direct approaches was 2.57 ± 1.06% as compared to 1.37 ± 1.16% using the indirect approaches.

CONCLUSIONS

We developed strategies of generating μ-maps for dedicated cardiac SPECT with small FOV. For both general purpose and dedicated SPECT, indirect approaches showed superior performance of AC than direct approaches.

"Virtual" attenuation correction: improving stress myocardial perfusion SPECT imaging using deep learning

PURPOSE

Myocardial perfusion imaging (MPI) using single-photon emission computed tomography (SPECT) is widely used for coronary artery disease (CAD) evaluation. Although attenuation correction is recommended to diminish image artifacts and improve diagnostic accuracy, approximately 3/4ths of clinical MPI worldwide remains non-attenuation-corrected (NAC). In this work, we propose a novel deep learning (DL) algorithm to provide "virtual" DL attenuation-corrected (DLAC) perfusion polar maps solely from NAC data without concurrent computed tomography (CT) imaging or additional scans.

METHODS

SPECT MPI studies (N = 11,532) with paired NAC and CTAC images were retrospectively identified. A convolutional neural network-based DL algorithm was developed and trained on half of the population to predict DLAC polar maps from NAC polar maps. Total perfusion deficit (TPD) was evaluated for all polar maps. TPDs from NAC and DLAC polar maps were compared to CTAC TPDs in linear regression analysis. Moreover, receiver-operating characteristic analysis was performed on NAC, CTAC, and DLAC TPDs to predict obstructive CAD as diagnosed from invasive coronary angiography.

RESULTS

DLAC TPDs exhibited significantly improved linear correlation (p < 0.001) with CTAC (R² = 0.85) compared to NAC vs. CTAC (R² = 0.68). The diagnostic performance of TPD was also improved with DLAC compared to NAC with an area under the curve (AUC) of 0.827 vs. 0.780 (p = 0.012) with no statistically significant difference between AUC for CTAC and DLAC. At 88% sensitivity, specificity was improved by 18.9% for DLAC and 25.6% for CTAC.

CONCLUSIONS

The proposed DL algorithm provided attenuation correction comparable to CTAC without the need for additional scans. Compared to conventional NAC perfusion imaging, DLAC significantly improved diagnostic accuracy.

Quantitative SPECT/CT for dosimetry of peptide receptor radionuclide therapy

Neuroendocrine tumors (NETs) are uncommon malignancies of increasing incidence and prevalence. As these slow growing tumors usually overexpress somatostatin receptors (SSTRs), the use of ⁶⁸Ga-DOTA-peptides (gallium-68 chelated with dodecane tetra-acetic acid to somatostatin), which bind to the SSTRs, allows for PET based imaging and selection of patients for peptide receptor radionuclide therapy (PRRT). PRRT with radiolabeled somatostatin analogues such as ¹⁷⁷Lu-DOTATATE (lutetium-177-[DOTA,Tyr³]-octreotate), is mainly used for the treatment of metastatic or inoperable NETs. However, PRRT is generally administered at a fixed injected activity in order not to exceed dose limits in critical organs, which is suboptimal given the variability in radiopharmaceutical uptake among patients. Advances in SPECT (single photon emission computed tomography) imaging enable the absolute quantitative measure of the true radiopharmaceutical distribution providing for PRRT dosimetry in each patient. Personalized PRRT based on patient-specific dosimetry could improve therapeutic efficacy by optimizing effective tumor absorbed dose while limiting treatment related radiotoxicity.

Automatic Inter-Frame Patient Motion Correction for Dynamic Cardiac PET Using Deep Learning

Patient motion during dynamic PET imaging can induce errors in myocardial blood flow (MBF) estimation. Motion correction for dynamic cardiac PET is challenging because the rapid tracer kinetics of 82Rb leads to substantial tracer distribution change across different dynamic frames over time, which can cause difficulties for image registration-based motion correction, particularly for early dynamic frames. In this paper, we developed an automatic deep learning-based motion correction (DeepMC) method for dynamic cardiac PET. In this study we focused on the detection and correction of inter-frame rigid translational motion caused by voluntary body movement and pattern change of respiratory motion. A bidirectional-3D LSTM network was developed to fully utilize both local and nonlocal temporal information in the 4D dynamic image data for motion detection. The network was trained and evaluated over motion-free patient scans with simulated motion so that the motion ground-truths are available, where one million samples based on 65 patient scans were used in training, and 600 samples based on 20 patient scans were used in evaluation. The proposed method was also evaluated using additional 10 patient datasets with real motion. We demonstrated that the proposed DeepMC obtained superior performance compared to conventional registration-based methods and other convolutional neural networks (CNN), in terms of motion estimation and MBF quantification accuracy. Once trained, DeepMC is much faster than the registration-based methods and can be easily integrated into the clinical workflow. In the future work, additional investigation is needed to evaluate this approach in a clinical context with realistic patient motion.

Artificial Intelligence and Cardiac PET/Computed Tomography Imaging

Artificial intelligence is an important technology, with rapidly expanding applications for cardiac PET. We review the common terminology, including methods for training and testing, which are fundamental to understanding artificial intelligence. Next, we highlight applications to improve image acquisition, reconstruction, and segmentation. Computed tomographic imaging is commonly acquired in conjunction with PET and various artificial intelligence methods have been applied, including methods to automatically extract anatomic information or generate synthetic attenuation images. Last, we describe methods to automate disease diagnosis or risk stratification. This summary highlights the current and future clinical applications of artificial intelligence to cardiovascular PET imaging.

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.

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.

Convolutional neural network based attenuation correction for123I-FP-CIT SPECT with focused striatum imaging

SPECT imaging with¹²³I-FP-CIT is used for diagnosis of neurodegenerative disorders like Parkinson's disease. Attenuation correction (AC) can be useful for quantitative analysis of¹²³I-FP-CIT SPECT. Ideally, AC would be performed based on attenuation maps (μ-maps) derived from perfectly registered CT scans. Suchμ-maps, however, are most times not available and possible errors in image registration can induce quantitative inaccuracies in AC corrected SPECT images. Earlier, we showed that a convolutional neural network (CNN) based approach allows to estimate SPECT-alignedμ-maps for full brain perfusion imaging using only emission data. Here we investigate the feasibility of similar CNN methods for axially focused¹²³I-FP-CIT scans. We tested our approach on a high-resolution multi-pinhole prototype clinical SPECT system in a Monte Carlo simulation study. Three CNNs that estimateμ-maps in a voxel-wise, patch-wise and image-wise manner were investigated. As the added value of AC on clinical¹²³I-FP-CIT scans is still debatable, the impact of AC was also reported to check in which cases CNN based AC could be beneficial. AC using the ground truthμ-maps (GT-AC) and CNN estimatedμ-maps (CNN-AC) were compared with the case when no AC was done (No-AC). Results show that the effect of using GT-AC versus CNN-AC or No-AC on striatal shape and symmetry is minimal. Specific binding ratios (SBRs) from localized regions show a deviation from GT-AC≤2.5% for all three CNN-ACs while No-AC systematically underestimates SBRs by 13.1%. A strong correlation (r≥0.99) was obtained between GT-AC based SBRs and SBRs from CNN-ACs and No-AC. Absolute quantification (in kBq ml^-1) shows a deviation from GT-AC within 2.2% for all three CNN-ACs and of 71.7% for No-AC. To conclude, all three CNNs show comparable performance in accurateμ-map estimation and¹²³I-FP-CIT quantification. CNN-estimatedμ-map can be a promising substitute for CT-basedμ-map.

Generation of synthetic PET images of synaptic density and amyloid from 18 F-FDG images using deep learning

PURPOSE

Positron emission tomography (PET) imaging with various tracers is increasingly used in Alzheimer's disease (AD) studies. However, access to PET scans using new or less-available tracers with sophisticated synthesis and short half-life isotopes may be very limited. Therefore, it is of great significance and interest in AD research to assess the feasibility of generating synthetic PET images of less-available tracers from the PET image of another common tracer, in particular 18 F-FDG.

METHODS

We implemented advanced deep learning methods using the U-Net model to predict 11 C-UCB-J PET images of synaptic vesicle protein 2A (SV2A), a surrogate of synaptic density, from 18 F-FDG PET data. Dynamic 18 F-FDG and 11 C-UCB-J scans were performed in 21 participants with normal cognition (CN) and 33 participants with Alzheimer's disease (AD). Cerebellum was used as the reference region for both tracers. For 11 C-UCB-J image prediction, four network models were trained and tested, which included 1) 18 F-FDG SUV ratio (SUVR) to 11 C-UCB-J SUVR, 2) 18 F-FDG Ki ratio to 11 C-UCB-J SUVR, 3) 18 F-FDG SUVR to 11 C-UCB-J distribution volume ratio (DVR), and 4) 18 F-FDG Ki ratio to 11 C-UCB-J DVR. The normalized root mean square error (NRMSE), structure similarity index (SSIM), and Pearson's correlation coefficient were calculated for evaluating the overall image prediction accuracy. Mean bias of various ROIs in the brain and correlation plots between predicted images and true images were calculated for ROI-based prediction accuracy. Following a similar training and evaluation strategy, 18 F-FDG SUVR to 11 C-PiB SUVR network was also trained and tested for 11 C-PiB static image prediction.

RESULTS

The results showed that all four network models obtained satisfactory 11 C-UCB-J static and parametric images. For 11 C-UCB-J SUVR prediction, the mean ROI bias was -0.3% ± 7.4% for the AD group and -0.5% ± 7.3% for the CN group with 18 F-FDG SUVR as the input, -0.7% ± 8.1% for the AD group, and -1.3% ± 7.0% for the CN group with 18 F-FDG Ki ratio as the input. For 11 C-UCB-J DVR prediction, the mean ROI bias was -1.3% ± 7.5% for the AD group and -2.0% ± 6.9% for the CN group with 18 F-FDG SUVR as the input, -0.7% ± 9.0% for the AD group, and -1.7% ± 7.8% for the CN group with 18 F-FDG Ki ratio as the input. For 11 C-PiB SUVR image prediction, which appears to be a more challenging task, the incorporation of additional diagnostic information into the network is needed to control the bias below 5% for most ROIs.

CONCLUSIONS

It is feasible to use 3D U-Net-based methods to generate synthetic 11 C-UCB-J PET images from 18 F-FDG images with reasonable prediction accuracy. It is also possible to predict 11 C-PiB SUVR images from 18 F-FDG images, though the incorporation of additional non-imaging information is needed.

Limited View Tomographic Reconstruction Using a Cascaded Residual Dense Spatial-Channel Attention Network With Projection Data Fidelity Layer

Limited view tomographic reconstruction aims to reconstruct a tomographic image from a limited number of projection views arising from sparse view or limited angle acquisitions that reduce radiation dose or shorten scanning time. However, such a reconstruction suffers from severe artifacts due to the incompleteness of sinogram. To derive quality reconstruction, previous methods use UNet-like neural architectures to directly predict the full view reconstruction from limited view data; but these methods leave the deep network architecture issue largely intact and cannot guarantee the consistency between the sinogram of the reconstructed image and the acquired sinogram, leading to a non-ideal reconstruction. In this work, we propose a cascaded residual dense spatial-channel attention network consisting of residual dense spatial-channel attention networks and projection data fidelity layers. We evaluate our methods on two datasets. Our experimental results on AAPM Low Dose CT Grand Challenge datasets demonstrate that our algorithm achieves a consistent and substantial improvement over the existing neural network methods on both limited angle reconstruction and sparse view reconstruction. In addition, our experimental results on Deep Lesion datasets demonstrate that our method is able to generate high-quality reconstruction for 8 major lesion types.

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.

Attenuation correction using deep learning for brain perfusion SPECT images

OBJECTIVE

Non-uniform attenuation correction using computed tomography (CT) improves the image quality and quantification of single-photon emission computed tomography (SPECT). However, it is not widely used because it requires a SPECT/CT scanner. This study constructs a convolutional neural network (CNN) to generate attenuation-corrected SPECT images directly from non-attenuation-corrected SPECT images.

METHODS

We constructed an auto-encoder (AE) using a CNN to correct the attenuation in brain perfusion SPECT images. SPECT image datasets of 270 (44,528 slices including augmentation), 60 (5002 slices), and 30 (2558 slices) cases were used for training, validation, and testing, respectively. The acquired projection data were reconstructed in three patterns: uniform attenuation correction using Chang's method (Chang-AC), non-uniform attenuation correction using CT (CT-AC), and no attenuation correction (No-AC). The AE learned an end-to-end mapping between the No-AC and CT-AC images. The No-AC images in the test dataset were loaded into the trained AE, which generated images simulating the CT-AC images as output. The generated SPECT images were employed as attenuation-corrected images using the AE (AE-AC). The accuracy of the AE-AC images was evaluated in terms of the peak signal-to-noise ratio (PSNR) and the structural similarity metric (SSIM). The intensities of the AE-AC and CT-AC images were compared by voxel-by-voxel and region-by-region analysis.

RESULTS

The PSNRs of the AE-AC and Chang-AC images, compared using CT-AC images, were 62.2, and 57.9, and their SSIM values were 0.9995 and 0.9985, respectively. The AE-AC and CT-AC images were visually and statistically in good agreement.

CONCLUSIONS

The proposed AE-AC method yields highly accurate attenuation-corrected brain perfusion SPECT images.