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.

SPECT and CT misregistration reduction in [99mTc]Tc-MAA SPECT/CT for precision liver radioembolization treatment planning.

PURPOSE: Respiration and body movement induce misregistration between static [99mTc]Tc-MAA SPECT and CT, causing lung shunting fraction (LSF) and tumor-to-normal liver ratio (TNR) errors for 90Y radioembolization planning. We aim to alleviate the misregistration between [99mTc]Tc-MAA SPECT and CT using two registration schemes on simulation and clinical data. METHODS: In the simulation study, 70 XCAT phantoms were modeled. The SIMIND Monte Carlo program and OS-EM algorithm were used for projection generation and reconstruction, respectively. Low-dose CT (LDCT) at end-inspiration was simulated for attenuation correction (AC), lungs and liver segmentation, while contrast-enhanced CT (CECT) was simulated for tumor and perfused liver segmentation. In the clinical study, 16 patient data including [99mTc]Tc-MAA SPECT/LDCT and CECT with observed SPECT and CT mismatch were analyzed. Two liver-based registration schemes were studied: SPECT registered to LDCT/CECT and vice versa. Mean count density (MCD) of different volumes-of-interest (VOIs), normalized mutual information (NMI), LSF, TNR, and maximum injected activity (MIA) based on the partition model before and after registration were compared. Wilcoxon signed-rank test was performed. RESULTS: In the simulation study, compared to before registration, registrations significantly reduced estimation errors of MCD of all VOIs, LSF (Scheme 1: - 100.28%, Scheme 2: - 101.59%), and TNR (Scheme 1: - 7.00%, Scheme 2: - 5.67%), as well as MIA (Scheme 1: - 3.22%, Scheme 2: - 2.40%). In the clinical study, Scheme 1 reduced 33.68% LSF and increased 14.75% TNR, while Scheme 2 reduced 38.88% LSF and increased 6.28% TNR compared to before registration. One patient may change from 90Y radioembolization untreatable to treatable and other patients may change the MIA up to 25% after registration. NMI between SPECT and CT was significantly increased after registrations in both studies. CONCLUSION: Registration between static [99mTc]Tc-MAA SPECT and corresponding CTs is feasible to reduce their spatial mismatch and improve dosimetric estimation. The improvement of LSF is larger than TNR. Our method can potentially improve patient selection and personalized treatment planning for liver radioembolization.

Fast myocardial perfusion SPECT denoising using an attention-guided generative adversarial network.

Purpose: Deep learning-based denoising is promising for myocardial perfusion (MP) SPECT. However, conventional convolutional neural network (CNN)-based methods use fixed-sized convolutional kernels to convolute one region within the receptive field at a time, which would be ineffective for learning the feature dependencies across large regions. The attention mechanism (Att) is able to learn the relationships between the local receptive field and other voxels in the image. In this study, we propose a 3D attention-guided generative adversarial network (AttGAN) for denoising fast MP-SPECT images. Methods: Fifty patients who underwent 1184 MBq 99mTc-sestamibi stress SPECT/CT scan were retrospectively recruited. Sixty projections were acquired over 180° and the acquisition time was 10 s/view for the full time (FT) mode. Fast MP-SPECT projection images (1 s to 7 s) were generated from the FT list mode data. We further incorporated binary patient defect information (0 = without defect, 1 = with defect) into AttGAN (AttGAN-def). AttGAN, AttGAN-def, cGAN, and Unet were implemented using Tensorflow with the Adam optimizer running up to 400 epochs. FT and fast MP-SPECT projection pairs of 35 patients were used for training the networks for each acquisition time, while 5 and 10 patients were applied for validation and testing. Five-fold cross-validation was performed and data for all 50 patients were tested. Voxel-based error indices, joint histogram, linear regression, and perfusion defect size (PDS) were analyzed. Results: All quantitative indices of AttGAN-based networks are superior to cGAN and Unet on all acquisition time images. AttGAN-def further improves AttGAN performance. The mean absolute error of PDS by AttcGAN-def was 1.60 on acquisition time of 1 s/prj, as compared to 2.36, 2.76, and 3.02 by AttGAN, cGAN, and Unet. Conclusion: Denoising based on AttGAN is superior to conventional CNN-based networks for MP-SPECT.

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.

Deep learning-based denoising in projection-domain and reconstruction-domain for low-dose myocardial perfusion SPECT.

BACKGROUND: Low-dose (LD) myocardial perfusion (MP) SPECT suffers from high noise level, leading to compromised diagnostic accuracy. Here we investigated the denoising performance for MP-SPECT using a conditional generative adversarial network (cGAN) in projection-domain (cGAN-prj) and reconstruction-domain (cGAN-recon). METHODS: Sixty-four noisy SPECT projections were simulated for a population of 100 XCAT phantoms with different anatomical variations and 99mTc-sestamibi distributions. Series of LD projections were obtained by scaling the full dose (FD) count rate to be 1/20 to 1/2 of the original. Twenty patients with 99mTc-sestamibi stress SPECT/CT scans were retrospectively analyzed. For each patient, LD SPECT images (7/10 to 1/10 of FD) were generated from the FD list mode data. All projections were reconstructed by the quantitative OS-EM method. A 3D cGAN was implemented to predict FD images from their corresponding LD images in the projection- and reconstruction-domain. The denoised projections were reconstructed for analysis in various quantitative indices along with cGAN-recon, Gaussian, and Butterworth-filtered images. RESULTS: cGAN denoising improves image quality as compared to LD and conventional post-reconstruction filtering. cGAN-prj can further reduce the dose level as compared to cGAN-recon without compromising the image quality. CONCLUSIONS: Denoising based on cGAN-prj is superior to cGAN-recon for MP-SPECT.

Pix2Pix generative adversarial network for low dose myocardial perfusion SPECT denoising.

Background: Myocardial perfusion (MP) SPECT is a well-established method for diagnosing cardiac disease, yet its radiation risk poses safety concern. This study aims to apply and evaluate the use of Pix2Pix generative adversarial network (Pix2Pix GAN) in denoising low dose MP SPECT images. Methods: One hundred male and female patients with different 99mTc-sestamibi activity distributions, organ and body sizes were simulated by a population of digital 4D Extended Cardiac Torso (XCAT) phantoms. Realistic noisy SPECT projections of full dose of 987 MBq injection and 16 min acquisition, and low dose ranged from 1/20 to 1/2 of the full dose, were generated by an analytical projector from the right anterior oblique (RAO) to the left posterior oblique (LPO) positions. Additionally, twenty patients underwent ~1,184 MBq 99mTc-sestamibi stress SPECT/CT scan were also retrospectively recruited for the study. For each patient, low dose SPECT images (7/10 to 1/10 of full dose) were generated from the full dose list mode data. Our Pix2Pix GAN model was trained with full dose and low dose reconstructed SPECT image pairs. Normalized mean square error (NMSE), structural similarity index (SSIM), coefficient of variation (CV), full-width-at-half-maximum (FWHM) and relative defect size differences (RSD) of Pix2Pix GAN processed images were evaluated along with a reference convolutional auto encoder (CAE) network and post-reconstruction filters. Results: NMSE values of 0.0233±0.004 vs. 0.0249±0.004 and 0.0313±0.007 vs. 0.0579±0.016 were obtained on 1/2 and 1/20 dose level for Pix2Pix GAN and CAE in the simulation study, while they were 0.0376±0.010 vs. 0.0433±0.010 and 0.0907±0.020 vs. 0.1186±0.025 on 7/10 and 1/10 dose level in the clinical study. Similar results were also obtained from the SSIM, CV, FWHM and RSD values. Overall, the use of Pix2Pix GAN was superior to other denoising methods in all physical indices, particular in the lower dose levels in the simulation and clinical study. Conclusions: The Pix2Pix GAN method is effective to reduce the noise level of low dose MP SPECT. Further studies on clinical performance are warranted to demonstrate its full clinical effectiveness.

A Novel Prognostic Signature Based on Glioma Essential Ferroptosis-Related Genes Predicts Clinical Outcomes and Indicates Treatment in Glioma.

Background: Ferroptosis is a form of programmed cell death (PCD) that has been implicated in cancer progression, although the specific mechanism is not known. Here, we used the latest DepMap release CRISPR data to identify the essential ferroptosis-related genes (FRGs) in glioma and their role in patient outcomes. Methods: RNA-seq and clinical information on glioma cases were obtained from the Chinese Glioma Genome Atlas (CGGA) and The Cancer Genome Atlas (TCGA). FRGs were obtained from the FerrDb database. CRISPR-screened essential genes (CSEGs) in glioma cell lines were downloaded from the DepMap portal. A series of bioinformatic and machine learning approaches were combined to establish FRG signatures to predict overall survival (OS) in glioma patients. In addition, pathways analysis was used to identify the functional roles of FRGs. Somatic mutation, immune cell infiltration, and immune checkpoint gene expression were analyzed within the risk subgroups. Finally, compounds for reversing high-risk gene signatures were predicted using the GDSC and L1000 datasets. Results: Seven FRGs (ISCU, NFS1, MTOR, EIF2S1, HSPA5, AURKA, RPL8) were included in the model and the model was found to have good prognostic value (p < 0.001) in both training and validation groups. The risk score was found to be an independent prognostic factor and the model had good efficacy. Subgroup analysis using clinical parameters demonstrated the general applicability of the model. The nomogram indicated that the model could effectively predict 12-, 36-, and 60-months OS and progression-free interval (PFI). The results showed the presence of more aggressive phenotypes (lower numbers of IDH mutations, higher numbers of EGFR and PTEN mutations, greater infiltration of immune suppressive cells, and higher expression of immune checkpoint inhibitors) in the high-risk group. The signaling pathways enriched closely related to the cell cycle and DNA damage repair. Drug predictions showed that patients with higher risk scores may benefit from treatment with RTK pathway inhibitors, including compounds that inhibit RTKs directly or indirectly by targeting downstream PI3K or MAPK pathways. Conclusion: In summary, the proposed cancer essential FRG signature predicts survival and treatment response in glioma.

Dual gating myocardial perfusion SPECT denoising using a conditional generative adversarial network.

PURPOSE: Dual respiratory-cardiac gating reduces respiratory and cardiac motion blur in myocardial perfusion single-photon emission computed tomography (MP-SPECT). However, image noise is increased as detected counts are reduced in each dual gate (DG). We aim to develop a denoising method for dual gating MP-SPECT images using a 3D conditional generative adversarial network (cGAN). METHODS: Twenty extended cardiac-torso phantoms with various 99m Tc-sestamibi distributions, defect characteristics, and body and organ sizes were used in the simulation, modeling six respiratory and eight cardiac gates (CGs), that is, 48 DGs for ordered subset expectation maximization reconstruction. Twenty clinical 99m Tc-sestamibi SPECT/CT datasets were re-binned into 7 respiratory gates and 8 CGs, that is, 56 DGs for maximum likelihood expectation maximization reconstruction. We evaluated the use of (i) phantoms' own datasets (patient-specific denoising [PD]) or other phantoms' datasets (cross-patient denoising) for training; (ii) the CG or the static (non-gated [NG]) data as the training references for cGAN; and (iii) cGAN as compared to conventional 3D post-reconstruction filtering, cardiac gating methods, and convolutional neural network. Normalized mean squared error, noise as assessed by normalized standard deviation, spatial blurring measured as the full-width-at-half-maximum of left ventricular wall, ejection fraction, joint correlation histogram, and defect size were analyzed as metrics of image quality. RESULTS: Training using patients' own dataset is superior to conventional training based on other patients' data. Using CG image as training reference provides a better trade-off in terms of noise and image blur as compared to the use of NG. cGAN-CG-PD provides superior performance as compared to other denoising methods for all physical and diagnostic indices evaluated in both simulation and clinical studies. CONCLUSIONS: cGAN denoising is promising for dual gating MP-SPECT based on the metrics mentioned earlier.

Clinical evaluation of three respiratory gating schemes for different respiratory patterns on cardiac SPECT.

PURPOSE: Respiratory gating reduces respiratory blur in cardiac single photon emission computed tomography (SPECT). It can be implemented as three gating schemes: (a) equal amplitude-based gating (AG); (b) phase or time-based gating (TG); or (c) equal count-based gating (CG), that is, a variant of amplitude-based method. The goal of this study is to evaluate the effectiveness of these respiratory gating methods for patients with different respiratory patterns in myocardial perfusion SPECT. METHODS: We reviewed 1274 anonymized patient respiratory traces obtained via the Vicon motion-tracking system during their 99m Tc-sestamibi SPECT scans and grouped them into four breathing categories: (a) regular respiration (RR); (b) periodic respiration (PR); (c) respiration with apnea (AR); and (d) unclassified respiration (UR). For each respiratory pattern, 15 patients were randomly selected and their list-mode data were rebinned using the three gating schemes. A preliminary reconstruction was performed for each gate with the heart region segmented and registered to a reference gate to estimate the respiratory motion. A final reconstruction incorporating respiratory motion correction was done to get a final image set. The estimated respiratory motion, the full-width-at-half-maxima (FWHM) measured across the image intensity profile of the left ventricle wall, as well as the normalized standard deviation measured in a uniform cuboid region of the thorax were analyzed. RESULTS: There are 47.1%, 24.3%, 13.5%, and 15.1% RR, PR, AR, and UR patients, respectively, among the 1274 patients in this study. The differences among the three gating schemes in RR were smaller than other respiratory patterns. The AG and CG methods showed statistically larger motion estimation than TG particularly in the AR and PR patterns. Noise of AG varied more in different gates, especially for AR and UR patterns. CONCLUSION: More than half of the patients reviewed exhibited nonregular breathing patterns. Amplitude-based gating, that is, AG and CG, is a preferred gating method for such patterns and is a robust respiratory gating implementation method given the respiratory pattern of the patients is unknown before data acquisition. Phase gating is also a feasible option for regular respiratory pattern.