SPICER: Self-supervised learning for MRI with automatic coil sensitivity estimation and reconstruction.

PURPOSE: To introduce a novel deep model-based architecture (DMBA), SPICER, that uses pairs of noisy and undersampled k-space measurements of the same object to jointly train a model for MRI reconstruction and automatic coil sensitivity estimation. METHODS: SPICER consists of two modules to simultaneously reconstructs accurate MR images and estimates high-quality coil sensitivity maps (CSMs). The first module, CSM estimation module, uses a convolutional neural network (CNN) to estimate CSMs from the raw measurements. The second module, DMBA-based MRI reconstruction module, forms reconstructed images from the input measurements and the estimated CSMs using both the physical measurement model and learned CNN prior. With the benefit of our self-supervised learning strategy, SPICER can be efficiently trained without any fully sampled reference data. RESULTS: We validate SPICER on both open-access datasets and experimentally collected data, showing that it can achieve state-of-the-art performance in highly accelerated data acquisition settings (up to 10 × $$ 10\times $$ ). Our results also highlight the importance of different modules of SPICER-including the DMBA, the CSM estimation, and the SPICER training loss-on the final performance of the method. Moreover, SPICER can estimate better CSMs than pre-estimation methods especially when the ACS data is limited. CONCLUSION: Despite being trained on noisy undersampled data, SPICER can reconstruct high-quality images and CSMs in highly undersampled settings, which outperforms other self-supervised learning methods and matches the performance of the well-known E2E-VarNet trained on fully sampled ground-truth data.

Respiratory motion management using a single rapid MRI scan for a 0.35 T MRI-Linac system.

BACKGROUND: MRI has a rapidly growing role in radiation therapy (RT) for treatment planning, real-time image guidance, and beam gating (e.g., MRI-Linac). Free-breathing 4D-MRI is desirable in respiratory motion management for therapy. Moreover, high-quality 3D-MRIs without motion artifacts are needed to delineate lesions. Existing MRI methods require multiple scans with lengthy acquisition times or are limited by low spatial resolution, contrast, and signal-to-noise ratio. PURPOSE: We developed a novel method to obtain motion-resolved 4D-MRIs and motion-integrated 3D-MRI reconstruction using a single rapid (35-45 s scan on a 0.35 T MRI-Linac. METHODS: Golden-angle radial stack-of-stars MRI scans were acquired from a respiratory motion phantom and 12 healthy volunteers (n = 12) on a 0.35 T MRI-Linac. A self-navigated method was employed to detect respiratory motion using 2000 (acquisition time = 5-7 min) and the first 200 spokes (acquisition time = 35-45 s). Multi-coil non-uniform fast Fourier transform (MCNUFFT), compressed sensing (CS), and deep-learning Phase2Phase (P2P) methods were employed to reconstruct motion-resolved 4D-MRI using 2000 spokes (MCNUFFT2000) and 200 spokes (CS200 and P2P200). Deformable motion vector fields (MVFs) were computed from the 4D-MRIs and used to reconstruct motion-corrected 3D-MRIs with the MOtion Transformation Integrated forward-Fourier (MOTIF) method. Image quality was evaluated quantitatively using the structural similarity index measure (SSIM) and the root mean square error (RMSE), and qualitatively in a blinded radiological review. RESULTS: Evaluation using the respiratory motion phantom experiment showed that the proposed method reversed the effects of motion blurring and restored edge sharpness. In the human study, P2P200 had smaller inaccuracy in MVFs estimation than CS200. P2P200 had significantly greater SSIMs (p < 0.0001) and smaller RMSEs (p < 0.001) than CS200 in motion-resolved 4D-MRI and motion-corrected 3D-MRI. The radiological review found that MOTIF 3D-MRIs using MCNUFFT2000 exhibited the highest image quality (scoring > 8 out of 10), followed by P2P200 (scoring > 5 out of 10), and then motion-uncorrected (scoring < 3 out of 10) in sharpness, contrast, and artifact-freeness. CONCLUSIONS: We have successfully demonstrated a method for respiratory motion management for MRI-guided RT. The method integrated self-navigated respiratory motion detection, deep-learning P2P 4D-MRI reconstruction, and a motion integrated reconstruction (MOTIF) for 3D-MRI using a single rapid MRI scan (35-45 s) on a 0.35 T MRI-Linac system.

Deformation-Compensated Learning for Image Reconstruction Without Ground Truth.

Deep neural networks for medical image reconstruction are traditionally trained using high-quality ground-truth images as training targets. Recent work on Noise2Noise (N2N) has shown the potential of using multiple noisy measurements of the same object as an alternative to having a ground-truth. However, existing N2N-based methods are not suitable for learning from the measurements of an object undergoing nonrigid deformation. This paper addresses this issue by proposing the deformation-compensated learning (DeCoLearn) method for training deep reconstruction networks by compensating for object deformations. A key component of DeCoLearn is a deep registration module, which is jointly trained with the deep reconstruction network without any ground-truth supervision. We validate DeCoLearn on both simulated and experimentally collected magnetic resonance imaging (MRI) data and show that it significantly improves imaging quality.

MR-assisted PET respiratory motion correction using deep-learning based short-scan motion fields.

PURPOSE: We evaluated the impact of PET respiratory motion correction (MoCo) in a phantom and patients. Moreover, we proposed and examined a PET MoCo approach using motion vector fields (MVFs) from a deep-learning reconstructed short MRI scan. METHODS: The evaluation of PET MoCo was performed in a respiratory motion phantom study with varying lesion sizes and tumor to background ratios (TBRs) using a static scan as the ground truth. MRI-based MVFs were derived from either 2000 spokes (MoCo2000 , 5-6 min acquisition time) using a Fourier transform reconstruction or 200 spokes (MoCoP2P200 , 30-40 s acquisition time) using a deep-learning Phase2Phase (P2P) reconstruction and then incorporated into PET MoCo reconstruction. For six patients with hepatic lesions, the performance of PET MoCo was evaluated using quantitative metrics (SUVmax , SUVpeak , SUVmean , lesion volume) and a blinded radiological review on lesion conspicuity. RESULTS: MRI-assisted PET MoCo methods provided similar results to static scans across most lesions with varying TBRs in the phantom. Both MoCo2000 and MoCoP2P200 PET images had significantly higher SUVmax , SUVpeak , SUVmean and significantly lower lesion volume than non-motion-corrected (non-MoCo) PET images. There was no statistical difference between MoCo2000 and MoCoP2P200 PET images for SUVmax , SUVpeak , SUVmean or lesion volume. Both radiological reviewers found that MoCo2000 and MoCoP2P200 PET significantly improved lesion conspicuity. CONCLUSION: An MRI-assisted PET MoCo method was evaluated using the ground truth in a phantom study. In patients with hepatic lesions, PET MoCo images improved quantitative and qualitative metrics based on only 30-40 s of MRI motion modeling data.

Learning-based motion artifact removal networks for quantitative R2∗ mapping.

PURPOSE: To introduce two novel learning-based motion artifact removal networks (LEARN) for the estimation of quantitative motion- and B0 -inhomogeneity-corrected R2∗ maps from motion-corrupted multi-Gradient-Recalled Echo (mGRE) MRI data. METHODS: We train two convolutional neural networks (CNNs) to correct motion artifacts for high-quality estimation of quantitative B0 -inhomogeneity-corrected R2∗ maps from mGRE sequences. The first CNN, LEARN-IMG, performs motion correction on complex mGRE images, to enable the subsequent computation of high-quality motion-free quantitative R2∗ (and any other mGRE-enabled) maps using the standard voxel-wise analysis or machine learning-based analysis. The second CNN, LEARN-BIO, is trained to directly generate motion- and B0 -inhomogeneity-corrected quantitative R2∗ maps from motion-corrupted magnitude-only mGRE images by taking advantage of the biophysical model describing the mGRE signal decay. RESULTS: We show that both CNNs trained on synthetic MR images are capable of suppressing motion artifacts while preserving details in the predicted quantitative R2∗ maps. Significant reduction of motion artifacts on experimental in vivo motion-corrupted data has also been achieved by using our trained models. CONCLUSION: Both LEARN-IMG and LEARN-BIO can enable the computation of high-quality motion- and B0 -inhomogeneity-corrected R2∗ maps. LEARN-IMG performs motion correction on mGRE images and relies on the subsequent analysis for the estimation of R2∗ maps, while LEARN-BIO directly performs motion- and B0 -inhomogeneity-corrected R2∗ estimation. Both LEARN-IMG and LEARN-BIO jointly process all the available gradient echoes, which enables them to exploit spatial patterns available in the data. The high computational speed of LEARN-BIO is an advantage that can lead to a broader clinical application.

Phase2Phase: Respiratory Motion-Resolved Reconstruction of Free-Breathing Magnetic Resonance Imaging Using Deep Learning Without a Ground Truth for Improved Liver Imaging.

OBJECTIVES: Respiratory binning of free-breathing magnetic resonance imaging data reduces motion blurring; however, it exacerbates noise and introduces severe artifacts due to undersampling. Deep neural networks can remove artifacts and noise but usually require high-quality ground truth images for training. This study aimed to develop a network that can be trained without this requirement. MATERIALS AND METHODS: This retrospective study was conducted on 33 participants enrolled between November 2016 and June 2019. Free-breathing magnetic resonance imaging was performed using a radial acquisition. Self-navigation was used to bin the k-space data into 10 respiratory phases. To simulate short acquisitions, subsets of radial spokes were used in reconstructing images with multicoil nonuniform fast Fourier transform (MCNUFFT), compressed sensing (CS), and 2 deep learning methods: UNet3DPhase and Phase2Phase (P2P). UNet3DPhase was trained using a high-quality ground truth, whereas P2P was trained using noisy images with streaking artifacts. Two radiologists blinded to the reconstruction methods independently reviewed the sharpness, contrast, and artifact-freeness of the end-expiration images reconstructed from data collected at 16% of the Nyquist sampling rate. The generalized estimating equation method was used for statistical comparison. Motion vector fields were derived to examine the respiratory motion range of 4-dimensional images reconstructed using different methods. RESULTS: A total of 15 healthy participants and 18 patients with hepatic malignancy (50 ± 15 years, 6 women) were enrolled. Both reviewers found that the UNet3DPhase and P2P images had higher contrast (P < 0.01) and fewer artifacts (P < 0.01) than the CS images. The UNet3DPhase and P2P images were reported to be sharper than the CS images by 1 reviewer (P < 0.01) but not by the other reviewer (P = 0.22, P = 0.18). UNet3DPhase and P2P were similar in sharpness and contrast, whereas UNet3DPhase had fewer artifacts (P < 0.01). The motion vector lengths for the MCNUFFT800 and P2P800 images were comparable (10.5 ± 4.2 mm and 9.9 ± 4.0 mm, respectively), whereas both were significantly larger than CS2000 (7.0 ± 3.9 mm; P < 0.0001) and UNnet3DPhase800 (6.9 ± 3.2; P < 0.0001) images. CONCLUSIONS: Without a ground truth, P2P can reconstruct sharp, artifact-free, and high-contrast respiratory motion-resolved images from highly undersampled data. Unlike the CS and UNet3DPhase methods, P2P did not artificially reduce the respiratory motion range.

Protective effect of quercetin on the homocysteine-injured human umbilical vein vascular endothelial cell line (ECV304).

Homocysteine is responsible for the occurrence of many cardiovascular diseases for instance by injuring the vascular endothelial cells. Quercetin has many beneficial effects on the cardiovascular system, but it is unknown whether it provides protection against homocysteine-injured vascular endothelial cells. The aim of the present study was to investigate the protective effect and mechanism of quercetin on the homocysteine-injured human umbilical vein vascular endothelial cell line (ECV304) (i.e. morphology, viability and nuclear factor kappa B (NF-kappaB) expression of ECV304 injured with 1.0 mM homocysteine) by determination of lipid peroxidant and endothelium-derived factors in the cultural medium of homocysteine-injured ECV304. Quercetin at 6.25, 12.5, 25, 50 and 100 microM attenuated the morphological changes and increased viability of homocysteine-injured ECV304 in a dose-dependent manner (P < 0.05 or P < 0.01 versus the homocysteine-injured group). At the same time, quercetin at 12.5, 25 and 50 microM decreased malondialdehyde level, endothelin release and NF-kappaB expression, and increased superoxide dismutase activity, nitric oxide and 6-keto-prostaglandin F1alpha releases in homocysteine-injured ECV304 (P < 0.05 or P < 0.01 versus the homocysteine-injured group). These results suggest that quercetin has a protective effect on homocysteine-injured vascular endothelial cells by antioxidant and anti-inflammatory mechanisms.

Fenofibrate inhibits tumor necrosis factor-alpha-induced expression of CD40 and matrix metalloproteinase in human vascular endothelial cells.

OBJECTIVE: To investigate the regulatory effects of fenofibrate on TNF-alpha-induced CD40 expression and matrix metalloproteinase (MMP) activity in human vascular endothelial cells (HUVECs). METHODS: Quantitative RT-PCR and flow cytometry were employed to evaluate the effect of fenofibrate on TNF-alpha-induced CD40 mRNA and cell surface CD40 expression in HUVECs, and gelatin zymography was used to determine the effect of fenofibrate on the gelatinolytic activities of MMP-2 and MMP-9 in TNF-alpha-stimulated HUVECs. RESULTS: Fenofibrate at the concentrations of 5x10(-5), 1x10(-4) and 2x10(-4) mol/L significantly reduced TNF-alpha-induced increment of CD40 mRNA and cell surface CD40 expressions (P<0.01), with the maximal inhibition achieved at the concentration of 1x10(-4) mol/L. Fenofibrate at 2x10(-4) mol/L did not further decrease CD40 expression induced by TNF-alpha. Fenofibrate significantly inhibited the stimulatory effect of TNF-alpha on MMP-2 and MMP-9 activities in HUVECs. CONCLUSION: Fenofibrate reduces TNF-alpha-induced increment of CD40 expression and MMP-2 and MMP-9 activities in HUVECs.

Lovastatin reduces apoptosis and downregulates the CD40 expression induced by TNF-alpha in cerebral vascular endothelial cells.

Inflammation may be one of the independent risk factors contributing to many neurological diseases. Moreover, there is an emerging body of data indicating that statins may have neuroprotective action. Recent studies suggest that CD40-CD40 ligand (CD40L) system is proven to be an important mediator of several auto-immune and chronic inflammation diseases. To address whether lovastatin produces neuroprotection as a potential novel anti-inflammatory pathway through the inhibition of CD40 expression, we examined the possible effects of lovastatin on expression of CD40, apoptosis, level of nitric oxide (NO) and nitric oxide synthase (NOS) activity induced by tumor necrosis factor alpha (TNF-alpha) in the cerebral vascular endothelial cells (CVECs) involved in cerebrovascular diseases. Preincubation with lovastatin (10(-7), 10(-6) and 10(-5) mol/l) for 24 hours (h) protected CVECs from TNF-alpha-induced decrease of cellular viability. Further, lovastatin inhibited the TNF-alpha-induced increases of NO level, NOS activity, apoptotic cells and CD40 expression in a dose-dependent manner, and anti-CD40 antibody also inhibited the cellular apoptosis induced by TNF-alpha. In conclusion, our data provide evidence to support a direct pro-inflammatory effect of CD40-CD40L signaling pathway in CVECs, and lovastatin possesses an anti-inflammatory effect independent of its lipid-lowering action involved in the cerebrovascular diseases.

[Liposome-mediated human CD40 gene transfection and human umbilical vein endothelial ECV-304 cells].

OBJECTIVE: To construct an eukaryotic expression vector containing human CD40 gene for its efficient, continuous and stable expression in human umbilical vein endothelial ECV-304 cells. METHODS: The recombinant plasmid pUCD40 was digested with endonucleases to obtain human CD40 gene fragment, which was cloned into pCDNA3.1 vector to construct recombinant eukaryotic expression vector pCDNA3.1(+)/CD40. The recombinant vector was identified by enzyme digestion before introduced into ECV-304 cells via liposome, with the positive cell clones selected with G418. The stable transfection and expression of CD40 in ECV-304 cells were identified by reverse transcription (RT)-PCR, Western blotting and flow cytometry, respectively. RESULTS: Enzyme digestion analysis showed that target gene had been cloned into the recombinant vector. The transfected ECV-304 cells successfully expressed human CD40 as determined by RT-PCR and Western-blotting, and 95% of the cells were CD40-positive as shown by flow cytometry. CONCLUSION: The recombinant eukaryotic expression vector pCDNA3.1(+)/CD40 has been successfully constructed, which is capable of stable transfection and expression of CD40 in ECV-304 cells to facilitate further investigation of the roles of CD40 molecule in antiatherosclerotic drug development.

Lovastatin reduces nuclear factor kappaB activation induced by C-reactive protein in human vascular endothelial cells.

The role of C-reactive protein (CRP) in atherogenesis has been supported by more recent data. Some studies have demonstrated marked up-regulation inflammatory responses in endothelial cells subjected to CRP. The nuclear factor-kappaB (NF-kappaB) signal transduction is known to play a key role in the expression of these proatherogenic entities. Statins have anti-inflammatory properties independent of their cholesterol-lowering effects. Therefore, we studied the effects of CRP and lovastatin on NF-kappaB activation in human umbilical vein endothelial cells (HUVECs). By using an electrophoretic mobility shift assays (EMSA), we found that CRP (50 microg/ml) increased activation of NF-kappaB and degradation of inhibitory kappa B (IkappaB) in HUVECs, reaching a maximal effect after the incubation with CRP for 1 h. Lovastatin (10(-5) mol/l) diminished NF-kappaB activation induced by CRP. Furthermore, lovastatin may block NF-kappaB activation by causing a stabilization of the IkappaB-alpha in cellular cytoplasm with western blotting analysis. Preincubation of HUVECs with pyrrolidinethiocarbamate (PDTC, NF-kappaB inhibitor) diminished CD40 expression induced by CRP with flow cytometry. Our results suggest that CRP increases activation of NF-kappaB and induces CD40 expression in HUVECs partly via activation of NF-kappaB. Lovastatin, through the inhibition of NF-kappaB activation, reduces the inflammation involved in the pathogenesis of atherosclerosis.

C-reactive protein-induced expression of CD40-CD40L and the effect of lovastatin and fenofibrate on it in human vascular endothelial cells.

Inflammation plays a pivotal role in the formation of atherosclerosis. In addition to being a risk marker for cardiovascular diseases, the role of C-reactive protein (CRP) in atherogenesis has been supported by more recent data. CD40-CD40L system is proven to be an important mediator of several auto-immune and chronic inflammation diseases. Interruption of CD40-CD40L signaling pathway not only reduces the initiation and progression of atherosclerotic lesions, but also modulates plaque architecture. By using a flow cytometry and western blotting, we found that incubation of human umbilical vein endothelial cells (HUVECs) with CRP resulted in a time- and dose-dependent increase in the cell-surface expression of CD40 and CD40L. In addition, CRP (25 microg/ml) increased gelatinolytic activities of MMP-2 and MMP-9. Anti-CD40 antibody significantly reversed the upregulated activities of MMP-2 and MMP-9 induced by CRP with gelatin zymography. Furthermore, lovastatin (10(-7), 10(-6), 10(-5) mol/l) and fenofibrate (5 x 10(-5), 10(-4), 2 x 10(-4) mol/l) significantly diminished the expression of CD40, CD40L and gelatinase activities (MMP-2, MMP-9) induced by CRP in HUVECs. In conclusion, our data provide evidence to support the direct pro-inflammatory effects of CRP via CD40-CD40L signaling pathway involved in the pathogenesis of atherosclerosis, and lovastatin and fenofibrate possess anti-inflammatory effects independent of their lipid-lowering action.

[Effect of ligustrazine on injury of vascular endothelial cells ECV-304 induced by hypoxia and lack of glucose].

OBJECTIVE: To investigate the effects of ligustrazine on nitric oxide (NO), malonaldehyde (MDA) production, release of intracellular lactate dehydrogenase (LDH) and membrane fluidity of the injured human umbilical vein vascular endothelial cell line (ECV-304) with hypoxia and lack of glucose. METHOD: The experiments were performed in culture of ECV-304 injured with hypoxia and lack of glucose in vitro. The released LDH of ECV-304 was measured with automatic biochemistry analyse. NO content of ECV-304 was monitored with colorimetry. Lipid peroxidation of ECV-304 was monitored as MDA with a fluorometric assay. The membrane fluidity of ECV-304 was measured with the fluorescence polarization method. RESULT: After culture ECV-304 in hypoxia and lack of glucose for 24 h, the LDH release, MDA production and the membrane fluidity increased significantly and NO level was decreased. Preincubation of ECV-304 with ligustrazine for 24 h reduced LDH release, MDA production, membrane fluidity increasing and increased the level of NO in ECV-304 due to hypoxia and lack of glucose. CONCLUSION: Ligustrazine has protective effect on injury of ECV-304 induced by hypoxia and lack of glucose.

[Protection of vascular endothelial cells from TNF-alpha induced injury by quercetin].

OBJECTIVE: To investigate the effects of quercetin (Que) on the TNF-alpha injured human umbilical vein endothelial cell line (ECV-304). METHODS: The experiments were performed in culture of TNF-alpha-induced ECV-304 in vitro. Cell viability was assessed with MTT assay. Lipid peroxidation of ECV-304 was monitored as malonaldehyde (MDA) with a fluorometric assay. NO and SOD of ECV-304 were monitored with colorimetry. The expression of NF-kappaB was detected with immunocytochemistry. RESULTS: TNF-alpha inhibited the ECV-304 prolifteration. Preincubation of ECV-304 with Que for 24h before TNF-alpha exposure increased the cell viability. Que significantly decreased the level of NO and MDA and the expression of NF-kappaB, and increased the activity of SOD in TNF-alpha-induced ECV-304. CONCLUSION: These results demonstrate that Que can produce the protective action on TNF-alpha-induced cultured ECV-304 and its mechanism of action may be related to the decrease of NO, antioxidant effect of lipids. However, antioxidant effect might be exerted through NF-kappaB activation pathway.

[Effect of ligustrazine on cell cycle and prostacyclin of injured vascular endothelial cells induced by hypoxia and lack of glucose].

OBJECTIVE: To study the effects of ligustrazine on cell cycle and prostacyclin of injured human umbilical vein vascular endothelial cell line induced by hypoxia and lack of glucose. METHODS: The experiments were performed by culturing vascular endothelial cell induced by hypoxia and lack of glucose in vitro. Cell viability was assessed by MTT assay and cell cycle was observed by flow cytometry. Radioimmunoassay (RIA) was used to assess the amount of prostacyclin (PGI2). RESULTS: The vascular endothelial cell viability and PGI2 were decreased in condition of hypoxia and lack of glucose-injury. Preincubation of vascular endothelial cell with Ligustrazine for 24 h significantly increased the cell viability, S-Phase cell, G2M-phase cell and PGI2 level. CONCLUSION: These results demonstrated that ligustrazine had protective effect on hypoxia and lack of glucose-injured vascular endothelial cell and the effect may be related to increasing cell viability, S-phase cell, G2M-phase cell and level of PGI2.