Synthesizing PET images from high-field and ultra-high-field MR images using joint diffusion attention model.

BACKGROUND: Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) stand as pivotal diagnostic tools for brain disorders, offering the potential for mutually enriching disease diagnostic perspectives. However, the costs associated with PET scans and the inherent radioactivity have limited the widespread application of PET. Furthermore, it is noteworthy to highlight the promising potential of high-field and ultra-high-field neuroimaging in cognitive neuroscience research and clinical practice. With the enhancement of MRI resolution, a related question arises: can high-resolution MRI improve the quality of PET images? PURPOSE: This study aims to enhance the quality of synthesized PET images by leveraging the superior resolution capabilities provided by high-field and ultra-high-field MRI. METHODS: From a statistical perspective, the joint probability distribution is considered the most direct and fundamental approach for representing the correlation between PET and MRI. In this study, we proposed a novel model, the joint diffusion attention model, namely, the joint diffusion attention model (JDAM), which primarily focuses on learning information about the joint probability distribution. JDAM consists of two primary processes: the diffusion process and the sampling process. During the diffusion process, PET gradually transforms into a Gaussian noise distribution by adding Gaussian noise, while MRI remains fixed. The central objective of the diffusion process is to learn the gradient of the logarithm of the joint probability distribution between MRI and noise PET. The sampling process operates as a predictor-corrector. The predictor initiates a reverse diffusion process, and the corrector applies Langevin dynamics. RESULTS: Experimental results from the publicly available Alzheimer's Disease Neuroimaging Initiative dataset highlight the effectiveness of the proposed model compared to state-of-the-art (SOTA) models such as Pix2pix and CycleGAN. Significantly, synthetic PET images guided by ultra-high-field MRI exhibit marked improvements in signal-to-noise characteristics when contrasted with those generated from high-field MRI data. These results have been endorsed by medical experts, who consider the PET images synthesized through JDAM to possess scientific merit. This endorsement is based on their symmetrical features and precise representation of regions displaying hypometabolism, a hallmark of Alzheimer's disease. CONCLUSIONS: This study establishes the feasibility of generating PET images from MRI. Synthesis of PET by JDAM significantly enhances image quality compared to SOTA models.

The Prognostic Value of Left Ventricular Entropy From T1 Mapping in Patients With Hypertrophic Cardiomyopathy.

Background: The prognostic value of left ventricular (LV) entropy in hypertrophic cardiomyopathy (HCM) is unclear. Objectives: This study aimed to assess the prognostic value of LV entropy from T1 mapping in HCM. Methods: A total of 748 participants with HCM, who underwent cardiovascular magnetic resonance (CMR), were consecutively enrolled. LV entropy was quantified by native T1 mapping. A competing risk analysis and a Cox proportional hazards regression analysis were performed to identify potential associations of LV entropy with sudden cardiac death (SCD) and cardiovascular death (CVD), respectively. Results: A total of 40 patients with HCM experienced SCD, and 65 experienced CVD during a median follow-up of 43 months. Participants with increased LV entropy (≥4.06) were more likely to experience SCD and CVD (all P < 0.05) in the entire study cohort or the subgroup with low late gadolinium enhancement (LGE) extent (<15%). After adjustment for the European Society of Cardiology predictors and the presence of high LGE extent (≥15%), LV mean entropy was an independent predictor for SCD (HR: 1.03; all P < 0.05) by the multivariable competing risk analysis and CVD (HR: 1.06; 95% CI: 1.03-1.09; P < 0.001) by multivariable Cox regression analysis. Conclusions: LV mean entropy derived from native T1 mapping, reflecting myocardial tissue heterogeneity, was an independent predictor of SCD and CVD in participants with HCM. (Cardiac Magnetic Resonance Imaging Clinical Application Registration Study; ChiCTR1900024094).

An orally administered bacterial membrane protein nanodrug ameliorates doxorubicin cardiotoxicity through alleviating impaired intestinal barrier.

The cardiotoxicity caused by Dox chemotherapy represents a significant limitation to its clinical application and is a major cause of late death in patients undergoing chemotherapy. Currently, there are no effective treatments available. Our analysis of 295 clinical samples from 132 chemotherapy patients and 163 individuals undergoing physical examination revealed a strong positive correlation between intestinal barrier injury and the development of cardiotoxicity in chemotherapy patients. We developed a novel orally available and intestinal targeting protein nanodrug by assembling membrane protein Amuc_1100 (obtained from intestinal bacteria Akkermansia muciniphila), fluorinated polyetherimide, and hyaluronic acid. The protein nanodrug demonstrated favorable stability against hydrolysis compared with free Amuc_1100. The in vivo results demonstrated that the protein nanodrug can alleviate Dox-induced cardiac toxicity by improving gut microbiota, increasing the proportion of short-chain fatty acid-producing bacteria from the Lachnospiraceae family, and further enhancing the levels of butyrate and pentanoic acids, ultimately regulating the homeostasis repair of lymphocytes in the spleen and heart. Therefore, we believe that the integrity of the intestinal barrier plays an important role in the development of chemotherapy-induced cardiotoxicity. Protective interventions targeting the intestinal barrier may hold promise as a general clinical treatment regimen for reducing Dox-induced cardiotoxicity.

DPP: deep phase prior for parallel imaging with wave encoding.

Objective.In Magnetic Resonance (MR) parallel imaging with virtual channel-expanded Wave encoding, limitations are imposed on the ability to comprehensively and accurately characterize the background phase. These limitations are primarily attributed to the calibration process relying solely on center low-frequency Auto-Calibration Signals (ACS) data for calibration.Approach.To tackle the challenge of accurately estimating the background phase in wave encoding, a novel deep neural network model guided by deep phase priors is proposed with integrated virtual conjugate coil (VCC) extension. Concretely, within the proposed framework, the background phase is implicitly characterized by employing a carefully designed decoder convolutional neural network, leveraging the inherent characteristics of phase smoothness and compact support in the transformed domain. Furthermore, the proposed model with wave encoding benefits from additional priors, which incorporate transmission sparsity of the latent image and coil sensitivity smoothness.Main results.Ablation experiments were conducted to ascertain the proposed method's capability to implicitly represent CSM and the background phase. Subsequently, the superiority of the proposed method is demonstrated through confidence comparisons with competing methods, employing 4-fold and 5-fold acceleration experiments. In achieving 4-fold and 5-fold acceleration, the optimal quantitative metrics (PSNR/SSIM/NMSE) are 44.1359 dB/0.9863/0.0008 (4-fold) and 41.2074/0.9846/0.0017 (5-fold), respectively. Furthermore, the generalizability of the proposed method is further validated by conducting acceleration experiments with T1, T2, T2*, and various undersampling patterns. In addition, the DPP delivered much better performance than the conventional methods by exploring accelerated phase-sensitive SWI imaging. In SWI accelerated imaging, it also surpasses the optimal competing method in terms of (PSNR/SSIM/NMSE) with 0.096%/0.009%/0.0017%.Significance.The proposed method enables precise characterization of the background phase in the integrated VCC and wave encoding framework, supported via theoretical analysis and empirical findings. Our code is available at:https://github.com/sober235/DPP.

Integrating data distribution prior via Langevin dynamics for end-to-end MR reconstruction.

PURPOSE: To develop a novel deep learning-based method inheriting the advantages of data distribution prior and end-to-end training for accelerating MRI. METHODS: Langevin dynamics is used to formulate image reconstruction with data distribution before facilitate image reconstruction. The data distribution prior is learned implicitly through the end-to-end adversarial training to mitigate the hyper-parameter selection and shorten the testing time compared to traditional probabilistic reconstruction. By seamlessly integrating the deep equilibrium model, the iteration of Langevin dynamics culminates in convergence to a fix-point, ensuring the stability of the learned distribution. RESULTS: The feasibility of the proposed method is evaluated on the brain and knee datasets. Retrospective results with uniform and random masks show that the proposed method demonstrates superior performance both quantitatively and qualitatively than the state-of-the-art. CONCLUSION: The proposed method incorporating Langevin dynamics with end-to-end adversarial training facilitates efficient and robust reconstruction for MRI. Empirical evaluations conducted on brain and knee datasets compellingly demonstrate the superior performance of the proposed method in terms of artifact removing and detail preserving.

A Two-Stage Generative Model with CycleGAN and Joint Diffusion for MRI-based Brain Tumor Detection.

Accuratedetection and segmentation of brain tumors is critical for medical diagnosis. However, current supervised learning methods require extensively annotated images and the state-of-the-art generative models used in unsupervised methods often have limitations in covering the whole data distribution. In this paper, we propose a novel framework Two-Stage Generative Model (TSGM) that combines Cycle Generative Adversarial Network (CycleGAN) and Variance Exploding stochastic differential equation using joint probability (VE-JP) to improve brain tumor detection and segmentation. The CycleGAN is trained on unpaired data to generate abnormal images from healthy images as data prior. Then VE-JP is implemented to reconstruct healthy images using synthetic paired abnormal images as a guide, which alters only pathological regions but not regions of healthy. Notably, our method directly learned the joint probability distribution for conditional generation. The residual between input and reconstructed images suggests the abnormalities and a thresholding method is subsequently applied to obtain segmentation results. Furthermore, the multimodal results are weighted with different weights to improve the segmentation accuracy further. We validated our method on three datasets, and compared with other unsupervised methods for anomaly detection and segmentation. The DSC score of 0.8590 in BraTs2020 dataset, 0.6226 in ITCS dataset and 0.7403 in In-house dataset show that our method achieves better segmentation performance and has better generalization.

Coumarin-embedded [5]helicene derivatives: synthesis, X-ray analysis and photoconducting properties.

Two novel coumarin-embedded π-extended [5]helicene derivatives (3a and 6a) have been strategically synthesized and characterized, and the structure of 3a was determined via single crystal X-ray analysis. Both of them exhibit green fluorescence in dichloromethane. In addition, molecule 3a can aggregate to form a large quantity of nanowires through the re-precipitation method. More importantly, the photoelectric conversion properties of 3a nanowire-C60 based films are much better than those of the thin film of bulk 3a-C60, indicating that the ordered nanostructures are a crucial factor for enhancing device performance.

Clinical Outcomes of Percutaneous Transforaminal Endoscopic Discectomy Assisted with Selective Nerve Root Block for Treating Radicular Pain with Diagnostic Uncertainty in the Elderly.

Purpose: To investigate the clinical outcomes of percutaneous transforaminal endoscopic discectomy assisted with selective nerve root block for treating radicular pain with diagnostic uncertainty in the elderly. Methods: A total number of 36 elderly patients were included in the study. Clinical outcomes collected for analysis include operative time, hospital stay time, Visual Analog Scale, and Oswestry Disability Index before and after the surgery, the global outcome based on the Macnab outcome criteria. Results: Seventeen males and nineteen females with a mean age of 73.72 ± 7.15 were included in this study. Radicular pain was the main complaint of all the patients with the least symptom duration of two months. Radiological findings showed that 80.6% of the patients with multilevel disc herniation, 16.7% received lumbar fusion surgery before, and 8.3% with degenerative scoliosis. Besides, 69.4% of the patients have at least one comorbidity. 85.4% of the patients showed a positive response to selective nerve root block, and 91.6% of the patients reported a favorable outcome at the last follow-up. The mean value of pre-operative leg pain was 7.56 ± 0.74 and dramatically decreased after surgery (2.47 ± 0.81, P < 0.001). Besides, the mean value of Oswestry Disability Index decreased from 43.03 ± 4.43 to 5.92 ± 5.24 (P < 0.001) one year after the surgery. Conclusion: Multilevel degeneration of the lumbar spine is common in elderly patients. Identifying the responsible segment and decompressing the nerve root through minimally invasive surgery can provide a satisfactory clinical outcome for those with radicular pain as their primary complaint. And selective nerve root block is a reliable diagnostic tool for those with an ambiguous diagnosis.

High efficiency free-breathing 3D thoracic aorta vessel wall imaging using self-gating image reconstruction.

PURPOSE: To improve the scan efficiency of thoracic aorta vessel wall imaging using a self-gating (SG)-based motion correction scheme. MATERIALS AND METHODS: A slab-selective variable-flip-angle 3D turbo spin-echo (SPACE) sequence was modified to acquire SG signals and imaging data. Cartesian sampling with a tiny golden-step spiral profile ordering was used to obtain the imaging data during the systolic period, and then the image data were subsequently corrected based on the SG signals and binned to different respiratory cycles. Finally, respiratory artifacts were estimated from image-based registration of 3D undersampled respiratory bins that were reconstructed with L1 iterative self-consistent parallel imaging reconstruction (SPIRiT). This method was evaluated in 11 healthy volunteers and compared against conventional diaphragmatic navigator-gated acquisition to assess the feasibility of the proposed framework. RESULTS: Results showed that the proposed method achieved image quality comparable to that of conventional diaphragmatic navigator-gated acquisition with an average scan time of 4 min. The sharpness of the vessel wall and the definition of the liver boundary were in good agreement with the navigator-gated acquisition, which took approximately above 8.5 min depend on the respiratory rate. Further valuation of this technique in patients will be conducted to determine its clinical use.

High-Frequency Space Diffusion Model for Accelerated MRI.

Diffusion models with continuous stochastic differential equations (SDEs) have shown superior performances in image generation. It can serve as a deep generative prior to solving the inverse problem in magnetic resonance (MR) reconstruction. However, low-frequency regions of k -space data are typically fully sampled in fast MR imaging, while existing diffusion models are performed throughout the entire image or k -space, inevitably introducing uncertainty in the reconstruction of low-frequency regions. Additionally, existing diffusion models often demand substantial iterations to converge, resulting in time-consuming reconstructions. To address these challenges, we propose a novel SDE tailored specifically for MR reconstruction with the diffusion process in high-frequency space (referred to as HFS-SDE). This approach ensures determinism in the fully sampled low-frequency regions and accelerates the sampling procedure of reverse diffusion. Experiments conducted on the publicly available fastMRI dataset demonstrate that the proposed HFS-SDE method outperforms traditional parallel imaging methods, supervised deep learning, and existing diffusion models in terms of reconstruction accuracy and stability. The fast convergence properties are also confirmed through theoretical and experimental validation. Our code and weights are available at https://github.com/Aboriginer/HFS-SDE.

Three in One: Three Different Molybdates Trapped in a Thiacalix[4]arene Protected Ag72 Nanocluster for Structural Transformation and Photothermal Conversion.

Polyoxometalates (POMs) represent crucial intermediates in the formation of insoluble metal oxides from soluble metal ions, however, the rapid hydrolysis-condensation kinetics of MoVI or WVI makes the direct characterization of coexisted molecular species in a given medium extremely difficult. Silver nanoclusters have shown versatile capacity to encapsulate diverse POMs, which provides an alternative scene to appreciate landscape of POMs in atomic precision. Here, we report a thiacalix[4]arene protected silver nanocluster (Ag72b) that simultaneously encapsulates three kinds of molybdates (MoO4 2- , Mo6 O22 8- and Mo7 O25 8- ) in situ transformed from classic Lindqvist Mo6 O19 2- , providing more deep understanding on the structural diversity and condensation growth route of POMs in solution. Ag72b is the first silver nanocluster trapping so many kinds of molybdates, which in turn exert collective template effect to aggregate silver atoms into a nanocluster. The post-reaction of Ag72b with AgOAc or PhCOOAg produces a discrete Ag24 nanocluster (Ag24a) or an Ag28 nanocluster based 1D chain structure (Ag28a), respectively. Moreover, the post-synthesized Ag28a can be utilized as potential ignition material for further application. This work not only provides an important model for unlocking dynamic features of POMs at atom-precise level but also pioneers a promising approach to synthesize silver nanoclusters from known to unknown.

Annihilation-Net: Learned annihilation relation for dynamic MR imaging.

BACKGROUND: Deep learning methods driven by the low-rank regularization have achieved attractive performance in dynamic magnetic resonance (MR) imaging. The effectiveness of existing methods lies mainly in their ability to capture interframe relationships using network modules, which are lack interpretability. PURPOSE: This study aims to design an interpretable methodology for modeling interframe relationships using convolutiona networks, namely Annihilation-Net and use it for accelerating dynamic MRI. METHODS: Based on the equivalence between Hankel matrix product and convolution, we utilize convolutional networks to learn the null space transform for characterizing low-rankness. We employ low-rankness to represent interframe correlations in dynamic MR imaging, while combining with sparse constraints in the compressed sensing framework. The corresponding optimization problem is solved in an iterative form with the semi-quadratic splitting method (HQS). The iterative steps are unrolled into a network, dubbed Annihilation-Net. All the regularization parameters and null space transforms are set as learnable in the Annihilation-Net. RESULTS: Experiments on the cardiac cine dataset show that the proposed model outperforms other competing methods both quantitatively and qualitatively. The training set and test set have 800 and 118 images, respectively. CONCLUSIONS: The proposed Annihilation-Net improves the reconstruction quality of accelerated dynamic MRI with better interpretability.

Developing a Dynamic Graph Network for Interpretable Analysis of Multi-Modal MRI Data in Parkinson's Disease Diagnosis.

Following the aging of the population, Parkinson's disease (PD) poses a severe challenge to public health. For the diagnosis of PD and the prediction of its progression, numerous computer-aided diagnosis procedures have been developed. Recently, Graph Convolutional Networks (GCN) are widely applied in deep learning to effectively integrate multi-modal features and model subject correlation. However, many GCNs which are used for node classification build large-scale fixed graph topologies using the entire dataset, which could make them impossible to verify independently. Furthermore, past GCN algorithms would need more interpretability, limiting their real-world applications. In this paper, an Interpretable Graph-Learning Convolutional Network (iGLCN) is proposed to enhance the performance of personalized diagnosis for PD while simultaneously producing interpretable results. The proposed method can dynamically adjust the graph structure for GCN to better diagnose outcomes by learning the optimal underlying latent graph. Through interpretable feature learning, the proposed network can interpret diagnosis outcomes. The experiments showed that the proposed method increased flexibility while maintaining a high level of classification performance and could be interpretable for PD diagnosis.Clinical Relevance- The proposed method is expected to have good performance in its strong practicability, feasibility, and interpretability for Parkinson's disease diagnosis.

Predicting Antiseizure Medication Treatment in Children with Rare Tuberous Sclerosis Complex-Related Epilepsy Using Deep Learning.

BACKGROUND AND PURPOSE: Tuberous sclerosis complex disease is a rare, multisystem genetic disease, but appropriate drug treatment allows many pediatric patients to have positive outcomes. The purpose of this study was to predict the effectiveness of antiseizure medication treatment in children with tuberous sclerosis complex-related epilepsy. MATERIALS AND METHODS: We conducted a retrospective study involving 300 children with tuberous sclerosis complex-related epilepsy. The study included the analysis of clinical data and T2WI and FLAIR images. The clinical data consisted of sex, age of onset, age at imaging, infantile spasms, and antiseizure medication numbers. To forecast antiseizure medication treatment, we developed a multitechnique deep learning method called WAE-Net. This method used multicontrast MR imaging and clinical data. The T2WI and FLAIR images were combined as FLAIR3 to enhance the contrast between tuberous sclerosis complex lesions and normal brain tissues. We trained a clinical data-based model using a fully connected network with the above-mentioned variables. After that, a weighted-average ensemble network built from the ResNet3D architecture was created as the final model. RESULTS: The experiments had shown that age of onset, age at imaging, infantile spasms, and antiseizure medication numbers were significantly different between the 2 drug-treatment outcomes (P < .05). The hybrid technique of FLAIR3 could accurately localize tuberous sclerosis complex lesions, and the proposed method achieved the best performance (area under the curve = 0.908 and accuracy of 0.847) in the testing cohort among the compared methods. CONCLUSIONS: The proposed method could predict antiseizure medication treatment of children with rare tuberous sclerosis complex-related epilepsy and could be a strong baseline for future studies.

PEARL: Cascaded Self-supervised Cross-fusion Learning For Parallel MRI Acceleration.

Supervised deep learning (SDL) methodology holds promise for accelerated magnetic resonance imaging (AMRI) but is hampered by the reliance on extensive training data. Some self-supervised frameworks, such as deep image prior (DIP), have emerged, eliminating the explicit training procedure but often struggling to remove noise and artifacts under significant degradation. This work introduces a novel self-supervised accelerated parallel MRI approach called PEARL, leveraging a multiple-stream joint deep decoder with two cross-fusion schemes to accurately reconstruct one or more target images from compressively sampled k-space. Each stream comprises cascaded cross-fusion sub-block networks (SBNs) that sequentially perform combined upsampling, 2D convolution, joint attention, ReLU activation and batch normalization (BN). Among them, combined upsampling and joint attention facilitate mutual learning between multiple-stream networks by integrating multi-parameter priors in both additive and multiplicative manners. Long-range unified skip connections within SBNs ensure effective information propagation between distant cross-fusion layers. Additionally, incorporating dual-normalized edge-orientation similarity regularization into the training loss enhances detail reconstruction and prevents overfitting. Experimental results consistently demonstrate that PEARL outperforms the existing state-of-the-art (SOTA) self-supervised AMRI technologies in various MRI cases. Notably, 5-fold  âˆ¼ 6-fold accelerated acquisition yields a 1 %  âˆ¼  2 % improvement in SSIM ROI and a 3 %  âˆ¼  6 % improvement in PSNR ROI, along with a significant 15 %  âˆ¼  20 % reduction in RLNE ROI.

A route to metalloligands consolidated silver nanoclusters by grafting thiacalix[4]arene onto polyoxovanadates.

Metalloligands provide a potent strategy for manipulating the surface metal arrangements of metal nanoclusters, but their synthesis and subsequent installation onto metal nanoclusters remains a significant challenge. Herein, two atomically precise silver nanoclusters {Ag14[(TC4A)6(V9O16)](CyS)3} (Ag14) and {Ag43S[(TC4A)2(V4O9)]3(CyS)9(PhCOO)3Cl3(SO4)4(DMF)3·6DMF} (Ag43) are synthesized by controlling reaction temperature (H4TC4A = p-tert-butylthiacalix[4]arene). Interestingly, the 3D scaffold-like [(TC4A)6(V9O16)]11- metalloligand in Ag14 and 1D arcuate [(TC4A)2(V4O9)]6- metalloligand in Ag43 exhibit a dual role that is the internal polyoxovanadates as anion template and the surface TC4A4- as the passivating agent. Furthermore, the thermal-induced structure transformation between Ag14 and Ag43 is achieved based on the temperature-dependent assembly process. Ag14 shows superior photothermal conversion performance than Ag43 in solid state indicating its potential for remote laser ignition. Here, we show the potential of two thiacalix[4]arene modified polyoxovanadates metalloligands in the assembly of metal nanoclusters and provide a cornerstone for the remote laser ignition applications of silver nanoclusters.

Recognizing Pediatric Tuberous Sclerosis Complex Based on Multi-Contrast MRI and Deep Weighted Fusion Network.

Multi-contrast magnetic resonance imaging (MRI) is wildly applied to identify tuberous sclerosis complex (TSC) children in a clinic. In this work, a deep convolutional neural network with multi-contrast MRI is proposed to diagnose pediatric TSC. Firstly, by combining T2W and FLAIR images, a new synthesis modality named FLAIR3 was created to enhance the contrast between TSC lesions and normal brain tissues. After that, a deep weighted fusion network (DWF-net) using a late fusion strategy is proposed to diagnose TSC children. In experiments, a total of 680 children were enrolled, including 331 healthy children and 349 TSC children. The experimental results indicate that FLAIR3 successfully enhances the visibility of TSC lesions and improves the classification performance. Additionally, the proposed DWF-net delivers a superior classification performance compared to previous methods, achieving an AUC of 0.998 and an accuracy of 0.985. The proposed method has the potential to be a reliable computer-aided diagnostic tool for assisting radiologists in diagnosing TSC children.

MicroRNA-124-3p Attenuated Retinal Neovascularization in Oxygen-Induced Retinopathy Mice by Inhibiting the Dysfunction of Retinal Neuroglial Cells through STAT3 Pathway.

MicroRNA (miRNA) is a non-coding RNA that can regulate the expression of many target genes, and it is widely involved in various important physiological activities. MiR-124-3p was found to associate with the normal development of retinal vessels in our previous study, but the mechanism of its anti-angiogenic effect on pathological retinal neovascularization still needed to be explored. Therefore, this study aimed to investigate the effect and mechanism of miR-124-3p on retinal neovascularization in mice with oxygen-induced retinopathy (OIR). Here, we found that intravitreal injection of miR-124-3p agomir attenuated pathological retinal neovascularization in OIR mice. Moreover, miR-124-3p preserved the astrocytic template, inhibited reactive gliosis, and reduced the inflammatory response as well as necroptosis. Furthermore, miR-124-3p inhibited the signal transducer and activator of transcription 3 (STAT3) pathway and decreased the expression of hypoxia-inducible factor-1α and vascular endothelial growth factor. Taken together, our results revealed that miR-124-3p inhibited retinal neovascularization and neuroglial dysfunction by targeting STAT3 in OIR mice.

Equilibrated Zeroth-Order Unrolled Deep Network for Parallel MR Imaging.

In recent times, model-driven deep learning has evolved an iterative algorithm into a cascade network by replacing the regularizer's first-order information, such as the (sub)gradient or proximal operator, with a network module. This approach offers greater explainability and predictability compared to typical data-driven networks. However, in theory, there is no assurance that a functional regularizer exists whose first-order information matches the substituted network module. This implies that the unrolled network output may not align with the regularization models. Furthermore, there are few established theories that guarantee global convergence and robustness (regularity) of unrolled networks under practical assumptions. To address this gap, we propose a safeguarded methodology for network unrolling. Specifically, for parallel MR imaging, we unroll a zeroth-order algorithm, where the network module serves as a regularizer itself, allowing the network output to be covered by a regularization model. Additionally, inspired by deep equilibrium models, we conduct the unrolled network before backpropagation to converge to a fixed point and then demonstrate that it can tightly approximate the actual MR image. We also prove that the proposed network is robust against noisy interferences if the measurement data contain noise. Finally, numerical experiments indicate that the proposed network consistently outperforms state-of-the-art MRI reconstruction methods, including traditional regularization and unrolled deep learning techniques.

K-UNN: k-space interpolation with untrained neural network.

Recently, untrained neural networks (UNNs) have shown satisfactory performances for MR image reconstruction on random sampling trajectories without using additional full-sampled training data. However, the existing UNN-based approaches lack the modeling of physical priors, resulting in poor performance in some common scenarios (e.g., partial Fourier (PF), regular sampling, etc.) and the lack of theoretical guarantees for reconstruction accuracy. To bridge this gap, we propose a safeguarded k-space interpolation method for MRI using a specially designed UNN with a tripled architecture driven by three physical priors of the MR images (or k-space data), including transform sparsity, coil sensitivity smoothness, and phase smoothness. We also prove that the proposed method guarantees tight bounds for interpolated k-space data accuracy. Finally, ablation experiments show that the proposed method can characterize the physical priors of MR images well. Additionally, experiments show that the proposed method consistently outperforms traditional parallel imaging methods and existing UNNs, and is even competitive against supervised-trained deep learning methods in PF and regular undersampling reconstruction.