本文提出了一种结合物理和深度生成先验的非线性盲逆框架，用于带标签MRI，实现了解剖恢复、高分辨率 cine 图像合成和运动估计的统一处理。
摘要：
带标签MRI能够非侵入性地跟踪内部组织运动。它通过用周期性标签调制解剖结构来编码运动，这些标签会随着组织变形。然而，解剖结构、标签和运动之间的纠缠对后处理构成了重大挑战。标签的存在和成像模糊阻碍了诸如解剖结构分割等下游任务。由于T1弛豫，标签褪色破坏了运动跟踪中的亮度恒定假设。几十年来，这些挑战都是孤立处理且次优的。相比之下，我们引入了一种盲非线性逆框架用于带标签MRI，首次统一了这些任务：解剖图像恢复、高分辨率 cine 图像合成和运动估计。其核心在于MR物理和生成先验的协同作用，使我们能够盲估计未知的正向成像模型和高分辨率的下层解剖结构，同时跟踪时间内的3D diffeomorphic Lagrangian运动。在带标签脑MRI上的实验表明，我们的方法产生了高分辨率解剖图像、cine图像以及比专门方法更准确的运动。
英文摘要：
Tagged MRI enables tracking internal tissue motion non-invasively. It encodes motion by modulating anatomy with periodic tags, which deform along with tissue. However, the entanglement between anatomy, tags and motion poses significant challenges for post-processing. The existence of tags and imaging blur hinders downstream tasks such as segmenting anatomy. Tag fading, due to T1-relaxation, disrupts the brightness constancy assumption for motion tracking. For decades, these challenges have been handled in isolation and sub-optimally. In contrast, we introduce a blind and nonlinear inverse framework for tagged MRI that, for the first time, unifies these tasks: anatomical image recovery, high-resolution cine image synthesis, and motion estimation. At its core, the synergy of MR physics and generative priors enables us to blindly estimate the unknown forward imaging models and high-resolution underlying anatomy, while simultaneously tracking 3D diffeomorphic Lagrangian motion over time. Experiments on tagged brain MRI demonstrate that our approach yields high-resolution anatomy images, cine images, and more accurate motion than specialized methods.
Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；University of Maryland School of Dentistry（马里兰大学牙科学院）
Tagged magnetic resonance imaging (MRI) is a widely used imaging technique for measuring tissue deformation in moving organs. Due to tagged MRI's intrinsic low anatomical resolution, another matching set of cine MRI with higher resolution is sometimes acquired in the same scanning session to facilitate tissue segmentation, thus adding extra time and cost. To mitigate this, in this work, we propose a novel dual-cycle constrained bijective VAE-GAN approach to carry out tagged-to-cine MR image synthesis. Our method is based on a variational autoencoder backbone with cycle reconstruction constrain…
Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；University of Maryland（马里兰大学）
Understanding the underlying relationship between tongue and oropharyngeal muscle deformation seen in tagged-MRI and intelligible speech plays an important role in advancing speech motor control theories and treatment of speech related-disorders. Because of their heterogeneous representations, however, direct mapping between the two modalities -- i.e., two-dimensional (mid-sagittal slice) plus time tagged-MRI sequence and its corresponding one-dimensional waveform -- is not straightforward. Instead, we resort to two-dimensional spectrograms as an intermediate representation, which contains bot…
Echocardiography to Cardiac MRI View Transformation for Real-Time Blind Restoration
Tampere University（坦佩雷大学）；Qatar University（卡塔尔大学）；VTT Technical Research Centre of Finland（芬兰国家技术研究中心）；Hamad Medical Corporation Hospital（哈马德医疗公司医院）；Hayatabad Medical Complex（海亚特阿巴德医疗中心）
Accelerating Stroke MRI with Diffusion Probabilistic Models through Large-Scale Pre-training and Target-Specific Fine-Tuning
Chandra Family Department of Electrical and Computer Engineering（查克拉家族电气与计算机工程系）；Department of Imaging Physics（成像物理系）；Dell Medical School Department of Neurology（德莱尔医学院神经学系）；Dell Medical School Department of Neurosurgery（德莱尔医学院神经外科系）；Dell Medical School Department of Diagnostic Medicine（德莱尔医学院诊断医学系）；Oden Institute for Computational Engineering and Sciences（奥登计算工程与科学研究院）
DRIMET: Deep Registration for 3D Incompressible Motion Estimation in Tagged-MRI with Application to the Tongue
Johns Hopkins University（约翰斯·霍普金斯大学）；Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）；University of Maryland School of Medicine（马里兰大学医学院）
Accurate tongue motion estimation is essential for tongue function evaluation. The harmonic phase processing (HARP) method and the phase vector incompressible registration algorithm (PVIRA) based on HARP can generate motion estimates from tagged MRI images, but they suffer from tag jumping due to large motions. This paper proposes a new registration method by combining the stationary velocity fields produced by PVIRA between successive time frames as a new initialization of the final registration stage to avoid tag jumping. The experiment results demonstrate the proposed method can avoid tag j…
Patch-Based Diffusion for Data-Efficient, Radiologist-Preferred MRI Reconstruction
Stanford University（斯坦福大学）；Stanford University School of Medicine（斯坦福大学医学院）；Stanford Center for Artificial Intelligence in Medicine and Imaging（斯坦福大学医学与成像人工智能中心）；Georgia Institute of Technology（佐治亚理工学院）
Verisk AI, Verisk Analytics（Verisk分析公司Verisk人工智能部门）；Vision and Security Technology (VAST) Lab（视觉与安全技术（VAST）实验室）
Detection and localization of image manipulations like splices are gaining in importance with the easy accessibility of image editing softwares. While detection generates a verdict for an image it provides no insight into the manipulation. Localization helps explain a positive detection by identifying the pixels of the image which have been tampered. We propose a deep learning based method for splice localization without prior knowledge of a test image's camera-model. It comprises a novel approach for learning rich filters and for suppressing image-edges. Additionally, we train our model on a …
Brightness-Invariant Tracking Estimation in Tagged MRI
Johns Hopkins University（约翰霍普金斯大学）；University of Maryland School of Medicine（马里兰大学医学院）；Uniformed Services University of the Health Sciences（统一服务卫生科学大学）；The Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc.（亨利·M·杰克逊军事医学促进基金会）；Massachusetts General Hospital and Harvard Medical School（麻省总医院和哈佛医学院）；Washington University（华盛顿大学）；University of South Carolina（南卡罗来纳大学）
Gordon Center for Medical Imaging, Massachusetts General Hospital and Harvard Medical School（戈登医学成像中心，麻省总医院与哈佛医学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；University of Maryland School of Dentistry（马里兰大学牙科学院）
Investigating the relationship between internal tissue point motion of the tongue and oropharyngeal muscle deformation measured from tagged MRI and intelligible speech can aid in advancing speech motor control theories and developing novel treatment methods for speech related-disorders. However, elucidating the relationship between these two sources of information is challenging, due in part to the disparity in data structure between spatiotemporal motion fields (i.e., 4D motion fields) and one-dimensional audio waveforms. In this work, we present an efficient encoder-decoder translation netwo…
Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；University of Maryland School of Dentistry（马里兰大学牙医学院）
Cycle reconstruction regularized adversarial training -- e.g., CycleGAN, DiscoGAN, and DualGAN -- has been widely used for image style transfer with unpaired training data. Several recent works, however, have shown that local distortions are frequent, and structural consistency cannot be guaranteed. Targeting this issue, prior works usually relied on additional segmentation or consistent feature extraction steps that are task-specific. To counter this, this work aims to learn a general add-on structural feature extractor, by explicitly enforcing the structural alignment between an input and it…
Vanderbilt University（范德堡大学）；Vanderbilt University Medical Center（范德堡大学医学中心）；National Institute on Aging（美国国家老龄化研究所）
An important task in image processing and neuroimaging is to extract quantitative information from the acquired images in order to make observations about the presence of disease or markers of development in populations. Having a lowdimensional manifold of an image allows for easier statistical comparisons between groups and the synthesis of group representatives. Previous studies have sought to identify the best mapping of brain MRI to a low-dimensional manifold, but have been limited by assumptions of explicit similarity measures. In this work, we use deep learning techniques to investigate …
Converting T1-weighted MRI from 3T to 7T quality using deep learning
Department of Clinical Sciences Malmö, SciLifeLab, Lund University（马尔摩临床科学系、SciLifeLab、吕勒奥大学）；Univ Rennes, CNRS, Inria, Inserm, IRISA UMR 6074, Empenn ERL U 1228（雷恩大学、CNRS、Inria、Inserm、IRISA UMR 6074、Empenn ERL U 1228）；Clinical Memory Research Unit, Department of Clinical Sciences Malmö, Lund University（临床记忆研究单位、马尔摩临床科学系、吕勒奥大学）；Centre for Mathematical Sciences, Lund University（数学科学中心、吕勒奥大学）；Diagnostic Radiology Unit, Department of Clinical Sciences Lund, Lund University（诊断放射学单位、临床科学系、吕勒奥大学）；Alzheimer Center Amsterdam, Vrije Universiteit Amsterdam, Amsterdam UMC（阿姆斯特丹阿尔茨海默病中心、阿姆斯特丹自由大学、阿姆斯特丹大学医学中心）；Amsterdam Neuroscience, Neurodegeneration, Amsterdam, The Netherlands（阿姆斯特丹神经科学、神经退行性疾病、阿姆斯特丹、荷兰）；Memory Clinic, Skåne University Hospital, Malmö, Sweden（记忆诊所、斯德哥尔摩大学医院、马尔摩、瑞典）；German Center for Neurodegenerative Diseases, Magdeburg, Germany（德国神经退行性疾病中心、马格德堡、德国）；Center for Behavioral Brain Sciences, Otto-von-Guericke University Magdeburg, Magdeburg, Germany（行为脑科学中心、奥托·冯·格里克大学马格德堡、马格德堡、德国）；Visual Recognition Group, Faculty of Electrical Engineering, Czech Technical University in Prague（视觉识别组、电气工程系、布拉格技术大学）
研究旨在用深度学习从 3T 脑 MRI 合成 7T 脑 MRI，训练了专门 U - Net 和 GAN U - Net 两个模型，其在评估指标上优于此前模型，合成图像在多方面表现良好，可提高图像质量与分割效果且不影响下游任务性能。
Speech motion anomaly detection via cross-modal translation of 4D motion fields from tagged MRI
Gordon Center for Medical Imaging, Massachusetts General Hospital and Harvard Medical School（戈登医学影像中心，麻省总医院与哈佛医学院）；University of Maryland School of Medicine（马里兰大学医学院）；University of Maryland School of Dentistry（马里兰大学牙医学院）；Johns Hopkins University（约翰斯·霍普金斯大学）；Yale University（耶鲁大学）
Image Quality Assessment (IQA) is of great value in the workflow of Magnetic Resonance Imaging (MRI)-based analysis. Blind IQA (BIQA) methods are especially required since high-quality reference MRI images are usually not available. Recently, many efforts have been devoted to developing deep learning-based BIQA approaches. However, the performance of these methods is limited due to the utilization of simple content-non-adaptive network parameters and the waste of the important 3D spatial information of the medical images. To address these issues, we design a 3D content-adaptive hyper-network f…
We present a method for estimating intravoxel parameters from a DW-MRI based on deep learning techniques. We show that neural networks (DNNs) have the potential to extract information from diffusion-weighted signals to reconstruct cerebral tracts. We present two DNN models: one that estimates the axonal structure in the form of a voxel and the other to calculate the structure of the central voxel using the voxel neighborhood. Our methods are based on a proposed parameter representation suitable for the problem. Since it is practically impossible to have real tagged data for any acquisition pro…
This paper proposes a novel method for automatic MRI denoising that exploits last advances in deep learning feature regression and self-similarity properties of the MR images. The proposed method is a two-stage approach. In the first stage, an overcomplete patch-based convolutional neural network blindly removes the noise without specific estimation of the local noise variance to produce a preliminary estimation of the noise-free image. The second stage uses this preliminary denoised image as a guide image within a rotationally invariant non-local means filter to robustly denoise the original …
Is Registering Raw Tagged-MR Enough for Strain Estimation in the Era of Deep Learning?
Johns Hopkins University（约翰斯·霍普金斯大学）；University of South Carolina（南卡罗来纳大学）；Massachusetts General Hospital and Harvard Medical School（马萨诸塞总医院和哈佛医学院）；Uniformed Services University of the Health Sciences（美国健康科学统一服务大学）；University of Maryland School of Medicine（马里兰大学医学院）
Facebook AI Research（Facebook人工智能研究）；Facebook New York（纽约Facebook）；NYU Grossman School of Medicine（纽约大学格罗斯曼医学院）；Department of Radiology（放射科）
MRI images reconstructed from sub-sampled Cartesian data using deep learning techniques often show a characteristic banding (sometimes described as streaking), which is particularly strong in low signal-to-noise regions of the reconstructed image. In this work, we propose the use of an adversarial loss that penalizes banding structures without requiring any human annotation. Our technique greatly reduces the appearance of banding, without requiring any additional computation or post-processing at reconstruction time. We report the results of a blind comparison against a strong baseline by a gr…
Training Adaptive Reconstruction Networks for Blind Inverse Problems
Institut de Mathématiques de Toulouse (IMT)（图卢兹数学研究所）；Université de Toulouse（图卢兹大学）；CNRS（法国国家科学研究中心）；Institut de Recherche en Informatique de Toulouse (IRIT)（图卢兹计算机科学研究所）；Centre de Biologie Intégrative (CBI)（综合生物学中心）
NIMARC-MRI: Abdominal HASTE Dataset and a Baseline U-Net Exposing the Synthetic-to-Real Gap in Low-Resource Motion Correction
Federal University of Health Sciences, Azare（阿扎雷联邦健康科学大学）；African Institute for Research Advancement and Innovation (AIRA Africa)（非洲研究与创新促进研究所）；Bayero University Kano（卡诺拜鲁克大学）；Medserve Kano Diagnostic Center（卡诺医疗服务中心）；Radiology, University of Maiduguri Teaching Hospital（迈杜古里教学医院放射科）；PET Imaging Centre, King’s College London, University of London（伦敦大学国王学院正电子发射断层扫描成像中心）；Northwest University kano（卡诺西北大学）
Anatomically Faithful Artifact Suppression in SENSE Accelerated Brain MRI
East China Normal University（华东师范大学）；Affiliated Hospital of Guizhou Medical University（贵州医科大学附属医院）；West China Hospital, Sichuan University（四川大学华西医院）；Chinese Academy of Medical Sciences（中国医学科学院）；RadioDynamic Medical（镭动医疗）；Lucile Packard Children’s Hospital（露西尔·帕卡德儿童医院）
A cross-modal generative model for incomplete and degraded prostate MRI with multicentre clinical validation
College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算与数据科学学院）；School of Computer Science and Technology, Tongji University（同济大学计算机科学与技术学院）；Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine（上海交通大学医学院附属仁济医院）；National University of Singapore（新加坡国立大学）；Shanghai East Hospital, Tongji University School of Medicine（同济大学医学院附属东方医院）；McWilliams School of Biomedical Informatics, The University of Texas Health Science Center at Houston (UTHealth Houston)（德克萨斯大学休斯顿健康科学中心麦克威廉姆斯生物信息学院）；School of Transportation, Southeast University（东南大学交通学院）；Ningbo Hangzhou Bay Hospital（宁波杭州湾医院）；Changhai Hospital, Naval Medical University（海军军医大学长海医院）
Nanyang Technological University（南洋理工大学）；Michigan State University（密歇根州立大学）；University of Illinois, Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Magnetic resonance imaging (MRI) is widely used in clinical practice, but it has been traditionally limited by its slow data acquisition. Recent advances in compressed sensing (CS) techniques for MRI reduce acquisition time while maintaining high image quality. Whereas classical CS assumes the images are sparse in known analytical dictionaries or transform domains, methods using learned image models for reconstruction have become popular. The model could be pre-learned from datasets, or learned simultaneously with the reconstruction, i.e., blind CS (BCS). Besides the well-known synthesis dicti…
MedForj: An open, large-scale foundational generative prior for high-resolution 3D brain MRI
Image Analysis and Communications Laboratory, Johns Hopkins University（约翰霍普金斯大学图像分析与通信实验室）；Department of Neurology, Johns Hopkins Hospital for the Alzheimer’s Disease Neuroimaging Initiative（约翰霍普金斯医院阿尔茨海默病神经影像计划神经科）；for the Australian Imaging Biomarkers and Lifestyle Flagship Study of Aging（澳大利亚成像生物标志物与生活方式老龄化旗舰研究）；for the Health and Aging Brain Study (HABS-HD) Study Team（健康与老龄化大脑研究（HABS-HD）研究团队）；for the Mayo Clinic Study of Aging（梅奥诊所老龄化研究）
Evaluating the Safety of Deep Learning-Based Brain MRI Reconstruction
School of Science, Engineering & Technology, RMIT University Vietnam（RMIT越南大学科学、工程与技术学院）；School of Computing Technologies, RMIT University（RMIT大学计算技术学院）；School of Health and Biomedical Science, RMIT University（RMIT大学健康与生物医学科学学院）
University of Tehran（德黑兰大学）；University of Nottingham（诺丁汉大学）
In this study, an automated three dimensional (3D) deep segmentation approach for detecting gliomas in 3D pre-operative MRI scans is proposed. Then, a classi-fication algorithm based on random forests, for survival prediction is presented. The objective is to segment the glioma area and produce segmentation labels for its different sub-regions, i.e. necrotic and the non-enhancing tumor core, the peri-tumoral edema, and enhancing tumor. The proposed deep architecture for the segmentation task encompasses two parallel streamlines with two different reso-lutions. One deep convolutional neural net…
Deep Separable Spatiotemporal Learning for Fast Dynamic Cardiac MRI
Xiamen University（厦门大学）；Fudan University（复旦大学）；Xiamen Cardiovascular Hospital of Xiamen University（厦门大学附属厦门心血管病医院）；School of Medicine, Xiamen University（厦门大学医学院）；Ruijin Hospital Lu Wan Branch, Shanghai Jiaotong University School of Medicine（上海交通大学医学院附属瑞金医院卢湾分院）；Shanghai Institute of Cardiovascular Diseases（上海市心血管病研究所）；Zhongshan Hospital, Fudan University（复旦大学附属中山医院）；Shanghai Medical School, Fudan University（复旦大学上海医学院）；Shanghai Institute of Medical Imaging（上海市医学影像研究所）；Xiamen University of Technology（厦门理工学院）；Imperial College London（帝国理工学院）
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）；Massachusetts Institute of Technology（麻省理工学院）
Image corruption by motion artifacts is an ingrained problem in Magnetic Resonance Imaging (MRI). In this work, we propose a neural network-based regularization term to enhance Autofocusing, a classic optimization-based method to remove motion artifacts. The method takes the best of both worlds: the optimization-based routine iteratively executes the blind demotion and deep learning-based prior penalizes for unrealistic restorations and speeds up the convergence. We validate the method on three models of motion trajectories, using synthetic and real noisy data. The method proves resilient to n…
Deep Learning-based Intraoperative MRI Reconstruction
Oslo University Hospital（奥斯陆大学医院）；University of Oslo（奥斯陆大学）；University of South-Eastern Norway（挪威东南大学）；NTNU（挪威科技大学）；Vestfold Hospital Trust（维斯特福德医院信托）；Amsterdam UMC, University of Amsterdam（阿姆斯特丹大学医学中心，阿姆斯特丹大学）
