Bias-Reduced Neural Networks for Parameter Estimation in Quantitative MRI
New York University Grossman School of Medicine（纽约大学格罗斯曼医学院）；Bernard and Irene Schwartz Center for Biomedical Imaging（伯纳德和艾琳·施瓦茨生物医学成像中心）；Vilcek Institute of Graduate Biomedical Sciences（维尔cek研究生生物医学科学研究院）
Purpose: To develop neural network (NN)-based quantitative MRI parameter estimators with minimal bias and a variance close to the Cramér-Rao bound.
Theory and Methods: We generalize the mean squared error loss to control the bias and variance of the NN's estimates, which involves averaging over multiple noise realizations of the same measurements during training. Bias and variance properties of the resulting NNs are studied for two neuroimaging applications.
Results: In simulations, the proposed strategy reduces the estimates' bias throughout parameter space and achieves a variance close to the Cramér-Rao bound. In vivo, we observe good concordance between parameter maps estimated with the proposed NNs and traditional estimators, such as non-linear least-squares fitting, while state-of-the-art NNs show larger deviations.
Conclusion: The proposed NNs have greatly reduced bias compared to those trained using the mean squared error and offer significantly improved computational efficiency over traditional estimators with comparable or better accuracy.
Quantitative mapping from conventional MRI using self-supervised physics-guided deep learning: applications to a large-scale, clinically heterogeneous dataset
胎儿肺弥散加权磁共振成像（DWI）数据的定量分析显示，在提供定量成像生物标志物，间接反映胎儿肺成熟的潜力。然而，采集期间的胎儿运动妨碍了对采集的DWI数据的定量分析，因此妨碍了可靠的临床应用。我们介绍了qDWI-morph，这是一种用于运动补偿定量DWI（qDWI）分析的无监督深度神经网络架构。我们的方法将配准子网络与定量DWI模型拟合子网络耦合。我们同时估计的qDWI参数和运动模型，通过最小化的生物物理通知损失函数集成的注册损失和模型拟合质量损失。我们证明了qDWI-morph的附加价值：1）没有运动补偿的基线qDWI分析和2）仅包含配准丢失的基线深度学习模型。通过胎儿肺DWI数据的体内qDWI分析，qDWI-形态实现了与胎龄的显著改善的相关性（R平方=0.32 vs. 0.13，0.28）。我们的qDWI-形态有可能实现DWI数据的运动补偿定量分析，并为非侵入性胎儿肺成熟度评估提供临床可行的生物标志物。我们的代码可在https://github.com/TechnionComputationalMRILab/qDWI-Morph上获取。
Spatio-Temporal Mixture-of-Modality-Experts Diffusion for Quantitative DCE-MRI Synthesis from Incomplete MR Sequences
Interdisciplinary Program in Cancer Biology, Seoul National University College of Medicine（首尔大学医学院癌症生物学跨学科项目）；Department of Radiology, Seoul National University Hospital（首尔大学医院放射科）；Department of Radiology, Seoul National University College of Medicine（首尔大学医学院放射科）；Healthcare AI Research Institute, Seoul National University Hospital（首尔大学医院医疗人工智能研究所）
Hybrid Multi-Dimensional MRI Prostate Cancer Detection via Hadamard Network-Based Bias Correction and Residual Networks
Electrical and Computer Engineering Department, University of Illinois Chicago（伊利诺伊大学芝加哥分校电子与计算机工程系）；Machine and Hybrid Intelligence Lab, Northwestern University（西北大学机器与混合智能实验室）；Department of Radiology, University of Chicago（芝加哥大学放射学系）
Improving accuracy and uncertainty quantification of deep learning based quantitative MRI using Monte Carlo dropout
Technical University of Munich（慕尼黑工业大学）；University of Oxford（牛津大学）；Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）；Tianjin University（天津大学）
针对深度学习定量 MRI 的精度与不确定性问题，提出训练和推理均使用 Monte Carlo dropout 并平均多次预测的方法，在小数据集下显著提升 FA、MD 图精度，还能生成辅助诊断的置信图。
Coordinate-Based Neural Representation Enabling Zero-Shot Learning for 3D Multiparametric Quantitative MRI
School of Biomedical Engineering, Shanghai Jiao Tong University（上海交通大学生物医学工程学院）；School of Biomedical Engineering, ShanghaiTech University（上海科技大学生物医学工程学院）；Department of Neurosurgery, Shanghai Jiao Tong University School of Medicine（上海交通大学医学院神经外科）；Department of Electrical Engineering and Computer Sciences, University of California, Berkeley（加州大学伯克利分校电气工程与计算机科学系）；School of Information Science and Technology, ShanghaiTech University（上海科技大学信息科学与技术学院）
Female-RHINO: A Real-Time Scanner-Integrated Framework for Automated Quantitative Uterine MRI Analysis and Structured Reporting
Radiologisches Institut, Universitätsklinikum Erlangen（埃朗根大学医院放射学研究所）；School of Biomedical Engineering and Imaging, KCL（KCL生物医学工程与成像学院）；Institute for Information Processing, Leibniz University Hannover（汉诺威莱布尼茨大学信息处理研究所）；CAIMed, L3S research institute（CAIMed，L3S研究机构）
MRI Parameter Mapping via Gaussian Mixture VAE: Breaking the Assumption of Independent Pixels
UCL（伦敦大学学院）；UCL Hawkes Institute（伦敦大学学院霍克斯研究所）；UCL Satsuma Lab（伦敦大学学院萨摩亚实验室）；UCL Department of Medical Physics and Biomedical Engineering（伦敦大学学院医学物理与生物医学工程系）；UCL Medical School（伦敦大学学院医学院）；UCL Respiratory（伦敦大学学院呼吸科）；UCL Department of Computer Science（伦敦大学学院计算机科学系）；Cardiff University（卡迪夫大学）；Cardiff University Brain Research Imaging Center（卡迪夫大学脑研究成像中心）
Non-Invasive Detection of PROState Cancer with Novel Time-Dependent Diffusion MRI and AI-Enhanced Quantitative Radiological Interpretation: PROS-TD-AI
University of Chile（智利大学）；Clínica Dávila（达维拉诊所）；Zhejiang University（浙江大学）；Hospital Clinic, Universitat de Barcelona（巴塞罗那大学医院诊所）；San Sebastián University（圣塞巴斯蒂安大学）
Unsupervised learning of MRI tissue properties using MRI physics models
Harvard Medical School（哈佛医学院）；Massachusetts General Hospital（麻省总医院）；California Institute of Technology（加州理工学院）；Massachusetts Institute of Technology（麻省理工学院）
Pre- to Post-Contrast Breast MRI Synthesis for Enhanced Tumour Segmentation
Universitat de Barcelona（巴塞罗那大学）；Helmholtz Center Munich（慕尼黑亥姆霍兹中心）；Technical University of Munich（慕尼黑工业大学）；Karolinska Institutet（卡罗林斯卡学院）；King’s College London（伦敦国王学院）
Comparative Study of Probabilistic Atlas and Deep Learning Approaches for Automatic Brain Tissue Segmentation from MRI Using N4 Bias Field Correction and Anisotropic Diffusion Pre-processing Techniques
University of Girona（赫罗纳大学）；University of Burgundy（勃艮第大学）；University of Cassino and Southern Lazio（卡西诺和南拉齐奥大学）
School of Biomedical Engineering, Shanghai Jiao Tong University（上海交通大学生物医学工程学院）；School of Information Science and Technology, ShanghaiTech University（上海交通大学信息科学与技术学院）；School of Biomedical Engineering and the National Engineering Research Center of Advanced Magnetic Resonance Technologies for Diagnosis and Therapy (NERCAMRT), Shanghai Jiao Tong University（上海交通大学生物医学工程学院和先进磁共振诊断与治疗国家工程研究中心）
Artificial Intelligence-assisted Pixel-level Lung (APL) Scoring for Fast and Accurate Quantification in Ultra-short Echo-time MRI
Australian E-health Research Centre, CSIRO（澳大利亚电子健康研究中心，联邦科学与工业研究组织）；The University of Queensland（昆士兰大学）；Queensland Children's Hospital（昆士兰儿童医院）；Siemens Healthcare Pty Ltd（西门子医疗有限公司）；Siemens Healthineers AG（西门子医疗股份公司）
Bias in Unsupervised Anomaly Detection in Brain MRI
Technical University of Munich（慕尼黑工业大学）；Helmholtz AI（亥姆霍兹人工智能研究所）；Helmholtz Center Munich（亥姆霍兹慕尼黑中心）；King’s College London（伦敦国王学院）；Klinikum Rechts der Isar（伊萨尔河畔临床教学中心）；Imperial College London（帝国理工学院）
Joint PET-MRI Reconstruction with Diffusion Stochastic Differential Model
Inner Mongolia University（内蒙古大学）；Inner Mongolia Medical University（内蒙古医科大学）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；Inner Mongolia Medical University Affiliated Hospital（内蒙古医科大学附属医院）
针对 PET-MRI 系统中 MRI 采集耗时长且 PET 信噪比低的问题，本文提出一种基于扩散随机微分方程的联合重建模型，通过学习两者的联合概率分布加速 MRI 并提升 PET 质量，在定性和定量上超越现有最先进方法。
