Ultrasonic Medical Tissue Imaging Using Probabilistic Inversion: Leveraging Variational Inference for Speed Reconstruction and Uncertainty Quantification
Full Waveform Inversion (FWI) is a promising technique for achieving high-resolution imaging in medical ultrasound. However, conventional FWI methods suffer from issues related to computational efficiency, dependence on initial models, and the inability to quantify uncertainty. This study aims to enhance inversion performance and provide a reliable method for uncertainty quantification in medical FWI imaging. This study integrates the Stein Variational Gradient Descent (SVGD) algorithm into the FWI framework by deriving the posterior gradient for probabilistic inversion. To evaluate the proposed method, numerical experiments are conducted on synthetic datasets, including a breast tissue model with realistic anatomical structure. Imaging accuracy and uncertainty quantification are assessed to compare the performance of SVGD-based FWI with conventional FWI and Stochastic Variational Inference (SVI) methods. Markov Chain Monte Carlo (MCMC) is implemented as a benchmark to evaluate the quality of uncertainty estimates. For synthetic data, the SVGD-based FWI framework yields more precise estimates in the region of interest (ROI) and demonstrates faster convergence compared to the conventional FWI. For the anatomically realistic breast tissue simulation, SVGD produces a maximum relative error of 1.10\% and a mean relative error of 0.09\% in the ROI. The estimated uncertainty is spatially consistent, with most values below 0.01 and a mean of approximately 0.003. Compared to SVI, SVGD provides improved structural resolution and stronger agreement with the MCMC benchmark, indicating more reliable uncertainty quantification. The SVGD-based FWI method improves inversion quality, enhances uncertainty quantification. These findings indicate that probabilistic inversion is a promising tool for addressing the limitations of traditional FWI methods in ultrasonic imaging of medical tissues.
Real-time 3D Ultrasonic Needle Tracking with a Photoacoustic Beacon
King’s College London（伦敦国王学院）；University College London（伦敦大学学院）；GBA Institute of Collaborative Innovation（粤港澳大湾区协同创新研究院）；University College London Hospitals NHS Foundation Trust（伦敦大学学院医院NHS基金会信托）；Guys and St Thomas’ NHS Trust（盖伊和圣托马斯NHS信托）
Automatic quantification of abdominal subcutaneous and visceral adipose tissue in children, through MRI study, using total intensity maps and Convolutional Neural Networks
In vivo quantification of 3D displacement in sacral soft tissues under compression: Relevance of 2D US-based measurements for pressure ulcer risk assessment
Ultra-high frequency ultrasound imaging and quantification of microvascular flow in xenograft renal cell carcinoma in an avian chorioallantoic membrane model
University of Toronto（多伦多大学）；Sunnybrook Research Institute（桑尼布鲁克研究所）
First Standard Quantification of Ultrasound Attenuation in Healthy Periodontal Soft Tissues In Vivo
University of Michigan（密歇根大学）；University of Illinois Chicago（伊利诺伊大学芝加哥分校）；University of Iowa College of Dentistry & Dental Clinics（爱荷华大学牙学院与牙科诊所）；Sun Nuclear Corporation（太阳核能公司）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；The Ohio State University（俄亥俄州立大学）
Enhanced Instantaneous Elastography in Tissues and Hard Materials Using Bulk Modulus and Density Determined without Externally Applied Material Deformation
Automated quantification of myocardial tissue characteristics from native T1 mapping using neural networks with Bayesian inference for uncertainty-based quality-control
King’s College London（伦敦国王学院）；Guy’s and St Thomas’ NHS Foundation Trust（盖伊和圣托马斯国民保健服务基金会信托）；ETH Zürich（苏黎世联邦理工学院）；Imperial College London（伦敦帝国理工学院）
Automated HER2 scoring with uncertainty quantification using lensfree holography and deep learning
Electrical and Computer Engineering Department, University of California, Los Angeles, CA, 90095, USA（加州大学洛杉矶分校电气与计算机工程系）；Bioengineering Department, University of California, Los Angeles, CA, 90095, USA（加州大学洛杉矶分校生物工程系）；California NanoSystems Institute (CNSI), University of California, Los Angeles, CA, 90095, USA（加州大学洛杉矶分校加州纳米系统研究所）；Department of Computer Science, University of California, Los Angeles, CA, 90095, USA（加州大学洛杉矶分校计算机科学系）
In Vivo Quantification of Glioma-Induced Solid Stress Using MR Elastography and Deformable Image Registration
Charité – Universitätsmedizin Berlin（柏林夏里特医科大学）；Berlin Institute of Health at Charité – Universitätsmedizin Berlin（柏林健康研究所（隶属于柏林夏里特医科大学））；Humboldt University, Berlin（柏林洪堡大学）；Physikalisch-Technische Bundesanstalt, Berlin（德国联邦物理技术研究院）；Technical University Berlin, Einstein Center Digital Future（柏林工业大学爱因斯坦数字中心）
Eigenspectra optoacoustic tomography achieves quantitative blood oxygenation imaging deep in tissues
Institute for Biological and Medical Imaging (IBMI), Helmholtz Zentrum München（亥姆霍兹慕尼黑中心生物与医学成像研究所）；Department of Radiation Oncology, Klinikum rechts der Isar, Technische Universität München（慕尼黑工业大学伊萨尔河畔医院放射肿瘤科）；Chair for Biological Imaging, Technische Universität München（慕尼黑工业大学生物成像教席）；CCG – Innate immunity in Tumor Biology, Helmholtz Zentrum München（亥姆霍兹慕尼黑中心肿瘤生物学先天免疫合作研究组）
Fully Automated and Standardized Segmentation of Adipose Tissue Compartments by Deep Learning in Three-dimensional Whole-body MRI of Epidemiological Cohort Studies
University Hospital Tübingen（蒂宾根大学医院）；University of Stuttgart（斯图加特大学）；Max-Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；Eberhard Karls University（埃伯哈德·卡尔斯大学）；Helmholtz Zentrum München（慕尼黑亥姆霍兹中心）；German Center for Diabetes Research（德国糖尿病研究中心）；University of Freiburg（弗赖堡大学）
目的：通过全身MRI快速、可靠地评估皮下和内脏脂肪组织。研究方法：从全身MR图像中对不同脂肪组织区室进行定量和定位对于检查代谢状况具有高度意义。为了正确识别和代谢疾病风险增加的个体的表型，需要将脂肪组织可靠地自动分割成皮下和内脏脂肪组织。在这项工作中，我们提出了一个三维卷积神经网络（DCNet），以提供一个强大的和客观的分割。在这项回顾性研究中，我们收集了1000例来自Tuebingen家族研究和德国糖尿病研究中心（TUEF/DZD）的66例$\pm$ 13岁; 523例女性），以及来自德国国家队列（NAKO）数据库的300例病例（53例$\pm$ 11岁; 152例女性），进行模型训练、验证和队列间迁移学习测试。这些数据集具有可变的成像序列、成像对比度、接收器线圈布置、扫描仪和成像场强。在灵敏度、特异性、精确度、准确度和Dice重叠方面，将拟议的DCNet与可比的3D UNet分割进行比较。结果如下：可以从3D全身MR数据集（视野覆盖范围450 x 450 x 2000 mm ${}^3$）获得快速（5- 7秒）和可靠的脂肪组织分割，具有高Dice重叠（0.94）、灵敏度（96.6%）、特异性（95.1%）、精确度（92.1%）和准确度（98.4%）。分割掩模和脂肪组织轮廓被自动报告给转诊医生。结论：自动脂肪组织分割在3D全身MR数据集中是可行的，并且可推广到使用所提出的DCNe…
