Patient outcome prediction is critical in management of ischemic stroke. In this paper, a novel machine learning model is proposed for stroke outcome prediction using multimodal Magnetic Resonance Imaging (MRI). The proposed model consists of two serial levels of Autoencoders (AEs), where different AEs at level 1 are used for learning unimodal features from different MRI modalities and a AE at level 2 is used to combine the unimodal features into compressed multimodal features. The sequences of multimodal features of a given patient are then used by an LSTM network for predicting outcome score. The proposed AE2-LSTM model is proved to be an effective approach for better addressing the multimodality and volumetric nature of MRI data. Experimental results show that the proposed AE2-LSTM outperforms the existing state-of-the art models by achieving highest AUC=0.71 and lowest MAE=0.34.
Stroke Locus Net: Occluded Vessel Localization from MRI Modalities
Department of Electrical Engineering, College of Engineering, Qatar University, Doha, Qatar(电气工程系,工程学院,卡塔尔大学,多哈,卡塔尔);Department of Mathematics and Statistics, College of Arts and Sciences, Qatar University, Doha, Qatar(数学与统计学系,艺术与科学学院,卡塔尔大学,多哈,卡塔尔)
本研究提出端到端深度学习流水线 Stroke Locus Net,结合 nnUNet 分割、动脉图谱与 pGAN 生成,仅用 MRI 扫描实现闭塞血管定位,为更快速的中风诊断提供潜力。
Fusion of Diffusion Weighted MRI and Clinical Data for Predicting Functional Outcome after Acute Ischemic Stroke with Deep Contrastive Learning
Queens College of the City University of NY(纽约市立大学皇后学院);National Chung Cheng University(台湾中正大学);Ditmanson Medical Foundation Chia-Yi Christian Hospital(戴德森医疗财团法人嘉义基督教医院)
Comparing Baseline and Day-1 Diffusion MRI Using Multimodal Deep Embeddings for Stroke Outcome Prediction
1 Department of Computer Science, University of Geneva, Switzerland 2 Department of Electrical \& Computer Engineering, University of Florida, FL, USA 3 Service of Medical Informatics, University Hospital of Geneva, Switzerland 4 Department of Imaging;Medical Informatics, University of Geneva, Switzerland
Computational Image-based Stroke Assessment for Evaluation of Cerebroprotectants with Longitudinal and Multi-site Preclinical MRI
USC Stevens Neuroimaging and Informatics Institute(南加州大学史蒂文斯神经影像与信息学研究所);Zilkha Neurogenetic Institute, Keck School of Medicine of USC(南加州大学凯克医学院齐尔卡神经遗传学研究所);Martinos Center for Biomedical Imaging, Massachusetts General Hospital(麻省总医院马蒂诺斯生物医学成像中心);Yale University(耶鲁大学);University of Iowa(爱荷华大学);Augusta University(奥古斯塔大学);University of Texas(德克萨斯大学);Kennedy Krieger Institute, Johns Hopkins University(约翰斯·霍普金斯大学肯尼迪·克里格研究所)
Local Gamma Augmentation for Ischemic Stroke Lesion Segmentation on MRI
University of Copenhagen(哥本哈根大学);Cerebriu A/S(Cerebriu公司);Rigshospitalet, Copenhagen, Denmark(哥本哈根大学医院);Pioneer Centre for Artificial Intelligence, University of Copenhagen(哥本哈根大学人工智能先锋中心)
Dual-Encoder Transformer-Based Multimodal Learning for Ischemic Stroke Lesion Segmentation Using Diffusion MRI
Department of Anesthesiology, Perioperative and Pain Medicine, Stanford University(麻醉学、围术期与疼痛医学系,斯坦福大学);Department of Biomedical Sciences, Seoul National University(生物医学科学系,首尔国立大学);Department of Computer Engineering, National University of Sciences and Technology (NUST)(计算机工程系,国立科学与技术大学(NUST));Department of Computer Science, The University of Alabama(计算机科学系,阿拉巴马大学);Department of Computer Science, Hanyang University(计算机科学系,翰阳大学)
A multi-path 2.5 dimensional convolutional neural network system for segmenting stroke lesions in brain MRI images
New Jersey Institute of Technology(新泽西理工学院);Kessler Foundation(凯斯勒基金会);Rutgers – New Jersey Medical School(罗格斯大学新泽西医学院);Medical College of Wisconsin(威斯康星医学院);Rutgers University – Newark(罗格斯大学纽瓦克分校)
ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset
Technical University of Munich(慕尼黑工业大学);Klinikum rechts der Isar(右岸医院);School of Medicine, Technical University of Munich(慕尼黑工业大学医学院);icometrix(icometrix公司);University Medical Center Hamburg-Eppendorf(汉堡埃彭多夫大学医学中心);University of Bern(伯尔尼大学);ARTORG Center for Biomedical Engineering Research, Univ. of Bern(伯尔尼大学生物医学工程研究ARTORG中心);University of Southern California(南加州大学);Chan Division of Occupational Science and Occupational Therapy, University of Southern California(南加州大学陈职业科学与职业治疗系);TranslaTUM – Central Institute for Translational Cancer Research, Technical University of Munich(慕尼黑工业大学TranslaTUM转化癌症研究中心)
Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
Integrating uncertainty in deep neural networks for MRI based stroke analysis
University of Zurich (UZH)(苏黎世大学);Zurich University of Applied Sciences (ZHAW)(苏黎世应用科技大学);Konstanz University of Applied Sciences(康斯坦茨应用科技大学);University Hospital Zurich (UHZ)(苏黎世大学医院)
Unsupervised Non Linear Dimensionality Reduction Machine Learning methods applied to Multiparametric MRI in cerebral ischemia: Preliminary Results
The Johns Hopkins University School of Medicine(约翰斯·霍普金斯大学医学院);Sidney Kimmel Comprehensive Cancer Center(西德尼·金梅尔综合癌症中心);The Johns Hopkins University(约翰斯·霍普金斯大学)
Automatic detection of acute ischemic stroke using non-contrast computed tomography and two-stage deep learning model
Kobe University Hospital(神户大学医院);Research Center for Advanced Science and Technology, The University of Tokyo(东京大学先进科学技术研究中心);Ichinomiyanishi Nishi Hospital(一宮西医院);Osaka Red Cross Hospital(大阪红十字医院)
背景和目的:我们的目标是开发和评估一种涉及两阶段深度学习模型的自动急性缺血性卒中相关(AIS)检测系统。方法:我们纳入了来自两个不同机构的238例病例。通过参考头部磁共振成像(MRI)图像,在CT扫描后24小时内进行MRI检查,在238组头部CT图像中的每一组上注释AIS相关结果。这238个标注的案例被分成包括189个案例的训练集和包括49个案例的测试集。随后,使用You Only Look Once v3模型和Visual Geometry Group 16分类模型从训练集构建了两阶段深度学习检测模型。然后,两阶段模型在测试集中执行AIS检测过程。为了评估检测模型的结果,委员会认证的放射科医生还在有和没有检测模型的帮助下评估了测试集的头部CT图像。计算AIS检测的灵敏度和假阳性数,以评价测试集检测结果。使用McNemar检验比较了放射科医生在有和没有软件检测结果的情况下的灵敏度。p值小于0.05被认为具有统计学显著性。结果如下:对于两阶段模型和放射科医生,没有和使用软件结果,灵敏度分别为37.3%,33.3%和41.3%,每例病例的假阳性数分别为1.265,0.327和0.388。在使用两阶段检测模型的结果时,委员会认证的放射科医生的检测灵敏度显著提高(p值= 0.0313)。结论:我们的检测系统涉及两阶段深度学习模型,显着提高了放射科医生在AIS检测中的灵敏度。
Synthetic Perfusion Maps: Imaging Perfusion Deficits in DSC-MRI with Deep Learning
University of Bern(伯尔尼大学);Department of Neurology, Inselspital, University of Bern(伯尔尼大学 Inselspital 神经科);Department of Neurology, University of California Los Angeles (UCLA)(加州大学洛杉矶分校(UCLA)神经科);Support Center for Advanced Neuroimaging, Institute for Diagnostic and Interventional Neuroradiology, Inselspital, University of Bern(伯尔尼大学 Inselspital 诊断与介入神经放射学研究所高级神经影像支持中心)
A fully 3D multi-path convolutional neural network with feature fusion and feature weighting for automatic lesion identification in brain MRI images
New Jersey Institute of Technology(新泽西理工学院);Kessler Foundation(凯斯勒基金会);Emory University(埃默里大学);Atlanta VA Medical Center(亚特兰大退伍军人事务医疗中心);Medical College of Wisconsin(威斯康星医学院);Rutgers University - Newark(罗格斯大学纽瓦克分校)
我们提出了一个全三维多路径卷积网络来预测中风病变的三维脑MRI图像。我们的多路径模型具有针对包含残余卷积块的不同模态的独立编码器、来自不同模态的加权多路径特征融合以及用于组合编码器和解码器特征的加权融合模块。与DeepMedic、3D U-Net和AnatomyNet等现有的3D CNN相比,我们的网络在220名患者的大型ATLAS基准测试中实现了最高的统计学显著交叉验证准确率(60.5%)。我们还在来自Kessler Foundation和Medical College Wisconsin的多模态图像上测试了我们的模型,并实现了65%的统计学显著交叉验证准确率,显著优于多模态3D U-Net和DeepMedic。总的来说,我们的模型提供了一个原则性的,可扩展的多路径方法,优于多通道的替代品,并在现有的基准实现高骰子精度。
A Long Short-term Memory Based Recurrent Neural Network for Interventional MRI Reconstruction
Shanghai Jiao Tong University(上海交通大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine(上海交通大学医学院附属瑞金医院);KTH Royal Institute of Technology(瑞典皇家理工学院);Mayo Clinic in Arizona(亚利桑那州妙佑医疗国际);ShanghaiTech University(上海科技大学)
用于手术引导的介入性磁共振成像(i-MRI)可以帮助可视化介入过程,如脑深部电刺激(DBS),改善手术性能和患者结局。与传统动态成像中的回顾性重建不同,i-MRI for DBS必须在线顺序采集和重建介入图像。在这里,我们提出了一种基于卷积长短期记忆(Conv-LSTM)的循环神经网络(RNN)或ConvLR,以重建具有黄金角径向采样的介入图像。通过使用初始化器和Conv-LSTM块,利用术前参考图像和术中帧的先验信息重建当前帧。径向采样的数据一致性通过软投影方法实现。为了提高重建精度,采用了对抗性学习策略。模拟了一组基于术前和术后MR图像的介入图像,以进行算法确认。结果显示,与最先进的方法相比,只有10个径向辐条,ConvLR提供了最佳性能,加速高达40倍。所提出的算法有可能实现DBS的实时i-MRI,并可用于通用MR引导介入。
Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning
Department of Imaging Neuroscience, University College London, UK(影像神经科学系,伦敦大学学院,英国);Institute of Cognitive Neuroscience, University College London, UK(认知神经科学研究所,伦敦大学学院,英国)
BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning
Department of Radiation Oncology and Winship Cancer Institute, Emory University(放射肿瘤科和Winship癌症研究所,埃默里大学);Department of Radiation and Cellular Oncology, The University of Chicago(放射肿瘤学与细胞肿瘤学部,芝加哥大学);Department of Electrical and Computer Engineering, Georgia Institute of Technology(电气与计算机工程系,佐治亚理工学院);Department of Biomedical Engineering, Georgia Institute of Technology(生物医学工程系,佐治亚理工学院);Department of Biomedical Informatics, Emory University(生物医学信息学系,埃默里大学);Department of Medical Physics, Memorial Sloan Kettering Cancer Center(医学物理系,纪念斯隆凯特琳癌症中心)
Universidad Técnica Particular de Loja(洛哈技术大学);University of Waterloo(滑铁卢大学);Universitat Politècnica de València(瓦伦西亚理工大学);Consejo Superior de Investigaciones Científicas (CSIC)(西班牙国家研究委员会)
State-of-the-Art Stroke Lesion Segmentation at 1/1000th of Parameters
Center for Data Science, Nell Hodgson Woodruff School of Nursing(内尔·霍奇森·伍德拉夫护理学院数据科学中心);Emory University(埃默里大学);University of South Carolina(南卡罗来纳大学);Georgia State University(佐治亚州立大学)
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)
Multi-task learning for the automatic grading of enlarged perivascular space burden using MRI
University of Edinburgh(爱丁堡大学);Canon Medical Research Europe(佳能医疗研究欧洲);UK Dementia Research Institute(英国痴呆症研究所);Chongqing General Hospital(重庆市人民医院);Chongqing University(重庆大学);University of California, San Francisco(加利福尼亚大学旧金山分校);University of Aberdeen(阿伯丁大学)
Improving Factuality of 3D Brain MRI Report Generation with Paired Image-domain Retrieval and Text-domain Augmentation
Cancer Biology, Seoul National University College of Medicine, Korea(首尔国立大学医学院癌症生物学系,韩国);Radiology, Massachusetts General Hospital(麻省总医院放射科);Harvard Medical School(哈佛医学院);Biomedical Systems Informatics, Yonsei University College of Medicine, Korea(延世大学医学院生物医学系统信息学系,韩国);Graduate School, Yonsei University, Korea(延世大学研究生院,韩国);Radiology, Seoul National University College of Medicine, Korea(首尔国立大学医学院放射科,韩国);Radiology, Seoul National University Hospital, Korea(首尔国立大学医院放射科,韩国);Radiology, Seoul National University Bundang Hospital, Korea(首尔国立大学 Bundang 医院放射科,韩国);Radiology, SMG-SNU Boramae Medical Center, Korea(SMG-SNU Boramae 医疗中心放射科,韩国);Psychiatry, Yonsei University College of Medicine, Korea(延世大学医学院精神病学系,韩国);Behavioral Sciences in Medicine, Yonsei University College of Medicine, Korea(延世大学医学院医学行为科学系,韩国);Yonsei Institute for Digital Health, Korea(延世大学数字健康研究院,韩国);Kim Jaechul Graduate School of AI, KAIST, Korea(金 Jaechul人工智能研究生院,韩国)
Full-Head Segmentation of MRI with Abnormal Brain Anatomy: Model and Data Release
The City College of New York(纽约市立学院);New York University(纽约大学);Georgetown University(乔治城大学);University of North Carolina School of Medicine(北卡罗来纳大学医学院);Soterix Medical
Modality-Agnostic Input Channels Enable Segmentation of Brain lesions in Multimodal MRI with Sequences Unavailable During Training
Department of Engineering Science, University of Oxford, Oxford, UK(工程科学系,牛津大学,牛津,英国);Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK(牛津大学临床神经科学系,牛津,英国)