Adapted Foundation Models for Breast MRI Triaging in Contrast-Enhanced and Non-Contrast Enhanced Protocols
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU)（弗里德里希-亚历山大大学埃尔兰根-纽伦堡分校）；Uniklinikum Erlangen（埃尔兰根大学医院）；Radiology München（慕尼黑放射科）；Polish Academy of Science（波兰科学院）
Tri-Thien Nguyen, Lorenz A. Kapsner, Tobias Hepp, Shirin Heidarikahkesh, Hannes Schreiter, Luise Brock, Dominika Skwierawska, Dominique Hadler, Julian Hossbach,… 展开作者
Tri-Thien Nguyen, Lorenz A. Kapsner, Tobias Hepp, Shirin Heidarikahkesh, Hannes Schreiter, Luise Brock, Dominika Skwierawska, Dominique Hadler, Julian Hossbach, Evelyn Wenkel, Sabine Ohlmeyer, Frederik B. Laun, Andrzej Liebert, Andreas Maier, Michael Uder, Sebastian Bickelhaupt
备注：
23 pages, 6 figures, 4 tables. Originally submitted to Radiology (RAD-25-2541); under consideration for transfer to Radiology: Artificial Intelligence (RSNA Portfolio Journal)
Background: Magnetic resonance imaging (MRI) has high sensitivity for breast cancer detection, but interpretation is time-consuming. Artificial intelligence may aid in pre-screening. Purpose: To evaluate the DINOv2-based Medical Slice Transformer (MST) for ruling out significant findings (Breast Imaging Reporting and Data System [BI-RADS] >=4) in contrast-enhanced and non-contrast-enhanced abbreviated breast MRI. Materials and Methods: This institutional review board approved retrospective study included 1,847 single-breast MRI examinations (377 BI-RADS >=4) from an in-house dataset and 924 from an external validation dataset (Duke). Four abbreviated protocols were tested: T1-weighted early subtraction (T1sub), diffusion-weighted imaging with b=1500 s/mm2 (DWI1500), DWI1500+T2-weighted (T2w), and T1sub+T2w. Performance was assessed at 90%, 95%, and 97.5% sensitivity using five-fold cross-validation and area under the receiver operating characteristic curve (AUC) analysis. AUC differences were compared with the DeLong test. False negatives were characterized, and attention maps of true positives were rated in the external dataset. Results: A total of 1,448 female patients (mean age, 49 +/- 12 years) were included. T1sub+T2w achieved an AUC of 0.77 +/- 0.04; DWI1500+T2w, 0.74 +/- 0.04 (p=0.15). At 97.5% sensitivity, T1sub+T2w had the highest specificity (19% +/- 7%), followed by DWI1500+T2w (17% +/- 11%). Missed lesions had a mean diameter <10 mm at 95% and 97.5% thresholds for both T1sub and DWI1500, predominantly non-mass enhancements. External validation yielded an AUC of 0.77, with 88% of attention maps rated good or moderate. Conclusion: At 97.5% sensitivity, the MST framework correctly triaged cases without BI-RADS >=4, achieving 19% specificity for contrast-enhanced and 17% for non-contrast-enhanced MRI. Further research is warranted before clinical implementation.
Automated triaging of head MRI examinations using convolutional neural networks
King’s College London（伦敦国王学院）；King’s College Hospital（国王学院医院）；Wrightington, Wigan and Leigh NHSFT（赖廷顿、维冈与利国家医疗服务基金会信托）；University College London（伦敦大学学院）
Evaluating ADC-only deep learning pipelines for breast cancer detection and segmentation using standalone diffusion-weighted MRI
University of Oviedo（奥维耶多大学）；University of A Coruña（拉科鲁尼亚大学）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）；Purdue University（普渡大学）；Asturias Central University Hospital (HUCA)（阿斯图里亚斯中央大学医院）
A Large Model for Non-invasive and Personalized Management of Breast Cancer from Multiparametric MRI
The Hong Kong University of Science and Technology（香港科技大学）；PLA Middle Military Command General Hospital（中国人民解放军中部战区总医院）；The University of Hong Kong（香港大学）；The Chinese University of Hong Kong（香港中文大学）；Yunnan Cancer Hospital（云南省肿瘤医院）；th Medical Center of Chinese PLA General Hospital（中国人民解放军总医院第五医学中心）；Southern Medical University（南方医科大学）
BreastSegNet: Multi-label Segmentation of Breast MRI
Department of Biostatistics \& Bioinformatics, Duke University Department of Electrical；Computer Engineering, Duke University Department of Radiology, Duke University Medical Center Department of Computer Science, Duke University
PD-DWI: Predicting response to neoadjuvant chemotherapy in invasive breast cancer with Physiologically-Decomposed Diffusion-Weighted MRI machine-learning model
Reichman University（赖希曼大学）；Technion（以色列理工学院）
乳腺癌新辅助化疗（NAC）后病理完全缓解（pCR）的早期预测在手术计划和优化治疗策略中起着至关重要的作用。最近，提出了基于机器和深度学习的方法，用于从多参数MRI（mp-MRI）数据（包括动态对比增强MRI和弥散加权MRI（DWI））中进行早期pCR预测，并取得了中等成功。我们引入了PD-DWI，这是一种生理分解的DWI机器学习模型，用于从DWI和临床数据预测pCR。我们的模型首先将原始DWI数据分解为影响DWI信号的各种生理线索，然后使用分解的数据以及临床变量作为基于放射学的XGBoost模型的输入特征。我们证明了我们的PD-DWI模型相对于传统机器学习方法的附加价值，用于使用公开可用的乳腺多参数MRI预测NAC反应（BMMR 2）挑战，从mp-MRI数据预测pCR。与挑战测试集排行榜上的当前最佳结果（0.8849 vs. 0.8397）相比，我们的模型大大改善了曲线下面积（AUC）。PD-DWI有可能改善乳腺癌NAC后pCR的预测，减少总体mp-MRI采集时间，并消除造影剂注射的需要。
SAFE-Diff: Scale-Aware Attention and Feature-Dispersive Diffusion with Uncertainty Estimation for Contrast-Enhanced Breast MRI Synthesis
Department of Medical Imaging, Radboud University Medical Center（鲁文大学医学中心医学影像部）；Department of Radiology, Netherlands Cancer Institute（荷兰癌症研究所放射科）；Maastro Clinic（马斯垂克诊所）；Faculty of Applied Science, Macao Polytechnic University（澳门理工大学应用科学学院）；Department of Radiation Oncology, Netherlands Cancer Institute（荷兰癌症研究所放射肿瘤科）
Automatic lesion detection, segmentation and characterization via 3D multiscale morphological sifting in breast MRI
School of Information Technology and Electrical Engineering, University of Queensland（昆士兰大学信息技术与电气工程学院）；Science and Engineering Faculty, Queensland University of Technology（昆士兰科技大学科学与工程学院）
Towards Affordable Tumor Segmentation and Visualization for 3D Breast MRI Using SAM2
Center for Biosystems and Biotech Data Science, Ghent University Global Campus, Incheon, Republic of Korea（生物系统与生物技术数据科学中心，格罗宁根大学全球校园，韩国Incheon）；Department of Mathematics, Computer Science and Statistics, Ghent University, Ghent, Belgium（数学、计算机科学与统计学系，格罗宁根大学，比利时Ghent）；IDLab, Department of Electronics and Information Systems, Ghent University, Ghent, Belgium（IDLab，电子与信息系统系，格罗宁根大学，比利时Ghent）
LesiOnTime -- Joint Temporal and Clinical Modeling for Small Breast Lesion Segmentation in Longitudinal DCE-MRI
Computational Imaging Research Lab, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Austria（医学影像研究实验室，生物医学影像与影像引导治疗系，维也纳医科大学）；Comprehensive Center for AI in Medicine, Medical University of Vienna, Austria（医学人工智能综合中心，维也纳医科大学）；Breast Health Center, Privatklinik Döbling, Austria（乳腺健康中心，多克林根私人诊所）；Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Austria（生物医学影像与影像引导治疗系，维也纳医科大学）；Department of Obstetrics and Gynecology, Medical University of Vienna, Austria（妇产科系，维也纳医科大学）；Department of Pathology, Medical University of Vienna, Austria（病理学系，维也纳医科大学）；Comprehensive Cancer Center, Medical University of Vienna, Austria（综合癌症中心，维也纳医科大学）
MeisenMeister: A Simple Two Stage Pipeline for Breast Cancer Classification on MRI
German Cancer Research Center (DKFZ) Heidelberg, Division of Medical Image Computing, Germany（德国癌症研究中心（DKFZ）海德堡，医学影像计算部，德国）；Medical Faculty, Heidelberg University, Germany（海德堡大学医学学院，德国）；Faculty of Mathematics and Computer Science, Heidelberg University（海德堡大学数学与计算机科学学院）；Learning Group, Department of Radiation Oncology, Heidelberg University Hospital, Heidelberg, Germany（放射肿瘤学系学习组，海德堡大学医院，海德堡，德国）
Interpretable deep learning regression for breast density estimation on MRI
Image Sciences Institute, University Medical Center Utrecht, Utrecht University（乌得勒支大学乌得勒支大学医学中心影像科学研究所）；Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital（荷兰癌症研究所——安东尼·范·列文虎克医院）
LoGo-MR: Screening Breast MRI for Cancer Risk Prediction by Efficient Omni-Slice Modeling
Netherlands Cancer Institute（荷兰癌症研究所）；Radboud University Medical Center（拉德堡德大学医学中心）；University Medical Center Utrecht（乌得勒支大学医学中心）；Macao Polytechnic University（澳门理工大学）
Model Agnostic Saliency for Weakly Supervised Lesion Detection from Breast DCE-MRI
The University of Adelaide（阿德莱德大学）；Australian Institute for Machine Learning（澳大利亚机器学习研究所）；Delft University of Technology（代尔夫特理工大学）；Queensland University of Technology（昆士兰科技大学）
PRISM-Net: Patient-specific reference-guided inter-breast symmetry matching for three-class breast DCE-MRI classification
Nankai University（南开大学）；Zhongguancun Academy（中关村学院）；The First Medical Center of Chinese PLA General Hospital（中国人民解放军总医院第一医学中心）；Beijing University of Posts and Telecommunications（北京邮电大学）；The Six Medical Center of Chinese PLA General Hospital（中国人民解放军总医院第六医学中心）；Tsinghua University（清华大学）
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）
Deep Angular Embedding and Feature Correlation Attention for Breast MRI Cancer Analysis
The Chinese University of Hong Kong（香港中文大学）；Imsight Medical Technology, Co., Ltd.（推想医疗科技股份有限公司）；Imperial College London（伦敦帝国理工学院）；The Fifth Medical Center of Chinese PLA General Hospital（中国人民解放军总医院第五医学中心）；Beijing Image Diagnostic Center of Rimag（北京一脉阳光影像诊断中心）
Mask-Guided Multi-Channel SwinUNETR Framework for Robust MRI Classification
Departament de Matemàtiques i Informàtica, Universitat de Barcelona, Spain（巴塞罗那大学数学与计算机系）；Computer Vision Center, Bellaterra, Spain（贝尔特拉拉计算机视觉中心）；Institució Catalana de Recerca i Estudis Avançats (ICREA), Passeig Lluís Companys 23, Barcelona, Spain（加泰罗尼亚高级研究与教育研究所）
Diffeomorphic Multi-Resolution Deep Learning Registration for Applications in Breast MRI
Auckland Bioengineering Institute（奥克兰生物工程研究所）；University of Auckland（奥克兰大学）；Faculty of Medical and Health Sciences（医学与健康科学学院）；MGH/HST Martinos Center for Biomedical Imaging（麻省总医院/哈佛健康科学技术生物医学成像马蒂诺斯中心）；Harvard Medical School（哈佛医学院）；MIT（麻省理工学院）
Predictive Enhancement Calibration for Latent Breast MRI Virtual Contrast Enhancement
The First Affiliated Hospital of Chongqing Medical University（重庆医科大学附属第一医院）；Chongqing Municipal Health Commission（重庆市卫生健康委员会）；Chongqing Translational Medicine Center（重庆市转化医学中心）
Response monitoring of breast cancer on DCE-MRI using convolutional neural network-generated seed points and constrained volume growing
Image Sciences Institute, University Medical Center Utrecht（乌得勒支大学医学中心影像科学研究所）；Department of Radiology, Antoni van Leeuwenhoek Hospital - The Netherlands Cancer Institute（荷兰癌症研究所安东尼·范·列文虎克医院放射科）
SpurBreast: A Curated Dataset for Investigating Spurious Correlations in Real-world Breast MRI Classification
Center for Biosystems；Biotech Data Science, Ghent University Global Campus, Incheon, Republic of Korea IDLab, ELIS, Ghent University, Ghent, Belgium Department of Applied Mathematics, Computer Science；Statistics, Ghent University, Ghent, Belgium ( )
Synthesis of Contrast-Enhanced Breast MRI Using Multi-b-Value DWI-based Hierarchical Fusion Network with Attention Mechanism
Netherlands Cancer Institute (NKI)（荷兰癌症研究所）；Maastricht University（马斯特里赫特大学）；Radboud University Medical Center（拉德堡德大学医学中心）；Fondazione Universitaria A. Gemelli, IRCCS Roma（罗马A.杰梅利大学基金会IRCCS）
GuidedMorph: Two-Stage Deformable Registration for Breast MRI
Department of Electrical and Computer Engineering, Duke University（电子工程系，杜克大学）；Department of Biostatistics & Bioinformatics, Duke University（生物统计学与生物信息学系，杜克大学）；Department of Health Outcomes and Biomedical Informatics, College of Medicine, University of Florida（健康结果与生物医学信息学系，医学院，佛罗里达大学）
Robustness of breast lesion segmentation under MRI undersampling improves with k-space-aware deep learning
Institute for AI in Medicine (IKIM), University Hospital Essen（人工智能医学研究所（IKIM），埃森大学医院）；Department of Physics, Technical University Dortmund（物理系，多特蒙德技术大学）；Division of Radiology, German Cancer Research Center (DKFZ)（放射学部，德国癌症研究中心（DKFZ））；Cancer Research Center Cologne Essen (CCCE), University Medicine Essen（科隆埃森癌症研究中心（CCCE），埃森大学医学中心）；RACOON Study Group, Site Essen（RACOON研究组，埃森站点）；German Cancer Consortium (DKTK), Partner Site Essen（德国癌症联合会（DKTK），埃森合作伙伴站点）；Medical Faculty and Faculty of Computer Science, University of Duisburg-Essen（医学系和计算机科学系，杜伊斯堡-埃森大学）
Temporal Neural Cellular Automata: Application to modeling of contrast enhancement in breast MRI
Institute of Machine Learning in Biomedical Imaging, Helmholtz Munich, Germany School of Computation, Information；Technology, Technical University of Munich, Germany Departament de Matematiques i Informatica, Universitat de Barcelona, Spain Institució Catalana de Recerca i Estudis Avan c ats (ICREA), Barcelona, Spain Institute for Diagnostic；Interventional Radiology, School of Medicine \& Health, Klinkum Rechts der Isar, Technical University of Munich, Germany German Cancer Consortium (DKTK), Partner Site Munich, Germany School of Biomedical Engineering \& Imaging Sciences, King's College London, UK
Deep learning-based prediction of response to HER2-targeted neoadjuvant chemotherapy from pre-treatment dynamic breast MRI: A multi-institutional validation study
Case Western Reserve University（凯斯西储大学）；University of Mons（蒙斯大学）；Cleveland Clinic（克利夫兰诊所）；Yale School of Medicine（耶鲁医学院）；Institut Jules Bordet（朱尔斯·博尔代研究所）；University Hospitals Cleveland Medical Center（大学医院克利夫兰医学中心）；Louis Stokes Cleveland VA Medical Center（路易斯·斯托克斯克利夫兰退伍军人事务医疗中心）
预测对新辅助治疗的反应是乳腺癌中一个令人烦恼的挑战。在这项研究中，我们评估了深度学习从治疗前获得的治疗前动态对比增强（DCE）MRI预测对HER 2靶向新辅助化疗（NAC）的反应的能力。在一项回顾性研究中，包括来自5家机构的157名HER 2+乳腺癌患者的DCE-MRI数据，我们开发并验证了一种深度学习方法，用于预测治疗前对HER 2靶向NAC的病理完全缓解（pCR）。在单一机构接受HER 2靶向新辅助化疗的100名患者用于训练（n=85）和调整（n=15）卷积神经网络（CNN）以预测pCR。确定了利用造影前和造影后DCE-MRI采集的多输入CNN，以实现验证集内的最佳响应预测（AUC=0.93）。然后，使用治疗前DCE-MRI数据在两个独立的测试队列中对该模型进行测试。在第二家机构的28例患者测试集中（AUC=0.85，95% CI 0.67-1.0，p= 0.0008）和包括来自另外3家机构的数据的29例患者多中心试验中（AUC=0.77，95% CI 0.58-0.97，p=0.006），它实现了强大的性能。发现基于深度学习的反应预测模型超过了包含预测临床变量（测试队列中AUC < .65）和半定量DCE-MRI药代动力学测量模型（测试队列中AUC < .60）的多变量模型。这项工作在多个研究中心提出的结果表明，通过进一步验证，深度学习可以提供一种有效可靠的工具来指导乳腺癌的…
TopoTxR: A topology-guided deep convolutional network for breast parenchyma learning on DCE-MRIs
State University of New York at Stony Brook（纽约州立大学石溪分校）；Newark Beth Israel Medical Center（纽瓦克贝斯以色列医疗中心）；Columbia University Irving Medical Center（哥伦比亚大学欧文医疗中心）；Emory University School of Medicine（埃默里大学医学院）；Georgia Institute of Technology（佐治亚理工学院）
Prototype Learning Guided Hybrid Network for Breast Tumor Segmentation in DCE-MRI
School of Health Science and Engineering, University of Shanghai for Science and Technology（上海理工大学健康科学与工程学院）；School of Biomedical Engineering, ShanghaiTech University（上海科技大学生物医学工程学院）；School of Computer Science and Engineering, Nanjing University of Science and Technology（南京理工大学计算机科学与工程学院）；Ruijin Hospital, Shanghai Jiaotong University School of Medicine（上海交通大学医学院附属瑞金医院）；Shanghai General Hospital（上海交通大学附属第一人民医院）
To deform or not: treatment-aware longitudinal registration for breast DCE-MRI during neoadjuvant chemotherapy via unsupervised keypoints detection
Radboud University Medical Centre（拉德堡德大学医学中心）；The Netherlands Cancer Institute（荷兰癌症研究所）；Macao Polytechnic University（澳门理工大学）；Maastricht University Medical Centre（马斯特里赫特大学医学中心）
