Multi-institutional efforts can facilitate training of deep MRI reconstruction models, albeit privacy risks arise during cross-site sharing of imaging data. Federated learning (FL) has recently been introduced to address privacy concerns by enabling distributed training without transfer of imaging data. Existing FL methods for MRI reconstruction employ conditional models to map from undersampled to fully-sampled acquisitions via explicit knowledge of the imaging operator. Since conditional models generalize poorly across different acceleration rates or sampling densities, imaging operators must be fixed between training and testing, and they are typically matched across sites. To improve generalization and flexibility in multi-institutional collaborations, here we introduce a novel method for MRI reconstruction based on Federated learning of Generative IMage Priors (FedGIMP). FedGIMP leverages a two-stage approach: cross-site learning of a generative MRI prior, and subject-specific injection of the imaging operator. The global MRI prior is learned via an unconditional adversarial model that synthesizes high-quality MR images based on latent variables. Specificity in the prior is preserved via a mapper subnetwork that produces site-specific latents. During inference, the prior is combined with subject-specific imaging operators to enable reconstruction, and further adapted to individual test samples by minimizing data-consistency loss. Comprehensive experiments on multi-institutional datasets clearly demonstrate enhanced generalization performance of FedGIMP against site-specific and federated methods based on conditional models, as well as traditional reconstruction methods.
Bilkent University（比尔肯特大学）；University of Southern California（南加州大学）
Multi-institutional collaborations are key for learning generalizable MRI synthesis models that translate source- onto target-contrast images. To facilitate collaboration, federated learning (FL) adopts decentralized training and mitigates privacy concerns by avoiding sharing of imaging data. However, FL-trained synthesis models can be impaired by the inherent heterogeneity in the data distribution, with domain shifts evident when common or variable translation tasks are prescribed across sites. Here we introduce the first personalized FL method for MRI Synthesis (pFLSynth) to improve reliabil…
CERN（欧洲核子研究组织）；Masaryk University（马萨里克大学）；National & Kapodistrian University of Athens（雅典国立与卡波迪斯特里亚大学）；Democritus University of Thrace（德谟克利特色雷斯大学）；University Hospital Brno（布尔诺大学医院）
Cross-Modal Vertical Federated Learning for MRI Reconstruction
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Tencent Jarvis Lab（腾讯Jarvis实验室）；The Hong Kong University of Science and Technology（香港科技大学）；Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳））
Federated Magnetic Resonance Imaging (MRI) reconstruction enables multiple hospitals to collaborate distributedly without aggregating local data, thereby protecting patient privacy. However, the data heterogeneity caused by different MRI protocols, insufficient local training data, and limited communication bandwidth inevitably impair global model convergence and updating. In this paper, we propose a new algorithm, FedPR, to learn federated visual prompts in the null space of global prompt for MRI reconstruction. FedPR is a new federated paradigm that adopts a powerful pre-trained model while …
Exploiting Test-Time Augmentation in Federated Learning for Brain Tumor MRI Classification
Institute of Exact and Technological Sciences, Federal University of Viçosa - UFV, Rio Paranaíba-MG, Brazil（精确与技术科学研究所，弗拉维亚联邦大学-UFV，里奥帕拉纳伊巴-MG，巴西）；Department of Computing, Federal University of São Carlos, São Carlos-SP, Brazil（计算系，萨o卡洛斯联邦大学，萨o卡洛斯-SP，巴西）
University of California, San Francisco（加州大学旧金山分校）；Duke University（杜克大学）
Early prostate cancer detection and staging from MRI are extremely challenging tasks for both radiologists and deep learning algorithms, but the potential to learn from large and diverse datasets remains a promising avenue to increase their generalization capability both within- and across clinics. To enable this for prototype-stage algorithms, where the majority of existing research remains, in this paper we introduce a flexible federated learning framework for cross-site training, validation, and evaluation of deep prostate cancer detection algorithms. Our approach utilizes an abstracted rep…
Federated Pseudo Modality Generation for Incomplete Multi-Modal MRI Reconstruction
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Institute of High Performance Computing, A*STAR（新加坡科技研究局高性能计算研究所）；Guangxi Medical University（广西医科大学）；The Hong Kong University of Science and Technology（香港科技大学）
Adaptive Aggregation Weights for Federated Segmentation of Pancreas MRI
Northwestern University（西北大学）；Istanbul University（伊斯坦布尔大学）；University of Catania（卡塔尼亚大学）；Erasmus Medical Center（伊拉斯姆斯医学中心）；University of Wisconsin（威斯康星大学）；Mayo Clinic（梅奥诊所）；New York University（纽约大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；NVIDIA（英伟达公司）
FedSynthCT-Brain: A Federated Learning Framework for Multi-Institutional Brain MRI-to-CT Synthesis
Institute of Biomedical Engineering（生物医学工程研究所）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工大学）；Department of Radiation Oncology（放射肿瘤科）；University Medical Center Schleswig-Holstein（什勒斯威-霍斯特大学医院）；Department of Experimental and Clinical Medicine（实验与临床医学系）；Magna Graecia University（马格拉西亚大学）；Department of Information Engineering（信息工程系）；Università Politecnica delle Marche（帕尔米拉理工大学）
In recent years, convolutional neural networks have demonstrated promising performance in a variety of medical image segmentation tasks. However, when a trained segmentation model is deployed into the real clinical world, the model may not perform optimally. A major challenge is the potential poor-quality segmentations generated due to degraded image quality or domain shift issues. There is a timely need to develop an automated quality control method that can detect poor segmentations and feedback to clinicians. Here we propose a novel deep generative model-based framework for quality control …
Federated Fine-tuning of SAM-Med3D for MRI-based Dementia Classification
Dept. of Radiology \& Nuclear Medicine, Erasmus MC, Rotterdam, the Netherlands The Hong Kong University of Science；Technology, Hong Kong SAR Dept. of Neurology, Erasmus MC, Rotterdam, the Netherlands Dept. of Neurology, UMC Utrecht, Utrecht, the Netherlands Dept. of Geriatrics, Radboud UMC, Nijmegen, the Netherlands Dept. of Neurology, Leiden UMC, Leiden, the Netherlands Dept. of Internal Medicine, UMC Groningen, Groningen, the Netherlands Dept. of Neurology, Amsterdam UMC location VUmc, Amsterdam, the Netherlands Dept. of Psychiatry \& Psychology, Maastricht UMC, Maastricht, the Netherlands
Metadata-Conditioned Generative Models to Synthesize Anatomically-Plausible 3D Brain MRIs
Stanford University（斯坦福大学）；University of Missouri（密苏里大学）；Missouri Institute of Mental Health（密苏里精神卫生研究所）；Weill Cornell Medicine（威尔康奈尔医学院）；SRI International（SRI国际研究所）
MRI-based and metabolomics-based age scores act synergetically for mortality prediction shown by multi-cohort federated learning
Maastricht University Medical Centre +（马斯特里赫特大学医学中心）；Leiden University Medical Center（莱顿大学医学中心）；Delft University of Technology（代尔夫特理工大学）；Erasmus MC - University Medical Center Rotterdam（伊拉斯姆斯大学医学中心鹿特丹分校）；Leiden University（莱顿大学）
Helmholtz Zentrum München（慕尼黑亥姆霍兹中心）；Technical University of Munich（慕尼黑工业大学）；University of Freiburg（弗莱堡大学）；University of Bonn（波恩大学）
Given the high incidence and effective treatment options for liver diseases, they are of great socioeconomic importance. One of the most common methods for analyzing CT and MRI images for diagnosis and follow-up treatment is liver segmentation. Recent advances in deep learning have demonstrated encouraging results for automatic liver segmentation. Despite this, their success depends primarily on the availability of an annotated database, which is often not available because of privacy concerns. Federated Learning has been recently proposed as a solution to alleviate these challenges by trainin…
Energy-Efficient Federated Learning via Adaptive Encoder Freezing for MRI-to-CT Conversion: A Green AI-Guided Research
Institute of Biomedical Engineering（生物医学工程研究所）；Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院）；Department of Political Science（政治学系）；Università Degli Studi Di Teramo（特拉莫大学）；Department of Radiation Oncology（放射肿瘤学系）；University Medical Center Schleswig-Holstein（什未林施瓦茨-霍斯特大学医院）；Department of Experimental and Clinical Medicine（实验与临床医学系）；Magna Graecia University（马格拉西亚大学）
Cyst-X: A Multi-Center MRI Benchmark and Federated Learning Framework for Malignancy-Risk Stratification of Pancreatic Cystic Neoplasm
Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University（机器与混合智能实验室，放射科，西北大学）；Istanbul Faculty of Medicine, Istanbul University（伊斯坦布尔大学医学学院）；Department of Biomedical Engineering and Radiology, University of Wisconsin-Madison（生物医学工程与放射科，威斯康星大学麦迪逊分校）；Department of Preventive Medicine, Northwestern University（预防医学系，西北大学）；Division of Gastroenterology and Hepatology, New York University（消化内科与肝病科，纽约大学）；Department of Electrical, Electronic and Computer Engineering, University of Catania（电气、电子和计算机工程系，卡塔尼亚大学）；NVIDIA；Department of Radiology, Columbia University（放射科，哥伦比亚大学）；Department of Radiology and Nuclear Medicine, Erasmus Medical Center（放射科与核医学科，埃因霍温医学院）；Department of Gastroenterology and Hepatology, Erasmus Medical Center（消化内科与肝病科，埃因霍温医学院）；Department of Radiology, New York University（放射科，纽约大学）；Division of Gastroenterology and Hepatology, Mayo Clinic Florida（消化内科与肝病科，迈阿密诊所佛罗里达分部）；Department of Gastroenterology and Hepatology, Northwestern University（消化内科与肝病科，西北大学）
The CSGM framework (Bora-Jalal-Price-Dimakis'17) has shown that deep generative priors can be powerful tools for solving inverse problems. However, to date this framework has been empirically successful only on certain datasets (for example, human faces and MNIST digits), and it is known to perform poorly on out-of-distribution samples. In this paper, we present the first successful application of the CSGM framework on clinical MRI data. We train a generative prior on brain scans from the fastMRI dataset, and show that posterior sampling via Langevin dynamics achieves high quality reconstructi…
IPMN Risk Assessment under Federated Learning Paradigm
Northwestern University（西北大学）；Istanbul University（伊斯坦布尔大学）；University of Catania（卡塔尼亚大学）；Erasmus Medical Center（伊拉斯姆斯医学中心）；University of Wisconsin（威斯康星大学）；Mayo Clinic（梅奥诊所）；New York University（纽约大学）；NVIDIA（英伟达公司）
CEPerFed: Communication-Efficient Personalized Federated Learning for Multi-Pulse MRI Classification
Xinjiang Engineering Research Center of Big Data and Intelligent Software, School of Software, Xinjiang University（大数据与智能软件工程研究中心，软件学院，新疆大学）；Hunan Province Key Laboratory on Bioinformatics, School of Computer Science and Engineering, Central South University（生物信息学湖南省重点实验室，计算机科学与工程学院，中南大学）；School of Science and Engineering, University of Dundee（科学与工程学院，邓迪大学）；Division of Biomedical Engineering, University of Saskatchewan（生物医学工程系，萨斯喀彻温大学）；Hunan Provincial Key Laboratory on Bioinformatics, School of Computer Science and Engineering, Central South University（生物信息学湖南省重点实验室，计算机科学与工程学院，中南大学）
Counterfactual MRI Data Augmentation using Conditional Denoising Diffusion Generative Models
Instituto Superior Técnico（高等技术学院）；Universidade de Lisboa（里斯本大学）；Digital Surgery LAB（数字外科实验室）；Champalimaud Foundation（尚帕利莫基金会）；Faculdade de Medicina, Universidade de Lisboa（里斯本大学医学院）；Instituto de Telecomunicações（电信研究所）
Federated learning (FL) was proposed to achieve collaborative machine learning among various clients without uploading private data. However, due to model aggregation strategies, existing frameworks require strict model homogeneity, limiting the application in more complicated scenarios. Besides, the communication cost of FL's model and gradient transmission is extremely high. This paper proposes Loosely Coupled Federated Learning (LC-FL), a framework using generative models as transmission media to achieve low communication cost and heterogeneous federated learning. LC-FL can be applied on sc…
Generative Priors for MRI Reconstruction Trained from Magnitude-Only Images Using Phase Augmentation
1 Institute for Diagnostic；Interventional Radiology, University Medical Center G\"ottingen, Germany 2 Institute of Biomedical Imaging, Graz University of Technology, Graz, Austria 3 Clinic for Neurology, University Medical Center Göttingen, Germany 4 German Centre for Cardiovascular Research (DZHK), partner site Lower Saxony, Germany 5 Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School
Generative AI like ChatGPT in Blockchain Federated Learning: use cases, opportunities and future
Southern Illinois University（南伊利诺伊大学）；University of Asia Pacific（亚太大学）；Green University of Bangladesh（孟加拉国绿色大学）；University of North Texas（北得克萨斯大学）；John Hopkins University（约翰霍普金斯大学）
Nanchang University（南昌大学）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；Southeast University（东南大学）
Decreasing magnetic resonance (MR) image acquisition times can potentially make MR examinations more accessible. Prior arts including the deep learning models have been devoted to solving the problem of long MRI imaging time. Recently, deep generative models have exhibited great potentials in algorithm robustness and usage flexibility. Nevertheless, none of existing schemes can be learned or employed to the k-space measurement directly. Furthermore, how do the deep generative models work well in hybrid domain is also worth being investigated. In this work, by taking advantage of the deep energ…
Med-DualLoRA: Local Adaptation of Foundation Models for 3D Cardiac MRI
Artificial Intelligence in Medicine Lab (BCN-AIM), Departament de Matemàtiques i Informàtica, Universitat de Barcelona, Spain（巴塞罗那大学人工智能医学实验室（BCN-AIM）、数学与计算机科学系，西班牙）；Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain（加泰罗尼亚高级研究机构（ICREA），巴塞罗那，西班牙）
A Calibrated Memorization Index (MI) for Detecting Training Data Leakage in Generative MRI Models
Department of Computer Science, University of York（约克大学计算机科学系）；School of Computer Science, University of Leeds（利兹大学计算机科学学院）；Department of Computing and Mathematics, Manchester Metropolitan University（曼彻斯特 Metropolitan 大学计算与数学系）；Department of Computer Science, University of Manchester（曼彻斯特大学计算机科学系）；Department of Cardiovascular Sciences, KU Leuven（鲁汶大学心血管科学系）
Federated learning (FL) enables distributed computation of machine learning models over various disparate, remote data sources, without requiring to transfer any individual data to a centralized location. This results in an improved generalizability of models and efficient scaling of computation as more sources and larger datasets are added to the federation. Nevertheless, recent membership attacks show that private or sensitive personal data can sometimes be leaked or inferred when model parameters or summary statistics are shared with a central site, requiring improved security solutions. In…
FedSemiDG: Domain Generalized Federated Semi-supervised Medical Image Segmentation
Medical Artificial Intelligence Lab, Westlake University（西交大学医学人工智能实验室）；Department of Information and Communication Engineering, School of Engineering, Institute of Science Tokyo（东京科学大学工程学院信息与通信工程系）；Department of Biomedical Engineering, The Chinese University of Hong Kong（香港中文大学生物医学工程系）
Generative Latent Representations of 3D Brain MRI for Multi-Task Downstream Analysis in Down Syndrome
HER - Human-Environment Research Group, La Salle - URL, Barcelona, Spain（HER-人类环境研究组，La Salle-URL，巴塞罗那，西班牙）；Memory Unit, Department of Neurology, Institut de Recerca Sant Pau – Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain（记忆单元，神经科，圣保罗研究所–圣十字与圣保罗医院，巴塞罗那自治大学，巴塞罗那，西班牙）；Neuroradiology Section, Department of Radiology – Hospital de la Santa Creu i Sant Pau, Universitat Autònoma de Barcelona, Barcelona, Spain（放射学部，放射科，圣保罗研究所–圣十字与圣保罗医院，巴塞罗那自治大学，巴塞罗那，西班牙）；Departament de Biologia Evolutiva, Ecologia i Ciències Ambientals (BEECA), Facultat de Biologia, Universitat de Barcelona (UB), Barcelona, Spain（进化生物学部，生态与环境科学（BEECA），生物学学院，巴塞罗那大学（UB），巴塞罗那，西班牙）
