To curate a high-quality dataset, identifying data variance between the internal and external sources is a fundamental and crucial step. However, methods to detect shift or variance in data have not been significantly researched. Challenges to this are the lack of effective approaches to learn dense representation of a dataset and difficulties of sharing private data across medical institutions. To overcome the problems, we propose a unified pipeline called MedShift to detect the top-level shift samples and thus facilitate the medical curation. Given an internal dataset A as the base source, we first train anomaly detectors for each class of dataset A to learn internal distributions in an unsupervised way. Second, without exchanging data across sources, we run the trained anomaly detectors on an external dataset B for each class. The data samples with high anomaly scores are identified as shift data. To quantify the shiftness of the external dataset, we cluster B's data into groups class-wise based on the obtained scores. We then train a multi-class classifier on A and measure the shiftness with the classifier's performance variance on B by gradually dropping the group with the largest anomaly score for each class. Additionally, we adapt a dataset quality metric to help inspect the distribution differences for multiple medical sources. We verify the efficacy of MedShift with musculoskeletal radiographs (MURA) and chest X-rays datasets from more than one external source. Experiments show our proposed shift data detection pipeline can be beneficial for medical centers to curate high-quality datasets more efficiently. An interface introduction video to visualize our results is available at https://youtu.be/V3BF0P1sxQE.
Multimodal Human-AI Synergy for Medical Imaging Quality Control: A Hybrid Intelligence Framework with Adaptive Dataset Curation and Closed-Loop Evaluation
Zhongda Hospital（中大医院）；School of Medicine, Southeast University（东南大学医学院）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
Precision-medicine-toolbox: An open-source python package for facilitation of quantitative medical imaging and radiomics analysis
The D-Lab, Department of Precision Medicine（精准医学系D实验室）；GROW—School for Oncology, Maastricht University（马斯特里赫特大学GROW肿瘤学院）；GIGA Cyclotrone Research Center（GIGA回旋加速器研究中心）；University of Liege（列日大学）；Department of Radiology and Nuclear Medicine（放射与核医学系）；Maastricht University Medical Centre（马斯特里赫特大学医学中心）
Efficient Large Scale Medical Image Dataset Preparation for Machine Learning Applications
German Cancer Research Center (DKFZ)（德国癌症研究中心（DKFZ））；University of Heidelberg（海德堡大学）；Goethe University Frankfurt（法兰克福歌德大学）；Charité - Universitätsmedizin Berlin（柏林夏里特医学院）
Confounder-Aware Medical Data Selection for Fine-Tuning Pretrained Vision Models
West China Hospital, Sichuan University（四川大学华西医院）；School of Artificial Intelligence, Beijing University of Posts and Telecommunications（北京邮电大学人工智能学院）；College of Computer Science, Sichuan University（四川大学计算机学院）；School of Biomedical Engineering, University of Science and Technology of China（中国科学技术大学生物医学工程学院）
Automatic classification of prostate MR series type using image content and metadata
Brigham and Women’s Hospital（布莱根妇女医院）；German Cancer Research Center（德国癌症研究中心）；Isomics（伊索米克斯公司）；PixelMed Publishing（PixelMed出版公司）；Massachusetts General Hospital（麻省总医院）
Specialty-Oriented Generalist Medical AI for Chest CT Screening
Rensselaer Polytechnic Institute（伦斯勒理工学院）；Wake Forest University School of Medicine（维克森林大学医学院）；Massachusetts General Hospital（麻省总医院）；Harvard Medical School（哈佛医学院）
Most computer aided pathology detection systems rely on large volumes of quality annotated data to aid diagnostics and follow up procedures. However, quality assuring large volumes of annotated medical image data can be subjective and expensive. In this work we present a novel standardization framework that implements three few-shot learning (FSL) models that can be iteratively trained by atmost 5 images per 3D stack to generate multiple regional proposals (RPs) per test image. These FSL models include a novel parallel echo state network (ParESN) framework and an augmented U-net model. Additio…
Stanford University（斯坦福大学）；The University of Hong Kong（香港大学）；University of California, Santa Cruz（加州大学圣克鲁兹分校）
The collection and curation of large-scale medical datasets from multiple institutions is essential for training accurate deep learning models, but privacy concerns often hinder data sharing. Federated learning (FL) is a promising solution that enables privacy-preserving collaborative learning among different institutions, but it generally suffers from performance deterioration due to heterogeneous data distributions and a lack of quality labeled data. In this paper, we present a robust and label-efficient self-supervised FL framework for medical image analysis. Our method introduces a novel T…
Medical Image De-Identification Resources: Synthetic DICOM Data and Tools for Validation
University of Arkansas for Medical Sciences（亚拉巴马医学科学大学）；Frederick National Laboratory for Cancer Research（癌症研究国家实验室）；Ellumen, Inc.（Ellumen公司）；Deloitte Consulting LLP（德勤咨询有限公司）；National Cancer Institute（国家癌症研究所）；Pixelmed Publishing（Pixelmed出版公司）；National Heart, Lung, and Blood Institute（国家心脏、肺部和血液研究所）
Mayo Clinic（妙佑医疗国际）；DASA；Brown University（布朗大学）；Imaging Center Onex, Groupe 3R（Groupe 3R Onex影像中心）；Singularity Consulting；Nemours/A.I. DuPont Hospital for Children（尼莫尔斯/杜邦儿童医院）；De La Salle University（德拉萨大学）；Hospital Selayang（士拉央医院）；GE Healthcare（GE医疗）；Memorial Sloan Kettering Cancer Center（纪念斯隆·凯特琳癌症中心）；Aidoc；Weill Medical College of Cornell University（康奈尔大学威尔医学院）；Microsoft Research（微软研究院）；NYU School of Medicine（纽约大学医学院）；Emory University School of Medicine（埃默里大学医学院）；Foundation Institute for Research and Study of Diagnostic Imaging (FIDI)（诊断影像研究与学习基金会研究所（FIDI））
With the recent advances in A.I. methodologies and their application to medical imaging, there has been an explosion of related research programs utilizing these techniques to produce state-of-the-art classification performance. Ultimately, these research programs culminate in submission of their work for consideration in peer reviewed journals. To date, the criteria for acceptance vs. rejection is often subjective; however, reproducible science requires reproducible review. The Machine Learning Education Sub-Committee of SIIM has identified a knowledge gap and a serious need to establish guid…
Graph Representation Learning of Longitudinal Medical Imaging Trajectories for Treatment Response Prediction
Technical University of Munich（慕尼黑工业大学）；TUM University Hospital Rechts der Isar（慕尼黑工业大学伊萨尔河右岸大学医院）；Helmholtz Munich（慕尼黑亥姆霍兹中心）；Universitat de Barcelona（巴塞罗那大学）；Medical University of Vienna（维也纳医科大学）；King’s College London（伦敦国王学院）；Munich Center for Machine Learning（慕尼黑机器学习中心）
A data- and compute-efficient chest X-ray foundation model beyond aggressive scaling
Stanford Center for Artificial Intelligence in Medicine and Imaging（斯坦福大学人工智能医学与成像中心）；Stanford University（斯坦福大学）；Department of Radiology（放射科）；Department of Computer Science（计算机科学系）；Department of Pediatrics（儿科）；Department of Biomedical Data Science（生物医学数据科学系）；Department of Medicine（医学系）
Kaapana: A Comprehensive Open-Source Platform for Integrating AI in Medical Imaging Research Environments
Division of Medical Image Computing, German Cancer Research Center, Heidelberg, Germany（德国癌症研究中心医学影像计算部，海德堡，德国）；Medical Faculty, University of Heidelberg, Heidelberg, Germany（海德堡大学医学系，海德堡，德国）；Computer Science Faculty, University of Heidelberg, Heidelberg, Germany（海德堡大学计算机科学系，海德堡，德国）；Pattern Analysis and Learning Group, Department of Radiation Oncology, Heidelberg University Hospital, Heidelberg, Germany（放射肿瘤学部门模式分析与学习组，海德堡大学医院，海德堡，德国）；German Cancer Consortium (DKTK), Partner Site Heidelberg, Heidelberg, Germany（德国癌症联盟（DKTK），海德堡合作伙伴站点，海德堡，德国）；Heidelberg Institute of Radiation Oncology (HIRO), National Center for Radiation Research in Oncology (NCRO), Heidelberg, Germany（海德堡放射肿瘤学研究所（HIRO），肿瘤放射研究国家中心（NCRO），海德堡，德国）；National Center for Tumor Diseases (NCT), NCT Heidelberg, a partnership between DKFZ and the University Medical Center Heidelberg, 69120 Heidelberg, Germany（肿瘤疾病国家中心（NCT），海德堡NCT，德克萨斯儿童医学研究中心（DKFZ）与海德堡大学医学中心之间的合作，海德堡，德国）
University of Toronto（多伦多大学）；University of New Brunswick（新不伦瑞克大学）；Sunnybrook Health Sciences Centre（桑尼布鲁克健康科学中心）；St. Michaels Hospital（圣迈克尔医院）；Huron Digital Pathology（休伦数字病理公司）；Kingston Health Sciences Center（金斯顿健康科学中心）
The lack of well-annotated datasets in computational pathology (CPath) obstructs the application of deep learning techniques for classifying medical images. %Since pathologist time is expensive, dataset curation is intrinsically difficult. Many CPath workflows involve transferring learned knowledge between various image domains through transfer learning. Currently, most transfer learning research follows a model-centric approach, tuning network parameters to improve transfer results over few datasets. In this paper, we take a data-centric approach to the transfer learning problem and examine t…
Yale University（耶鲁大学）；PAII Inc.（PAII公司）；Chang Gung Memorial Hospital（长庚纪念医院）；PingAn Technology（平安科技）
As the demand for more descriptive machine learning models grows within medical imaging, bottlenecks due to data paucity will exacerbate. Thus, collecting enough large-scale data will require automated tools to harvest data/label pairs from messy and real-world datasets, such as hospital PACS. This is the focus of our work, where we present a principled data curation tool to extract multi-phase CT liver studies and identify each scan's phase from a real-world and heterogenous hospital PACS dataset. Emulating a typical deployment scenario, we first obtain a set of noisy labels from our institut…
University of Toronto（多伦多大学）；Sunnybrook Research Institute（桑尼布鲁克研究所）
Two of the most common tasks in medical imaging are classification and segmentation. Either task requires labeled data annotated by experts, which is scarce and expensive to collect. Annotating data for segmentation is generally considered to be more laborious as the annotator has to draw around the boundaries of regions of interest, as opposed to assigning image patches a class label. Furthermore, in tasks such as breast cancer histopathology, any realistic clinical application often includes working with whole slide images, whereas most publicly available training data are in the form of ima…
Technische Universität München（慕尼黑工业大学）；Johns Hopkins University（约翰斯·霍普金斯大学）
Within medical imaging, manual curation of sufficient well-labeled samples is cost, time and scale-prohibitive. To improve the representativeness of the training dataset, for the first time, we present an approach to utilize large amounts of freely available web data through web-crawling. To handle noise and weak nature of web annotations, we propose a two-step transfer learning based training process with a robust loss function, termed as Webly Supervised Learning (WSL) to train deep models for the task. We also leverage search by image to improve the search specificity of our web-crawling an…
Amasya University（阿马西亚大学）；Bilkent University（比尔肯特大学）；National Magnetic Resonance Research Center, Bilkent University（比尔肯特大学国家磁共振研究中心）；Sabuncu Brain Research Center, Bilkent University（比尔肯特大学萨本库脑研究中心）
Broadspread use of medical imaging devices with digital storage has paved the way for curation of substantial data repositories. Fast access to image samples with similar appearance to suspected cases can help establish a consulting system for healthcare professionals, and improve diagnostic procedures while minimizing processing delays. However, manual querying of large data repositories is labor intensive. Content-based image retrieval (CBIR) offers an automated solution based on dense embedding vectors that represent image features to allow quantitative similarity assessments. Triplet learn…
Project Imaging-X: A Survey of 1000+ Open-Access Medical Imaging Datasets for Foundation Model Development
Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Shanghai Innovation Institute（上海创新研究院）；Shanghai Institute of Optics and Fine Mechanics（上海光学精密机械研究所）；Fudan University（复旦大学）；University of Cambridge（剑桥大学）；Shanghai Jiao Tong University（上海交通大学）；The University of Hong Kong（香港大学）；Fuzhou University（福州大学）；University of Washington（华盛顿大学）；Stanford University（斯坦福大学）；Incept Labs；Monash University（莫纳什大学）；Ruijin Hospital, Shanghai Jiao Tong University School of Medicine（上海交通大学医学院附属瑞金医院）；Alibaba DAMO Academy（阿里巴巴达摩院）；The Hong Kong Polytechnic University（香港理工大学）；South China University of Technology（华南理工大学）；The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；Yau Mathematical Sciences Center, Tsinghua University（清华大学丘成桐数学科学中心）；Chinese Academy of Sciences（中国科学院）；Tsinghua University（清华大学）；Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；Artificial Intelligence Innovation and Incubation Institute, Fudan University（复旦大学人工智能创新与产业研究院）；Shanghai Academy of Artificial Intelligence for Science（上海科学智能研究院）；Nankai University（南开大学）；The Chinese University of Hong Kong（香港中文大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Zhejiang University（浙江大学）；School of Informatics, Xiamen University（厦门大学信息学院）；University College London（伦敦大学学院）；Sun Yat-sen University（中山大学）；Alibaba Group, DAMO Academy, New York, NY, USA（阿里巴巴集团达摩院（纽约））；Tongji University（同济大学）；University of Toronto（多伦多大学）；Department of Psychological and Cognitive Sciences, Tsinghua University（清华大学心理与认知科学系）；Johns Hopkins University（约翰霍普金斯大学）；Arizona State University（亚利桑那州立大学）；Academy for Clinical Innovation and Translation of Shanghai（上海临床创新转化研究院）；University of California, Santa Cruz（加州大学圣克鲁兹分校）；ELLIS Institute Finland（芬兰ELLIS研究所）；Aalto University（阿尔托大学）；Shandong University（山东大学）；Xi’an Jiaotong University（西安交通大学）
University College London（伦敦大学学院）；InstaDeep；University of Westminster（威斯敏斯特大学）
Supervised machine learning-based medical image computing applications necessitate expert label curation, while unlabelled image data might be relatively abundant. Active learning methods aim to prioritise a subset of available image data for expert annotation, for label-efficient model training. We develop a controller neural network that measures priority of images in a sequence of batches, as in batch-mode active learning, for multi-class segmentation tasks. The controller is optimised by rewarding positive task-specific performance gain, within a Markov decision process (MDP) environment t…
Despite continued advancement in recent years, deep neural networks still rely on large amounts of training data to avoid overfitting. However, labeled training data for real-world applications such as healthcare is limited and difficult to access given longstanding privacy, and strict data sharing policies. By manipulating image datasets in the pixel or feature space, existing data augmentation techniques represent one of the effective ways to improve the quantity and diversity of training data. Here, we look to advance augmentation techniques by building upon the emerging success of text-to-…
