cs.CVMar 29, 2026

Project Imaging-X: A Survey of 1000+ Open-Access Medical Imaging Datasets for Foundation Model Development

Authors: Zhongying Deng, Cheng Tang, Ziyan Huang, Jiashi Lin, Ying Chen, Junzhi Ning, Chenglong Ma, Jiyao Liu, +119 more

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Foundation models have demonstrated remarkable success across diverse domains and tasks, primarily due to the thrive of large-scale, diverse, and high-quality datasets. However, in the field of medical imaging, the curation and assembling of such medical datasets are highly challenging due to the reliance on clinical expertise and strict ethical and privacy constraints, resulting in a scarcity of large-scale unified medical datasets and hindering the development of powerful medical foundation models. In this work, we present the largest survey to date of medical image datasets, covering over 1,000 open-access datasets with a systematic catalog of their modalities, tasks, anatomies, annotations, limitations, and potential for integration. Our analysis exposes a landscape that is modest in scale, fragmented across narrowly scoped tasks, and unevenly distributed across organs and modalities, which in turn limits the utility of existing medical image datasets for developing versatile and robust medical foundation models. To turn fragmentation into scale, we propose a metadata-driven fusion paradigm (MDFP) that integrates public datasets with shared modalities or tasks, thereby transforming multiple small data silos into larger, more coherent resources. Building on MDFP, we release an interactive discovery portal that enables end-to-end, automated medical image dataset integration, and compile all surveyed datasets into a unified, structured table that clearly summarizes their key characteristics and provides reference links, offering the community an accessible and comprehensive repository. By charting the current terrain and offering a principled path to dataset consolidation, our survey provides a practical roadmap for scaling medical imaging corpora, supporting faster data discovery, more principled dataset creation, and more capable medical foundation models.

Figures & tables

Figure 1: Evolution of medical foundation models and general domain foundation models. Medical foundation models are mostly trained using millions of images, while advanced general domain ones are trained using billions of natural images. Additionally, most medical foundation models cover only a few modalities like CT and MRI, which may introduce modality-specific bias that constrains clinical applicability.
Figure 2: Conceptual overview of moving from fragmented medical image data to integrated resources for medical foundation models. Our survey addresses the data fragmentation issue in public medical image datasets by introducing a metadata-driven dataset integration paradigm, which is crucial for the development of advanced foundation models that can tackle multi-modality, multi-task, and multi-anatomy challenges effectively, ultimately enhancing clinical AI applications.
Figure 3: Overview of the survey. We first introduce the overview of the medical imaging datasets, followed by three sections detailing 2D, 3D, and video datasets. We further implement integration strategies to merge the datasets for large-scale resources, which can potentially be leveraged for the development of foundation models. Finally, we discuss the challenges for foundation model development.
Figure 4: Overview of medical imaging datasets: representative modalities by anatomical region (left), dataset distribution across modalities, anatomical regions, and tasks (upper right), and temporal trends in dataset numbers (lower right).
Figure 5: Taxonomy of medical imaging datasets across data dimensions, modalities, tasks, and anatomical organs.
Figure 6: The overview of image number in medical image datasets released from 2000 to 2025. (a) Total image number; Image number of different (b) dimensions, (c) modalities, (d) tasks, and (e) top five organs.
Figure 7: The distribution of (a) imaging dimensionalities, (b) modalities, (c) tasks, and (d) anatomical regions.
Figure 8: Representative modalities in medical imaging datasets.
Figure 9: Representative samples of three medical image analysis tasks: (a) classification, (b) segmentation, and (c) detection. Each column shows images from diverse modalities and clinical applications, illustrating the characteristic outputs of the respective task types.
Figure 10: Anatomical structures of medical image datasets. We also show the total number of datasets and images for each anatomical structure.
Figure 11: The distribution of different (a) modalities, (b) anatomical structures, and (c) tasks for for 2D labeled datasets. Each slice of the pie chart shows the percentage and the actual number of images.
Figure 12: Demonstration of the collected 2D medical datasets across different modalities and anatomical regions.
Figure 13: The distribution of different (a) modalities, (b) anatomical structures, and (c) tasks for 3D datasets. Each slice of the pie chart shows the percentage and the actual number of images.
Figure 14: 3D visualization examples of medical imaging datasets across different modalities and anatomical structures.
Figure 15: The distribution of different (a) modalities, (b) anatomical structures, and (c) tasks for video datasets
Figure 16: Demonstration of the collected video datasets from different tasks. The figure for the video estimation task is from the EndoVis 2022-SimCol-to-3D dataset (# ), and the figure for the video registration task is from the EndoVis 2022-P2ILF dataset (# ).
Figure 17: Illustration of four major modalities in the collected video datasets.
IDFieldBrief description
1dataset_nameOfficial name or commonly used short name of the dataset.
2release_dateFirst public release date (YYYY-MM or YYYY-MM-DD; use NA if unknown).
3homepage_urlStable URL or DOI for the dataset homepage, paper, or repository.
4organizationInstitution(s) releasing or hosting the dataset; multiple entries allowed, separated by commas.
5challenge_seriesName of the associated challenge or benchmark series; NA if not challenge-based.
6licenseData usage license or access policy as specified by the download agreement.
Table 1: Definition of data_meta fields for dataset-level metadata.
Figure 18: Pipeline of our dataset collection, processing, fusion, and summarization system based on the proposed dataset fusion paradigm, implemented in our interactive discovery portal.
BlockRoleRepresentative fields and examples
recordSample identifiers and cross-referencesdataset_name , image_path , optional sample IDs used to locate the underlying media file and join annotations across tasks.
contextClinical and textual contextSubject- and acquisition-level metadata (e.g., subject ID, age, sex, site, modality, anatomy), extensible extra dictionary, free-text descriptions or reports.
media_geometryMedia attributes and geometryMedia-level task configuration and spatio-temporal metadata, including task_type , leaf_task , annotation_type , dimension , pixel spacing, orientation, slice/frame indices, timestamps, camera parameters.
tasksTask-specific annotation payloadsStructured labels for the predefined tasks, such as segmentation masks, detection boxes, class labels, polygons, or keypoints, grouped by dimension and schema_variant .
Table 2: Four logical blocks— record , context , media_geometry , and tasks —structuring annotations_{task}.jsonl . The blocks separate sample identity, context, media-level geometry, and task-specific annotation payloads.
Figure 19: Detailed process of the proposed metadata-driven fusion paradigm (MDFP).
PhaseObjectiveMetadata Fields
1. HarmonizationStandardize modalities, tasks, and anatomy.modality_primary , dimension , anatomical_structure , organization , challenge_series
2. AlignmentAlign semantic labels and tasks across datasets.task_type , modality_secondary , label_presence , notes
3. BlueprintCluster datasets; assess integrative potential and data scale.data_volume , valid_image_n , storage_size_gb
4. IndexingCreate public metadata indices and visualization tools for easy access.dataset_name , release_date , homepage_url , license
Table 3: MDFP Workflow Overview with Key Metadata Fields
modalityn_datasetssum_imagen_orgslabeled_ratio
CT101,173,96541.000
MR5681,02521.000
Fundus42280,311170.952
Table 4: MDFP-derived composition for the 2D CT/MR/Fundus goal. sum_image is sum(valid_image_n) ; labeled_ratio is the fraction of datasets that are labeled.
Appendix figures & tables20 assets

Supplementary material from the paper’s appendix.

Appendix

#DatasetYearDimModalityStructureImagesLabelTaskDiseases
1LoDoPaB-CT Leuschner et al. [2021]20202DCTLung28YesReconNA
25K+ CT Images on Fractured Limbs Ruikar et al. [2021]20212DCTLimbs24YesSegBone Fracture
3AREN0534 Ehrlich et al. [2021]20213D, 2DMulti aKidney, Lung239NoEstKidney Tumor
4CT Medical Images Kohli et al. [2018]20172DCTLung475YesSegNA
5National Lung Screening Trial National Lung Screening Trial Research Team [2013]20133D, 2DCT, PathologyLung26.7kNoClsLung Cancer
6RSNA Intracranial Hemorrhage Detection Flanders et al. [2020]20192DCTBrain874kYesLocIntracranial Hemorrhage
Appendix
Table 5: 2D CT datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
37AREN0534 Ehrlich et al. [2021]20212D, 3DMulti aKidney, Lung239NoEstKidney Tumor
38KNOAP2020 Hirvasniemi et al. [2023b]20202D, 3DMR, X-RayKnee30YesPredOsteoarthritis
39braimMRI Alibaba Tianchi [2022]20222DMRBrain110YesSegBrain Tumor
40Brain-MRI Alibaba Tianchi [2020a]20202DMRBrain110YesSegBrain Disease
41SpinalDisease2020 Alibaba Tianchi [2020b]20202DMRSpine150YesDetSpinal Disease
42The Visible Human Project National Library of Medicine [1994]19942D, 3DCT, MR, OthersFull Body2NoNASkin Lesion
Appendix
Table 6: 2D MRI datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
59AREN0534 Ehrlich et al. [2021]20213D, 2DMulti aKidney, Lung239NoEstKidney
60ImageCLEF 2016 de Herrera et al. [2016]20152DMulti bSkin, Cell, Breast31kYesClsH&N Tumor
61CMB-CRC Biobank [2022c]20223D, 2DMulti cColon472NoSeg, ClsColorectal Cancer (H&E stained tissue)
62CMB-GEC Cancer Moonshot Biobank [2022a]20223D, 2DCT, WSI, PETBrain14NoSeg, ClsMelanoma (Cerebral microbleeds)
63CMB-MEL Biobank [2022a]20223D, 2DMulti dBrain255NoSegMelanoma (Cerebral microbleeds)
64CMB-MML Cancer Moonshot Biobank [2022c]20212D, 3DMulti eNA60NoPredMultiple Myeloma
Appendix
Table 7: 2D PET datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
70HC18 van den Heuvel et al. [2018]20182DUSSkull1.3kYesMeasNA
71BUSI Al-Dhabyani et al. [2020]20192DUSBreast647YesSegBreast Cancer
72APOLLO-5 The Cancer Imaging Archive (TCIA) [2025]20222D, 3DMulti aNA6.2kNoNANA
73CMB-LCA Biobank [2025]20222D, 3DMulti bNA0NoNANA
74ImageCLEF 2015 Garcia Seco De Herrera et al. [2015]20152D, 3DMulti cSkin, Cell, Breast0YesClsNA
75ImageCLEF 2016 de Herrera et al. [2016]20162DMulti dSkin, Cell, Breast31kYesClsHead & Neck Tumor
Appendix
Table 8: 2D Ultrasound datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
94Chest X-ray Kermany et al. [2018b]20182DX-RayLung5.9kYesClsPneumonia
95CoronaHack Praveen Govi [2019]20202DX-RayLung5.9kYesClsCOVID-19, Pneumonia
96NIH Chest X-ray 14 Wang et al. [2017]20172DX-RayLung112.1kYesClsThorax diseases
97COVIDx CXR-2 Wang et al. [2020a]20202DX-RayLung30.9kYesClsCOVID-19
98Pneumothorax Masks X-Ray Zawacki et al. [2019]20202DX-RayLung12.0kYesSegPneumothorax
99IRMA X-ray Raddar [2020]20202DX-RayBrain, Lung14.7kYesClsNA
Appendix
Table 9: 2D X-Ray datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
150OCT2017 Kermany et al. [2018a]20182DOCTRetina83.5kYesClsNA
151Retinal OCT - C8 Yang [2023]20212DOCTRetina24kYesClsNA
152iChallenge - AGE19 Fu et al. [2020]20192DOCTRetina1.6kYesClsNA
153DRAC22 Qian et al. [2024]20222DOCTRetina174YesSegDiabetic Retinopathy Lesions
154iChallenge - GOALS Fang et al. [2022a]20222DOCTRetina300YesSegNA
155Eye OCT Datasets Jahromi et al. [2014] , Mahmudi et al. [2025]20212DOCTRetina148YesClsNA
Appendix
Table 10: 2D OCT datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
171DRISHTI-GS Sivaswamy et al. [2014]20142DFundus PhotoRetina101YesSegOptic Disc
172CHASE Fraz et al. [2012]20092DFundus PhotoRetina28YesSegNA
173STARE Hoover et al. [2000]20042DFundus PhotoRetina40YesSegNA
174DRIVE Staal et al. [2004]20032DFundus PhotoRetina40YesSegNA
175IDRID2018 Porwal et al. [2018]20182DFundus PhotoRetina81YesSeg, ClsDiabetic Retinopathy
176EyePACS Cuadros and Bresnick [2009]20152DFundus PhotoRetina88.7kYesClsDiabetic Retinopathy
Appendix
Table 11: 2D fundus datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
242ISIC18 Codella et al. [2019]20182DDermoscopySkin2.7kYesSegSkin lesion
243ISIC20 Rotemberg et al. [2021]20202DDermoscopySkin33.1kYesClsBenign melanoma, malignant melanoma
244ISIC16 Gutman et al. [2016]20162DDermoscopySkin1.3kYesSegSkin lesion
245ISIC17 Codella et al. [2018]20162DDermoscopySkin2.8kYesSegSkin lesion
246Derm7pt Kawahara et al. [2018]20212DDermoscopySkin2.0kYesClsSkin lesion
247ISIC19 Combalia et al. [2022]20192DDermoscopySkin25.3kYesClsCells
Appendix
Table 12: 2D dermoscopy datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
258PANDA_radboud Bulten et al. [2022]20202DHistopathology (Patch)Prostate5.1kYesSegProstate Cancer
259Gleason Nir et al. [2018]20192DHistopathology (Patch)Prostate331YesSegProstate Cancer
260PathologyVQA He et al. [2020]20202DHistopathology (Patch)Full Body5.0kYesVQANA
261SLN-Breast Campanella et al. [2019]20192DHistopathology (WSI)Lymph166YesClsBreast Lymph Node
262MoNuSeg Kumar et al. [2019]20182DHistopathology (Patch)Nuclei51YesSegNA
263MoNuSAC2020 Verma et al. [2021]20192DHistopathology (Patch)Lung, Prostate, etc. a914YesSegNA
Appendix
Table 13: 2D histopathology datasets. (part 1/2)
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
319TIGER-wsirois Shephard et al. [2022]20222DHistopathology (Patch)Breast2.0kYesSegBreast Cancer
320TIGER-wsitils Shephard et al. [2022]20222DHistopathology (Patch)Breast82YesRegBreast Cancer
321Breast Cancer Cell Seg 2 Gelasca et al. [2008a]20202DHistopathology (Patch)Breast58YesSegBreast cancer
322Malignant Lymphoma Cls Dataset Orlov et al. [2010]20202DHistopathology (Patch)Lymph374YesClsLymphoma
323Lung and Colon Histopathology Borkowski et al. [2019]20202DHistopathology (Patch)Lung, Colon25kYesClsLung and Colon Cancer
324FocusPath Hosseini et al. [2019]20202DHistopathology (Patch)NA864YesIQAHistopathology Image
Appendix
Table 14: 2D histopathology datasets. (part 2/2)
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
372CellTracking2019 Maška et al. [2023]20192DMicroscopyCell1.4MYesTrackingNA
373CREMI [2016]20162DMicroscopyBrain375YesSegNA
374Bacteria Detection Wieczorek et al. [2024]20212DMicroscopyNA366YesSegNA
375Blood Cell Images [2018]20212DMicroscopyBlood12.5kYesClsBlood
376Leukemia Classification Santos and colleagues [2020]20212DMicroscopyNA15.1kYesClsLeukemia
377CellTracking2021 Maška et al. [2023]20212D+3D+VideoMicroscopyCell0YesTracking, SegLung Disease
Appendix
Table 15: 2D microscopy datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
406RAVIR Hatamizadeh et al. [2022]20222DInfraredRetina42YesSegBlood vessel
407MRL Eye Glasses cls Sojka and others [2018]20182DInfraredRetina84.9kYesClsNA
408MRL Eye Eye state cls Sojka and others [2018]20182DInfraredRetina84.9kYesClsNA
409MRL Eye Reflections cls Sojka and others [2018]20182DInfraredRetina84.9kYesClsNA
410MRL Eye Image quality cls Sojka and others [2018]20182DInfraredRetina84.9kYesClsNA
411MRL Eye Sensor type cls Sojka and others [2018]20182DInfraredRetina84.9kYesClsNA
Appendix
Table 16: 2D infrared datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
412Kavsir Pogorelov et al. [2017]20172DEndoscopyColon14kYesClsNA
413EndoSlam Ozyoruk et al. [2020]20212DEndoscopyColon, Liver, Stomach, Kidney76.8kYesRecon, EstNA
414SARAS-MESAD Bawa et al. [2021]20212DEndoscopyProstate50.3kNoDetGI disease
415EAD19 Ali et al. [2019]20182DEndoscopyStomach, Bladder, Colon2.1kYesDetEndo Artifact
416EndoCV2020-Sub Challenge1 Ali and Ghatwary [2022]20192DEndoscopyColon2.3kYesDet, SegPolyp
417EndoVis15 Bernal et al. [2017]20152DEndoscopyColon612YesSegPolyp
Appendix
Table 17: 2D endoscopy datasets.
#DatasetYearDimModalityStructureImagesLabelTaskDiseases
449Intel & MobileODT Cervical Cancer Screening Kaggle [2017]20172DColposcopyVagina1,993YesClsCervical cancer
450ADDI ALZHEIMER’S DETECTION CHALLENGE AIcrowd and Alzheimer’s Disease Data Initiative [2021]20212DSeries DataNA34,614YesClsLung adenocarcinoma
451The Digital Mammography DREAM Challenge Schaffter and others [2020]20162DMammographyBreast640,000YesClsAtrophic AMD
452BigNeuron Peng and others [2015]20162DNABrain2,166YesReconDiabetic foot ulcer
453Human Activity Classification with Radar Khan and others [2023]20192DOthersNA1,854YesClsNA
454KvasirCapsule-SEG Jha and others [2021]20212DColposcopy (Capsule)Polyp55YesSegPolyp
Appendix
Table 18: 2D datasets of the other modalities.
#DatasetYearDimModalityStructureVolumesLabelTaskDiseases
472TotalSegmentator Wasserthal et al. [2023b]20223DCTFull Body1204YesSegVaried pathologies
473AutoPET Gatidis et al. [2022]20223DCT/PETWhole-body1014YesSegCancer
474AutoPET II Gatidis et al. [2023]20233DCT/ PETWhole-body1219YesSegCancer
475ULS23 de Grauw et al. [2025]20233DCTChest, Abdomen, Pelvis39468YesSegOncological lesions
476DeepLesion Yan et al. [2018]20182D/3DCTBone, Abdomen, Mediastinum, Liver, Lung, Kidney…32735YesDetVarious lesions
477CT-ORG Rister et al. [2019]20193DCTBladder, Bone, Brain, Kidney, Liver, Lung140YesSegLiver lesions, Metastatic disease
Appendix
Table 19: 3D CT datasets
#DatasetYearDimModalityStructureVolumesLabelTaskDiseases
733TotalSegmentator MRI Akinci D’Antonoli et al. [2025]20243DMRWhole-body616YesSegVarious pathologies
734ACDC Bernard et al. [2018]20173DMRHeart150YesSeg/ ClsCardiac conditions
735M&Ms Campello et al. [2021]20203DMRHeart375YesSegCardiac diseases
736M&Ms-2 Camara et al. [2021]20213DMRHeart360YesSegCardiac conditions
737LAScarQS 2022 Li et al. [2023a]20223DMRHeart194YesSeg/QuanLeft atrial scar quantification and segmentation
738LAScarQS++ 2024 Li et al. [2024c]20243DMRHeart200+YesSeg/QuanMulti-center left atrial and scar segmentation
Appendix
Table 20: 3D MR datasets
#DatasetYearDimModalityStructureVolumesLabelTaskDiseases
964TDSC-ABUS2023 Luo et al. [2025]20233DUSBreast200YesSeg/ Cls/ DetBreast cancer
965CETUS2014 Bernard et al. [2014]20143DUSHeart45YesSegCardiac conditions
966MVSeg-3DTEE2023 Carnahan [2023]20233DUSMitral valve175YesSegMitral valve disease
967AREN0532 Children’s Oncology Group [2022a]20222D/3DUS/ CR/ CT/ MR/ PET/ RTIMAGEKidney544NoClsWilms tumor
968AREN0533 Children’s Oncology Group [2022b]20223DUS/CR/CT/MRKidney294NoClsWilms tumor
969AREN0533-Tumor-Annotations Rozenfeld and Jordan [2023c]20233DUS/CR/CT/MRKidney294YesSegWilms tumor
Appendix
Table 21: 3D US datasets
#DatasetYearDimModalityStructureVolumesLabelTaskDiseases
991QIN PET Phantom Beichel et al. [2015]20143DPETBody2YesSegPhantom study
992AutoPET Gatidis et al. [2022]20223DPET/CTWhole-body1014YesSegCancer
993AutoPET II Gatidis et al. [2023]20233DPET/CTWhole-body1219YesSegCancer
994COVID-19-NY-SBU Saltz et al. [2021]20212D/ 3DPET/CT/ MR/ X-RAYBrain/ Chest1384NoClsCOVID-19
995HECKTOR 2020 Andrearczyk et al. [2020]20203DPET/CTHead and Neck254YesSegHead and Neck Cancer
996HECKTOR 2021 Andrearczyk et al. [2022]20213DPET/CTHead and Neck325YesSeg/ RegHead and Neck Cancer
Appendix
Table 22: 3D PET datasets
#DatasetYearDimModalityStructureVolumesLabelTaskDiseases
1056MitoEM Wei et al. [2020]20203D3D MICROSCOPYBrain2YesSegMitochondrial ultrastructure
10573D Platelet EM Guay et al. [2021]20213D3D MICROSCOPYPlatelet2YesSegPlatelet ultrastructure
1058PCa Bx 3Dpathology Xie et al. [2023]20233D3D MICROSCOPYProstate50YesSeg/ LocProstate cancer
1059CADA Ivantsits et al. [2021]20203D3D DSABrain131YesDet/ Seg/ ClsCerebral aneurysms
1060ISBI2023 challenge - SHINY-ICARUS Dazeo and Larrabide [2023]20223D3D DSAInternal Carotid Artery82YesSegInternal carotid artery aneurysms
1061CADA-AS Ivantsits et al. [2022]20203D3D DSABrain110YesSegCerebral aneurysms
Appendix
Table 23: 3D Other datasets
#DatasetYearDimModalityStructureNum of samplesLabelTaskDiseases
1082CholecT50 Nwoye et al. [2023b]2023VideoEndoscopyGallbladder50YesCls, Detsurgical instrument, action, target
1083CholecTriplet 2021 Nwoye et al. [2023a]2021VideoEndoscopyGallbladder45YesCls, Detsurgical instrument, action, target
1084SurgVisDom Zia et al. [2021]2020VideoEndoscopyBowel488YesClsskin lesion
1085CATARACTS Al Hajj et al. [2019]2017VideoMicroscopyRetina50YesCls, Detsurgical workflow
1086EndoVis 2018-SWAS [2018]2018VideoEndoscopyColon42YesClssurgical phase
1087EndoVis 2019-SWSA Wagner et al. [2023]2019VideoEndoscopyGallbladder30YesClssurgical phase, action, instrument
Appendix
Table 24: Video datasets.

Explore similar work

CardsList

  1. MedPMC: A Systematic Framework for Scaling High-Fidelity Medical Multimodal Data for Foundation Models

    Jul 8, 2026Hyunjae Kim, Dain Kim, Pan Xiao +25Medical Imaging Foundation ModelsTraining Data Curation

    Yale University, New Haven, CT, USA · Korea University, Seoul, South Korea · Washington University in St. Louis, St. Louis, MO, USA +4

  2. CoM3^3eT: A foundation model for medical image analysis through federated, multidimensional context integration

    Aug 17, 2026J. Raphael Schäfer, Kai Geissler, Till Nicke +27Medical Imaging Foundation ModelsMulti-Task Learning

    Fraunhofer Institute for Digital Medicine MEVIS · Institute for Experimental Molecular Imaging, RWTH Aachen University, Aachen, Germany · Institute of Diagnostic and Interventional Radiology, Medizinische Hochschule Hannover, Hannover, Germany +11

  3. Foundation Models for Medical Imaging: Status, Challenges, and Directions

    Feb 17, 2026Chuang Niu, Pengwei Wu, Bruno De Man +1Medical Imaging Foundation ModelsMedical Imaging

    Biomedical Imaging Center, Center for Computational Innovations, Center for Biotechnology & Interdisciplinary Studies, Department of Biomedical Engineering, School of Engineering, Rensselaer Polytechnic Institute, Troy, New York, USA · GE HealthCare Technology & Innovation Center, Niskayuna, NY, USA