OASIS-3: Longitudinal Neuroimaging, Clinical, and Cognitive Dataset for Normal Aging and Alzheimer Disease
Open accessThe DOI matches its Crossref record; 66 of the 66 references it cites were found in a registry or an index; 17 of the 34 sources its claims cite were checked for support.
Abstract
OASIS-3 is a compilation of MRI and PET imaging and related clinical data for 1098 participants who were collected across several ongoing studies in the Washington University Knight Alzheimer Disease Research Center over the course of 15 years. Participants include 605 cognitively normal adults and 493 individuals at various stages of cognitive decline ranging in age from 42 to 95 years. The OASIS-3 dataset contains over 2000 MR sessions, including multiple structural and functional sequences. PET metabolic and amyloid imaging includes over 1500 raw imaging scans and the accompanying post-processed files from the PET Unified Pipeline (PUP) are also available in OASIS-3. OASIS-3 also contains post-processed imaging data such as volumetric segmentations and PET analyses. Imaging data is accompanied by dementia and APOE status and longitudinal clinical and cognitive outcomes. OASIS-3 is available as an open access data set to the scientific community to answer questions related to healthy aging and dementia.
Overview
CoChat Summary
TL;DR
OASIS-3 is an open-access compilation of longitudinal MRI, PET, clinical, and cognitive data from 1098 participants collected over 15 years at the Washington University Knight ADRC, spanning cognitively normal aging and early-stage Alzheimer disease. The paper describes the dataset's contents, processing pipelines (FreeSurfer, PUP, Centiloid), and exploratory analyses demonstrating feasibility of volumetric and amyloid analyses.
Key findings
- The release includes 1098 participants aged 42.5–95.6 years: 605 cognitively normal adults and 493 at various stages of cognitive decline; 850 entered with CDR 0 (605 remained normal, 245 converted), and 248 entered with CDR > 0. The APOE ε4 allele was present in 439 participants.
- Imaging inventory: 2168 MR sessions (236 at 1.5T, 1932 at 3T) and 1608 PET sessions (999 PiB, 492 Florbetapir/AV45, 117 FDG); 1912 FreeSurfer segmentations passed QC, and PUP post-processing was included for 1356 PET sessions.
- Exploratory analyses showed a significant baseline whole-brain volume group difference (F(2,974) = 32.36, p < 0.001), with greater annual atrophy in those entering with dementia (−2.01%/year, SD = 2.15) and converters (−1.68%/year, SD = 1.91) versus stable controls (−1.13%/year, SD = 1.49).
- Baseline amyloid differed significantly across groups (F(2,724) = 191.39, p < 0.001), but rate of amyloid deposition did not differ (F(2,330) = 0.19, p > 0.05); baseline amyloid correlated with later conversion while baseline whole-brain volume did not.
Methods
- Design: data descriptor compiling participants from several ongoing Knight ADRC studies (Memory and Aging Project, Adult Children Study, Healthy Aging and Senile Dementia) recruited over 15 years, with clinical/cognitive assessments standardized via the Uniform Data Set (UDS) and dementia staged by CDR.
- Data: structural and functional MRI (T1w, T2w, FLAIR, DTI, BOLD, ASL, SWI, TOF, fieldmaps), amyloid (PiB, AV45) and metabolic (FDG) PET, APOE genotyping, and longitudinal clinical/cognitive outcomes; images converted to NIfTI/BIDS and de-identified.
- Analysis: MRI processed with FreeSurfer for volumetric segmentation; PET processed with the PET Unified Pipeline (PUP) including partial volume correction and Centiloid standardization; exploratory group comparisons (stable controls, converters, dementia at entry) used ANOVA and t-tests on whole-brain volume and amyloid measures.
Limitations
- Stated: a standardized visit structure was not maintained across data collection, so investigators must use "days from entry" or "age at visit" to link imaging and clinical visits, and no association between visits was imposed; there is no categorical variable indicating healthy versus Alzheimer Disease status.
- Stated: the published OASIS datasets (Cross-sectional, Longitudinal, OASIS-3) contain overlapping participant data and should never be combined or aggregated, and links across datasets cannot be provided.
- Evident from the text: the exploratory analyses are described by the authors as "cursory" and "exploratory descriptive analyses" rather than confirmatory tests, and imaging was acquired across multiple scanner models and field strengths over 10+ years.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
12 Valid · 0 Not valid · 8 Not checked · 80 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Quality control reviewed post-processing PUP output 42,43 , which includes uptake measure for FreeSurfer-based regions, was included for 1356 PET sessions.”
What the sources say
- Partly supportedcites [42] 42. Su, Y. et al. Partial volume correction in quantitative amyloid imaging. NeuroImage 107 , 55-64 (2015). sourceExperimental — accuracy not yet measured
The Image Analysis section (page unavailable) describes quality assurance of FreeSurfer segmentation and PET uptake measures for FreeSurfer-defined regions, supporting the description of the output. It does not identify that output as PUP; the 1356-session count describes the citing authors’ dataset, not a result attributed to this source. (LLM, conf=0.78)
- Supportedcites [43] 43. Su, Y. et al. Quantitative analysis of PiB-PET with FreeSurfer ROIs. PloS One 8 , e73377 (2013). source
The Software section describes a FreeSurfer-based PET quantification workflow with output data types and quality-control reports. The FreeSurfer based analysis section describes quantitative measurements from FreeSurfer-defined regions. The citation supports the description of the processing output; the 1,356-session count describes the citing study’s own data, not a result attributed to this source. (LLM, conf=0.86)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Centiloid values 53 have been provided to standardize values across amyloid tracers PiB and Florbetapir.”
What the sources say
- Partly supportedcites [53] 53. Su, Y. et al. Utilizing the Centiloid scale in cross-sectional and longitudinal PiB PET studies. NeuroImage Clin. 19 , 406-416 (2018). sourceExperimental — accuracy not yet measured
The Discussion (page unavailable) says the Centiloid scale puts measurements on a common scale to facilitate comparisons between tracers. The Methods describe Centiloid conversion for PiB measurements in GAAIN and Knight ADRC data, but the paper does not say that it provided Centiloid values for both PiB and florbetapir in OASIS-3. (LLM, conf=0.86)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“OASIS-3 is a compilation of MRI and PET imaging and related clinical data for 1098 participants who were collected across several ongoing studies in the Washington University Knight Alzheimer Disease Research Center over the course of 15 years.”
plainn = 1,098- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“OASIS-3 also contains post-processed imaging data such as volumetric segmentations and PET analyses.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The published OASIS datasets - OASIS Cross-sectional, OASIS Longitudinal, and OASIS-3 - contain overlapping participant data and should never be combined or aggregated.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
35 more claims
Create a free account to see all 100 claims and every source verdict.
- It may state two separate claims as one. 27
- Probably not a substantial claim. 45
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“we utilize cognitively normal data from the Open Access Series of Imaging Studies - 3 (OASIS3) (LaMontagne et al., 2019) and the Alzheimer's Disease Neuroimaging Initiative (ADNI) (Alzheimer's Disease Neuroimaging Initiative (ADNI), 2004) datasets, as they are free from major brain pathologies and artifacts.”
From IterMask3D: Unsupervised Anomaly Detection and Segmentation with Test-Time Iterative Mask Refinement in 3D Brain MR (2025) · cites this paper as LaMontagne et al., 2019
Experimental — accuracy not yet measured
Statistics that don’t add up
Each reported p-value recomputed from its test statistic and degrees of freedom, each partial η² from its F, and each ratio, mean difference or B printed with a p checked against its own confidence interval.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“t (370) = 3.02, p = 0.002”
Reported p = .002 · recomputed p = .003
Statistics recomputed
Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.
- Doesn’t add upt(370) = 3.02reported p = .002 · recomputed p = .003
“t (370) = 3.02, p = .002”
- ConsistentF(2, 974) = 32.36reported p < .001 · recomputed p < .001
“F(2,974) = 32.36, p < .001”
- Consistentt(749) = 7.98reported p < .001 · recomputed p < .001
“t (749) = 7.98, p < .001”
- Consistentt(424) = 5.77reported p < .001 · recomputed p < .001
“t (424) = 5.77, p < .001”
- Consistentt(771) = 1.54reported p = .125 · recomputed p = .124
“t (771) = 1.54, p = .125”
8 more recomputed tests
Create a free account to see every test the Stats check recomputed.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Exploratory MRI and PET Analyses · p. 22
“Dementia at aging started in the study with CDR > 0.”
§ Exploratory MRI and PET Analyses · p. 23
“Those who entered the study with dementia (defined as CDR > 0.5) had significantly higher amyloid levels”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Data description · p. 20
“OASIS-3 includes 1608 PET imaging sessions comprised of 999 PiB, 492 Florbetapir (AV45), and 117 FDG scans (Table 4).”
§ Data description · p. 20
“| PIB | 1000 | 937 |”
Citations
Cites
Desikan, R. S., Ségonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D., Buckner, R. L., Dale, A. M., Maguire, R. P., Hyman, B. T., et al. (2006) · NeuroImage · doi:10.1016/j.neuroimage.2006.01.021 · Cited by 13,324
Cortical Surface-Based Analysis
Dale, A. M., Fischl, B., & Sereno, M. I. (1999) · NeuroImage · doi:10.1006/nimg.1998.0395 · Cited by 10,910
The Clinical Dementia Rating (CDR): Current version and scoring rules
Morris, J. C. (1993) · Neurology · doi:10.1212/wnl.43.11.2412-a · Cited by 9,882 · cited 1×
“In the Trail Making Test participants were asked to make a trail by connecting a series of numbers (1-26) for part A and connecting a series of alternating numbers and letters (1-A-2-B) for part B.”
Self-citationFischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C., van der Kouwe, A., Killiany, R., Kennedy, D., Klaveness, S., et al. (2002) · Neuron · doi:10.1016/s0896-6273(02)00569-x · Cited by 8,610
The organization of the human cerebellum estimated by intrinsic functional connectivity
Buckner, R. L., Krienen, F. M., Castellanos, A., Diaz, J. C., & Yeo, B. T. T. (2011) · Journal of Neurophysiology · doi:10.1152/jn.00339.2011 · Cited by 7,304 · cited 1×
“…development of neuroimaging tools, such as brain atlases 3,4 , tissue segmentation algorithms 5 , and de-identification tools 6 , used to create models of healthy aging 7,8 and AD timecourse 7–11 , used to answer hypothesis driven questions 12,13 , and used for educational and training purposes.”
Sperling, R. A., Aisen, P. S., Beckett, L. A., Bennett, D. A., Craft, S., Fagan, A. M., Iwatsubo, T., Jack, C. R., Kaye, J., Montine, T. J., et al. (2011) · Alzheimer's & Dementia · doi:10.1016/j.jalz.2011.03.003 · Cited by 7,155
Cortical Surface-Based Analysis
Fischl, B., Sereno, M. I., & Dale, A. M. (1999) · NeuroImage · doi:10.1006/nimg.1998.0396 · Cited by 6,147 · cited 2×
“…cortical models are complete, a number of deformable procedures were performed for further data processing and analysis including surface inflation (Fischl et al., 1999a), registration to a spherical atlas which is based on individual cortical folding patterns to match cortical geometry across…”
BackgroundMethodMeasuring the thickness of the human cerebral cortex from magnetic resonance images
Fischl, B., & Dale, A. M. (2000) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.200033797 · Cited by 5,818 · cited 2×
“…white matter boundary, automated topology correction 25,36 , and surface deformation following intensity gradients to optimally place the gray/white and gray/cerebrospinal fluid borders at the location where the greatest shift in intensity defines the transition to the other tissue class 23,24 .”
Hypothetical model of dynamic biomarkers of the Alzheimer's pathological cascade
Jack, C. R., Knopman, D. S., Jagust, W. J., Shaw, L. M., Aisen, P. S., Weiner, M. W., Petersen, R. C., & Trojanowski, J. Q. (2010) · The Lancet Neurology · doi:10.1016/s1474-4422(09)70299-6 · Cited by 4,806 · cited 1×
“In the Trail Making Test participants were asked to make a trail by connecting a series of numbers (1-26) for part A and connecting a series of alternating numbers and letters (1-A-2-B) for part B.”
A nonparametric method for automatic correction of intensity nonuniformity in MRI data
Sled, J. G., Zijdenbos, A. P., & Evans, A. C. (1998) · IEEE Transactions on Medical Imaging · doi:10.1109/42.668698 · Cited by 4,731 · cited 1×
“…matter and deep gray matter volumetric structures (including hippocampus, amygdala, caudate, putamen, ventricles) 26,27 intensity normalization 35 , tessellation of the gray matter white matter boundary, automated topology correction 25,36 , and surface deformation following intensity…”
Cited by
Items 701–788 of 788
Brain volumes discriminate clinical dementia rating scale categories
Christova, P. S., James, L. M., & Georgopoulos, A. P. (2025) · Scientific Reports · doi:10.1038/s41598-025-23418-9
“The OASIS-3 is a large and publicly available set of imaging and cognitive data 33 .”
Machine Learning for Longitudinal Brain-Age PredictionPreprint
Wegmann, M., Ganz, M., Svensson, J. E., Plavén‐Sigray, P., & Dörfel, R. P. (2025) · medRxiv · doi:10.64898/2025.12.10.25341964 · Cited by 2
Probabilistic brain MR image transformation using generative models
Rezvani, S., Moazami, S., Azevedo, C. J., & Oberai, A. A. (2025) · Scientific Reports · doi:10.1038/s41598-025-31958-3 · Cited by 1
“Another example is the effectiveness of the models in performing several image transformations on the OASIS test subjects that exhibit a range of pathologies, such as brain atrophy and Alzheimer’s disease lesions.”
InfluentialThe Data Behind the Model: Gaps and Opportunities for Foundation Models in Brain Imaging
Ghamizi, S., Kanli, G., Perquin, M., & Keunen, O. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-07845-2_11 · Cited by 1
Functional Connectome Alterations Across the Spectrum of Alzheimer’s Disease
Ghaffari, A., Zhao, Y., Abouzaki, M., Romero, Y., Langley, J. A., & Hu, X. (2025) · Journal of dementia and Alzheimer's disease · doi:10.3390/jdad2040046 · Cited by 3
“…utilize open-source datasets such as the Alzheimer’s Disease Neuroimaging Initiative (ADNI) [151], Open Access Series of Imaging Studies (OASIS) [152], and PResymptomatic EValuation of Experimental or Novel Treatments for Alzheimer’s Disease (PREVENT-AD) [153] to perform AD-related FC analyses…”
TissueAgeNet: Quantitative Textual Guidance for Tissue Level Brain Age Estimation
Chen, S., Wu, W., Zhang, X., Xu, J., Yu, S., Lv, Y., Xiong, T., Guo, J., Yang, C., & Xu, X. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-09513-8_25
End-to-End Cortical Surface Reconstruction from Clinical Magnetic Resonance Images
Nielsen, J. D., Gopinath, K. K., Hoopes, A., Dalca, A. V., Magdamo, C., Arnold, S. E., Das, S., Thielscher, A., Iglesias, J. E., & Puonti, O. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-09513-8_21 · Cited by 1
Hadadian, Z., Lotfi, M.-S., & Rezaei-Shahsavarloo, Z. (2026) · Aging Clinical and Experimental Research · doi:10.1007/s40520-025-03299-8 · Cited by 1
Hüppi, R. M., Langer, N., Vieira, B. H., & Initiative, T. F. L. D. N. (2026) · medRxiv · doi:10.64898/2026.01.13.26343927
Xie, M., Xiao, Q., Wu, H., Zhang, Y., Han, H., Xie, X., Zhang, W., Cheng, J., & Xu, J. (2026) · npj Digital Medicine · doi:10.1038/s41746-026-02347-5 · Cited by 1
López-Ríos, E. E., & Padilla, F. J. A. (2026) · World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany · doi:10.1007/978-3-032-13729-6_27
Cánovas, R., Cox, T. O., Doré, V., Bourgeat, P., Fripp, J., Feizpour, A., Shishegar, R., Fowler, C. J., Laws, S. M., Porter, T., et al. (2026) · Annals of Neurology · doi:10.1002/ana.78163 · Cited by 4
“33 A β -PET Imaging A β -PET imaging was performed at ADNI, AIBL, J-ADNI, and OASIS sites using previously published protocols, 21,34 – 36 considering 4 radiotracers: [11C] PiB, 18F NAV4694 (NAV), 18F Flutemetamol (Flute), and 18F Florbetapir (FBP).”
Self-citationElleaume, C., Vieira, B. H., Floris, D. L., Langer, N., & ageing, T. A. I. B. A. L. F. S. O. (2026) · eLife · doi:10.7554/elife.108952.2
Liu, X., Le, X., Huang, Z., Wang, Y., Zhang, Q., Feng, J., & Li, C. (2026) · medRxiv · doi:10.64898/2026.01.26.26344800
Fiszer, J., Ciupek, D., Malawski, M., & Pieciak, T. (2026) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2026.109649 · Cited by 2
Accelerating MRI With Longitudinally‐Informed Latent Posterior Sampling
Urman, Y., Shah, Z., Kumar, A., Soares, B. P., & Setsompop, K. (2026) · Magnetic Resonance in Medicine · doi:10.1002/mrm.70257 · Cited by 3
“However, to our knowledge, existing public longitudinal datasets 18,19,20,21 only include magnitude images, typically in DICOM format, which are appropriate for post-processing applications but insufficient for training complex-valued reconstruction models 22 .”
Location patterns and longitudinal progression of white matter hyperintensitiesPreprint
Zhao, X., Malone, I. B., Brown, T. M., Wong, A., Cash, D. M., Chaturvedi, N., Hughes, A. D., Schott, J. M., Barkhof, F., Barnes, J., et al. (2026) · medRxiv · doi:10.64898/2026.02.20.26346709
“Validation dataset: OASIS-3 OASIS-3 is a single-site longitudinal retrospective neuroimaging cohort from the United States (US) aged 42-95 [26].”
Wei, X., Zhang, T., Xiong, R., Zhang, Q., Zhang, J., Jin, Z., & Li, L. (2026) · Translational Psychiatry · doi:10.1038/s41398-026-03902-0 · Cited by 6
“…normal individuals (CN) aged 18 to 92 years, sourced from two datasets: the Cambridge Centre for Ageing and Neuroscience (Cam-CAN; n = 645; http://www.mrc-cbu.cam.ac.uk/datasets/camcan) [25, 26] and the Open Access Series of Imaging Studies-3 (OASIS-3; n = 540; https:// www.oasis-brains.org) [ 27].”
Toward Resilient AI in Medical Imaging: Handling and Mining Multiple Sources of Data
Gravina, M., Simone, A. D., Pontillo, G., Shawa, Z., Marrone, S., Fusco, R., Granata, V., Petrillo, A., Cole, J. H., Sansone, M., et al. (2026) · SN Computer Science · doi:10.1007/s42979-026-04821-z
“…Study of Aging (MCSA) [ 75], the NKI Rockland Sample Initiative (NKI) [ 76], the third release of Open Access Series of Imaging Studies (OASIS) [18 ], the Parkinson’s Progression Markers Initiative (PPMI) [77], the Consortium for Reliability and Reproducibility (CORR) [ 78], and the Health and…”
Zhang, T., Nie, B., Liu, H., Mao, G., OuYang, C., Jiang, X., & Shan, B. (2026) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-026-07797-3 · Cited by 1
“Using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) [35], we established spatiotemporal trajectories of Aβ deposition, tau accumulation, and brain atrophy, and validated these trajectories using the independent Open Access Series of Imaging Studies-3 (OASIS-3) dataset [36].”
Tracking early cognitive decline in preclinical AD with brain MRI similarity
Sun, J., Gerbolés, B. Z., Passaretti, M., Volpe, G., Mijalkov, M., Pereira, J. B., & Initiative, A. D. N. (2026) · Alzheimer s & Dementia · doi:10.1002/alz.71170 · Cited by 1
“23,34,37 No additional preprocessing was performed prior to analysis.”
InfluentialMaleki, A., Moghadam, M. Z., & Pławiak, P. (2025) · doi:10.1145/3783862.3784088
“OASIS-3 (Open Access Series of Imaging Studies-3) [26] (reference for the dataset) is a comprehensive, publicly accessible neuroimaging dataset to advance research in brain ageing and Alzheimer’s disease.”
Single-Subject Multi-view MRI Super-Resolution via Implicit Neural Representations
Kim, H., Thanki, A., van Herten, R. L. M., Margolis, D., & Sabuncu, M. R. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38189-7_47 · Cited by 1
Yan, Y., Song, D., Li, G., Li, J., Tang, Y., Li, D., Mao, J., Li, H., Liu, X., Yu, D., et al. (2026) · Molecular Psychiatry · doi:10.1038/s41380-026-03550-2
“Furthermore, 4-month-old 5xFAD mice did not display anxiety-like behaviors at this stage (Supplementary Fig.”
Dibble, A. J., Dalby, C., Sevegnani, M., Fracasso, A., Lyall, D. M., Harvey, M., Svanera, M., Initiative, T. F. L. D. N., & Initiative, T. F. L. D. N. (2026) · medRxiv · doi:10.64898/2026.03.27.26349489 · Cited by 3
“…Degeneration Neuroimaging Initiative), IXI (n=569, 569 volumes), a publicly available structural MRI dataset, OASIS-3 (n=1,025; 2,633 volumes) [70], LDM100k (n=99,994; 99,994 volumes) [28, 29], ADHD200 (n=880; 880 volumes) [52], ABIDE (n=904; 904 volumes) [53, 54], FCON1000 (n=6,904; 7,820…”
Taha, B. R., Benson, A., Arko, L., Harel, N., Sandoval-Garcia, C., Guillaume, D. J., & McGovern, R. A. (2026) · Frontiers in Neurology · doi:10.3389/fneur.2025.1639381
“In the computer vision literature, models with new architectures are often trained on publicly available data due to its accessibility (16–18).”
Attention-Guided Multimodal Neuroimaging Fusion Network for Modeling Brain Aging PatternPreprint
Wan, Z., Fu, W., Hossain, J., Gollo, L. L., & Wu, K. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.03.28.713645 · Cited by 1
“Additionally, this work further evaluated the proposed method using neuroimaging data from the Open Access Series of Imaging Studies, Phase 3 (OASIS-3) [77].”
Fouille de texte et extraction d'informations dans les données cliniques
Dalloux, C. (2020) · theses.fr (ABES) · doi:10.70675/d09cc9c4z1370z404cz96b2zb4f16b7d15c8 · Cited by 3
Ameri, R., Shahmirzadi, D., Yarahmadi, A., & Nadimi-Shahraki, M. H. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.132331 · Cited by 1
Federated learning of biomedical data in multicentric imaging studies
Rincon, S. S. S. (2023) · theses.fr (ABES) · doi:10.70675/7dd57e78zc2d5z412az9574zee90bd35ea4c
Machine Learning Applications in Medical Diagnosis, case study : bone metastasis
Afnouch, M. (2023) · theses.fr (ABES) · doi:10.70675/66b05d1fz9550z4a42zae8fz01f09904c294
Robust Learning for Medical Image Segmentation
Galati, F. (2024) · theses.fr (ABES) · doi:10.70675/49df4b56zc3d3z4169z8388zbcb5428e83ca
Segmentation et modélisation du réseau vasculaire cérébral à partir d'images IRM
Rougé, P. (2025) · theses.fr (ABES) · doi:10.70675/18a3ccffzcd56z475ez9418zf6808e89c0e0
Dhinagar, N. J., Jagad, C., Senthilkumar, P., et al. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.04.10.717865
Khalilullah, K. M. I., Phadikar, S., Agcaoglu, O., Sui, J., Duda, M., Adalı, T., & Calhoun, V. D. (2026) · Network Neuroscience · doi:10.1162/netn.a.575 · Cited by 6
An Artifact-Based Agent Framework for Adaptive and Reproducible Medical Image Processing
Zuo, L., Liu, Y., Rudravaram, G., Ramadass, K., Krishnan, A. R., Phillips, M. D., Bodien, Y. G., Patel, M. B., Trujillo, P., Forero, Y. J. J., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38059-3_1 · Cited by 3
Multimodal diagnosis of Alzheimer’s disease through causal imaging markers and risk factorsPreprint
Chilla, G. S. V. N. (2026) · medRxiv · doi:10.64898/2026.05.01.26352207
Bi, P., Zhou, Y., & Hu, Z. (2026) · Applied Sciences · doi:10.3390/app16104634
Deep Computational Anatomy via Latent-Aligned Multiview Normalizing FlowsPreprint
Tustison, N. J., Avants, B., Cook, P. A., Gee, J. C., & Stone, J. R. (2026) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.64898/2026.05.05.723039
“…cohorts in the experiments below: the Dallas Life Brain Study 63,64 , the NIMH Healthy Research Volunteer Dataset 65 , the Queensland Twin IMaging dataset 66,67 , the Brain Tumor Sequence Registration Challenge dataset 68 , and the Open Access Series of Imaging Studies 3 (OASIS-540 3) cohort 69 .”
Ly, P., Milongo, J., Xie, Y., Tian, X., He, J. (., & Zhao, C. (2026) · ACM Southeast Regional Conference (ACMSE) · doi:10.1145/3746467.3801526
“To address this limitation, we explicitly focus on large-scale multi-cohort learning by aggregating baseline T1-weighted MRI data from four major sources: OASIS-3 [26], OASIS-4 [21], ADNI [39], and AIBL [8].”
Kazi, A., Mora, J., Fischl, B. R., Dalca, A. V., & Aganj, I. (2026) · Frontiers in Neuroscience · doi:10.3389/fnins.2026.1786355 · Cited by 1
“All are defined as: (3) h 4 is the Frobenius inner product of the G matrix by a weight matrix E ∈ φ learned by the model, followed by sigmoid non-linearity, σ .”
Geometric brain signatures of Alzheimer’s disease progression and subtypesPreprint
Tong, B., Cao, T., Duong-Tran, D., Davatzikos, C. A., Thompson, P. M., Saykin, A. J., Fornito, A., Shen, L., & Initiative, T. F. L. D. N. (2026) · medRxiv · doi:10.64898/2026.05.14.26353211 · Cited by 1
Kaviani, S. I., Sanaat, A., Mokri, M., Zaidi, H., & Carrier, J. (2026) · EJNMMI Research · doi:10.1186/s13550-026-01448-2 · Cited by 1
“For the OASIS-3 dataset [40], a dataset including 30 cognitively normal subjects was selected for assessing the model belonging to 18 F-Floebetapir.”
Gray matter brain age predicts cognitive outcome one year after ischemic stroke
Tseng, W. I., Hsieh, Y.-C., Hsu, Y., Huang, L.-K., Fu, C.-K., Chen, D. Y.-T., Chen, J.-H., Hong, C., Lu, Y., Chan, L., et al. (2026) · Journal of Alzheimer s Disease · doi:10.1177/13872877261450602
Towards generalisable foundation models for brain MRI
Mazher, M., Parker, G. J. M., & Alexander, D. C. (2026) · npj Imaging · doi:10.1038/s44303-026-00176-5 · Cited by 11
“[30] 4,134 3,514 620 HABS [29] 580 - - ADNI [31] 382 235 57 AIBL [33] 152 129 23 NIFD [34] 346 294 52 BraTS [24] 1,107 941 166 FeTA [35] 80 68 12 ReMIND [25] 114 - - UPENN-GBM [26] 1,008 - - UCSF-PDGM [27] 588 500 88 Burdenko-GBM-Progression [28] 360 - -”
Stage-aware disentangled conditional VAE for interpretable prediction of cognitive decline
Yarushev, S., Neelakandan, S., Banupriya, N., & Velmurugan, S. (2026) · Discover Computing · doi:10.1007/s10791-026-10178-x
“The dataset [31] is available for research purposes at https://sites.wustl.edu/oasisbrains/.”
Baek, G., Salat, D. H., Jang, I., & Initiative, A. D. N. (2026) · Human Brain Mapping · doi:10.1002/hbm.70548 · Cited by 1
Tang, C., Tang, A., Wan, M., Li, C., Ruan, W., Liu, F. M., Fang, H., Gai, Y., Jiang, D., Zhou, W., et al. (2026) · NeuroImage · doi:10.1016/j.neuroimage.2026.122039 · Cited by 1
Jis, J. K., Chen, K., Chen, Z., Chen, L., Pouriyeh, S., Xie, Z., & Xie, Y. (2026) · Bioengineering · doi:10.3390/bioengineering13060698
“OASIS provides imaging modalities such as DTI, FLAIR, SWI, PET, and T1w imaging, along with clinical and cognitive data for studying connectivity, brain structure, and pathology [132].”
Pereira, H. R., Diogo, V. S., Fonseca, J., Ferreira, H. A., & Prata, D. P. (2026) · Clinical Neuroradiology · doi:10.1007/s00062-026-01688-8 · Cited by 2
Decomposition‐based harmonization for quantitative PET imaging across scanners and radiotracers
Zhao, S., Huang, Q., Nie, B., Zhang, T., Zuo, C. Y., Zhang, G., Mao, G. G., & Shan, B. (2026) · Medical Physics · doi:10.1002/mp.70545
Mirror-Fusion Attention for Reflection-Aware Self-supervised Representation Learning
Li, R. K., Liu, J., Shi, Y., & Lodi, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37670-1_36
“We use five datasets in total: CheXpert [20] and BraTS [25] as medical down-stream benchmarks, OASIS-3 [23] as additional unlabeled MRI pretraining data, and CelebA-HQ [27] and WFLW [39] as natural-image downstream benchmarks.”
Kumari, A., Basha, N. K., Shaffi, N., Viswan, V., & Mahmud, M. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-9575-4_20 · Cited by 1
Tsuchiya, R., Usui, K., Sakamoto, H., Kyogoku, S., & Daida, H. (2026) · Radiological Physics and Technology · doi:10.1007/s12194-026-01108-z
“…T2-weighted imaging (T2WI) scans of 100 patients were acquired from a publicly available dataset (Open Access Series of Imaging Studies), an open-access information source created by the Knight Alzheimer’s Disease Research Center at the Washington University School of Medicine in St. Louis [32].”
Classification of tau status with machine learning models in amyloid‐positive cohorts
Lin, Y., Chiu, S.-C., Massey, R., Fang, Y.-H. D., & Initiative, A. D. N. (2026) · Alzheimer s & Dementia · doi:10.1002/alz.71683 · Cited by 1
Haddada, K., Zaabi, M., Khedher, M. I., & Jemai, O. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31930-2_5
Weng, F., Chen, G., Zan, Y., et al. (2026) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ae97b0
“These include the XCAT phantom (Segars et al 2010) for assessing MPI performance, an OASIS-derived brain perfusion phantom (LaMontagne et al 2019, Auer et al 2021) for evaluating cortical and gray-white matter boundary recovery, and a dedicated DaTscan phantom (Auer et al 2021) for assessing the…”
Deepika, P., Sunkari, S., Upadhyayula, S. K., Initiative, T. A. D. N., & Sundaresan, V. (2026) · medRxiv · doi:10.64898/2026.08.07.26359964
The Privacy–Utility Trade-Off in Brain MRI Synthesis: A Comparative Framework for Generative Models
Ninh, D. T., Kien, N. D., Huynh, T. N., Anh, L. Q., Dinh, T. H., Manh, B. D., Anh, N. H., D’haeseleer, M., Schependom, J. V., Denissen, S., et al. (2026) · Communications in computer and information science · doi:10.1007/978-981-92-2600-9_31
Saadawy, M., & Saadawy, E. (2026) · Discover Neuroscience · doi:10.1186/s13064-026-00315-z
“Data were obtained from two independent, publicly available neuroimaging cohorts: the Alzheimer’s Disease Neuroimaging Initiative (ADNI; adni.loni.usc.edu) [11] and the Open Access Series of Imaging Studies 3 (OASIS-3; oasis-brains.org) [12].”
NeuroMorph: A Unified Morphological Reference Space for Cross-Disease Brain ProfilingPreprint
Dalby, C., Dibble, A. J., Benini, S., Ferrari, D., Lyall, D. M., Harvey, M., Quinn, T., Muckli, L. F., Fracasso, A., Svanera, M., et al. (2026) · medRxiv · doi:10.64898/2026.08.13.26359403 · Cited by 2
Jahangir, R., اصل, ع. ک., Arabi, H., & Zaidi, H. (2026) · NeuroImage · doi:10.1016/j.neuroimage.2026.122178
A vascular component to the selective vulnerability of the medial temporal lobe to tau pathology
Doyon, V., Janelle, F., Sean, M., Vanderweyen, D., Cote, S. R., Huck, J., Tétreault, P., Lepage, J., Gutierrez, J., Bocti, C., et al. (2026) · Brain Communications · doi:10.1093/braincomms/fcag301
“37 Information about data acquisitions can be found at: https://sites.wustl.edu/oasisbrains/home/oasis-3/.”
A data-driven regional amyloid PET score predicts cognitive decline beyond CentiloidPreprint
Hirose, T., Akamatsu, W., Kato, T., & Initiative, A. D. N. (2026) · medRxiv · doi:10.64898/2026.08.26.26360248
Kamalov, F., & Ibrahim, A. K. (2026) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-026-02240-3 · Cited by 2
Volkmer, S., Fritze, S., Altinok, D. C. A., Brandt, G. A., Kocak, E., Wiegert, J., Berhe, O., Lin, Y., Tost, H., Meyer‐Lindenberg, A., et al. (2026) · Translational Psychiatry · doi:10.1038/s41398-026-04426-3
“The cohort includes 755 cognitively normal adults and 622 individuals spanning various stages of cognitive impairment, with ages ranging from 42-95 years [34].”
Advanced Brain Aging Associated With Migraine and Posttraumatic Headache: A Deep Learning Study
Shah, J., Dumkrieger, G. M., Chong, C. D., Schwedt, T. J., & Wu, T. (2026) · Neurology Open Access · doi:10.1212/wn9.0000000000000158
Schuck, P. N., & da Silva, A. M. M. (2026) · World Congress on Medical Physics and Biomedical Engineering, September 7 - 12, 2009, Munich, Germany · doi:10.1007/978-3-032-20287-1_8
Tsai, C. M., Lee, J.-D., & Initiative, A. D. N. (2026) · Electronics · doi:10.3390/electronics15184071
Fin, S., & Moayedikia, A. (2026) · medRxiv · doi:10.64898/2026.09.05.26362318
Distribution-Aware Feature Selection for Post-hoc Out-of-Distribution Detection
Gutbrod, M., Rauber, D., & Palm, C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37490-5_1
A Structure-Consistency-Constrained Method for Stroke MRI Lesion Inpainting
Zhao, S., Peng, B., Dai, Y., & Yu, W. (2026) · International Conference on Health Big Data · doi:10.1145/3833450.3833511
“0[13] and OASIS-3 [14].”
Longitudinal Reliability of Cross-Sectional Normative Models Under Measurement Error
Ji, W., Long, J., Guan, Q., Li, X., Jiang, S., Li, J., & Wang, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38172-9_39
Zhang, Y., Li, H., & Fan, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38172-9_31
Hassan, S., Salem, M., Elsayed, A., Oullier, O., & Yaqub, M. K. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38189-7_9
PRA-PoE: Robust Multimodal Alzheimer’s Disease Classification Under Arbitrary Modality Missingness
Yang, G., Du, Y., Hou, W., Niu, Q., Wang, S., & Initiative, A. D. N. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38098-2_51 · Cited by 2
Contrast-Invariant Reference-Based Slice Thickness Estimation in MRI via Spectral Matching
Remedios, S. W., Wei, S., Carass, A., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38260-3_8
Ding, S., Wu, F., Kong, Y., & Shu, H. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38069-2_40
Discovering Heterogeneous Neurodegenerative Disease Patterns From MRI Data for Improved Prediction
Zhang, Y., Li, H., & Fan, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38239-9_16
Anatomy- and Site-Guided 3D Patch Diffusion for Robust MRI Harmonization
Hache, B., Roca, V., Kuchcinski, G., Manouvriez, D., & Lopes, R. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38072-2_2
Resolution-Conditioned Representation Learning for Unified Brain MRI Modeling
Dong, K., Hu, X., Huang, S., Xu, Y., Lian, Z., Li, X., Yang, L., Fan, Y., Shen, D. G., & Sun, K. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38239-9_45
SIGMA-VAE: Sketched Isotropic Gaussian Multi-view Autoencoder for Normative Modeling
Zabihi, M., Dai, S., Schmidt, A., Preller, K. H., Watrous, A., Bornemann, K., Cole, J. H., Marquand, A. F., Yang, Y., & Wolfers, T. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38239-9_51
Brain-Specialized Self-supervised Learning: Neuroanatomy-Aware Spatial Priors for 3D Brain MRI
Kim, K. H., Kwon, M., Kim, Y.-M., Kim, M. K., Shin, M.-H., Chung, I., Koh, S. B., Kim, H. C., & Lee, J.-M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38069-2_5
Mehmood, A., Mehmood, F., & Kim, J. (2026) · Mathematics · doi:10.3390/math14193542
Navale, G. J., Wagh, K. S., Khan, A. A., et al. (2026) · doi:10.1109/ignite69592.2026.11706283
“Experiments utilize ADNI and OASIS datasets containing MRI, cognitive scores, and demographic information [3], [4], [25], [26].”
Ramya, C. N., & Keshavamurthy, K. (2026) · International Research Journal of Multidisciplinary Technovation · doi:10.54392/irjmt26519
“The OASIS-3 database [15] is a publicly available longitudinal neuroimaging cohort which was specifically chosen to further the study of aging, cognitive impairment, and Alzheimer disease (AD).”
Liang, Z., Guo, X., Xu, W., Ibrahim, Y., Voets, N., Pretorius, P. M., Noble, J. A., & Kamnitsas, K. (2025) · arXiv
Metadata
- Authors
- LaMontagne, P. J., Benzinger, T. L., Morris, J. C., Keefe, S., Hornbeck, R., Xiong, C., Grant, E., Hassenstab, J., Moulder, K., Vlassenko, A. G., Raichle, M. E., Cruchaga, C., & Marcus, D.
- Affiliations
- Washington University in St. Louis (US)
- Corresponding author
- LaMontagne, Pamela J.
- ORCID
- Venue
- medRxiv
- Publisher
- openRxiv
- Published
- 2019-12-15
- OpenAlex
- W2995808388
- Cited by
- 779
- References
- 62
- Citations by year
- 2026: 98 · 2025: 158 · 2024: 155 · 2023: 142 · 2022: 125 · 2021: 73 · 2020: 27
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Licence
- cc-by-nd
- Locations
- medRxiv (acceptedVersion, cc-by-nd)
- Funders
- Eli Lilly and Company
- National Institutes of Health (P30 NS09857781, UL1TR000448, P50 AG00561)
- Avid Radiopharmaceuticals
- Charles F. and Joanne Knight Alzheimer Disease Research Center, Washington University in St. Louis (P01AG003991)
- National Institute on Aging (R01AG043434, P01AG003991, P01AG026276)
- National Institute of Biomedical Imaging and Bioengineering (R01EB009352)
- Topics
- Health Sciences › Medicine › Psychiatry and Mental health › Dementia and Cognitive Impairment Research
- Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Alzheimer's disease research and treatments
- Life Sciences › Neuroscience › Cognitive Neuroscience › Functional Brain Connectivity Studies
- Keywords
- OASIS-3, Alzheimer's disease, normal aging, cognitive decline, longitudinal neuroimaging, structural MRI, functional MRI, positron emission tomography, amyloid imaging, glucose metabolism, 3D image segmentation, dementia, apolipoprotein E, cognitive outcomes, healthy aging
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.