{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T12:20:39Z","timestamp":1780489239864,"version":"3.54.1"},"posted":{"date-parts":[[2025,12,27]]},"group-title":"Neuroscience","reference-count":70,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,12,27]],"date-time":"2025-12-27T00:00:00Z","timestamp":1766793600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/publicdomain\/zero\/1.0\/"}],"funder":[{"DOI":"10.13039\/100000957","name":"Alzheimer\u2019s Association","doi-asserted-by":"crossref","award":["23AARF-1026883"],"award-info":[{"award-number":["23AARF-1026883"]}],"id":[{"id":"10.13039\/100000957","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,5,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Tau PET is a powerful tool to assess tau pathology in vivo; however, in comparison to MRI, its development is more recent, is rarely available at scale, and substantially more difficult to acquire. Here, we present Covariate-Modulated Attention UNet (CoMA-UNet) to synthesize subject-specific 3D tau PET from T1 MRI while incorporating in the synthesis procedure readily available covariates. Across six external validation datasets, CoMA-UNet reproduced regional patterns of tau PET uptake showing strong agreement with true PET that was generalizable across tracers. Next, we submitted the synthetic tau PET to a series of downstream clinically relevant tasks. First, MMSE associations between the synthetic tau PET were statistically indistinguishable from true PET. Second, the synthetic tau PET achieved out-of-sample diagnostic classification of dementia with an AUROC=0.99. Third, in two independent autopsy cohorts, voxel wise synthetic tau PET images closely followed neuropathologically defined Braak-stages. Fourth, when fine-tuned on a small subset of atypical presentations of AD, CoMA-UNet preserved phenotype-relevant tau topographies and recovered clinically meaningful structure across atypical clinical phenotypes. These findings demonstrate that the novel CoMA-UNet MRI-based synthesis augmented with covariate information can approximate tau PET with sufficient accuracy for downstream scientific and clinical applications.<\/jats:p>","DOI":"10.64898\/2025.12.27.696668","type":"posted-content","created":{"date-parts":[[2025,12,28]],"date-time":"2025-12-28T02:10:11Z","timestamp":1766887811000},"source":"Crossref","is-referenced-by-count":0,"title":["Neuropathologically validated MRI to tau PET synthesis via Covariate-modulated attention networks"],"prefix":"10.64898","author":[{"given":"Marcell","family":"Borhi","sequence":"first","affiliation":[{"name":"Division of Biostatistics, School of Public Health, University of California, Berkeley, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rita Qiuran","family":"Lyu","sequence":"additional","affiliation":[{"name":"Division of Biostatistics, School of Public Health, University of California, Berkeley, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"William J","family":"Jagust","sequence":"additional","affiliation":[{"name":"Department of Neuroscience, University of California, Berkeley, California, USA, 94720"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingshen","family":"Wang","sequence":"additional","affiliation":[{"name":"Division of Biostatistics, School of Public Health, University of California, Berkeley, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Joseph","family":"Giorgio","sequence":"additional","affiliation":[{"name":"Division of Biostatistics, School of Public Health, University of California, Berkeley, California, USA"},{"name":"Department of Neuroscience, University of California, Berkeley, California, USA, 94720"},{"name":"School of Psychological Sciences, College of Engineering, Science and the Environment, University of Newcastle, Newcastle, New South Wales, Australia, 2308"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"name":"Alzheimer\u2019s Disease Neuroimaging Initiative","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026060304550819000_2025.12.27.696668v2.1","doi-asserted-by":"crossref","unstructured":"2025 Alzheimer\u2019s disease facts and figures. Alzheimer\u2019s & Dementia 21, (2025).","DOI":"10.1002\/alz.70235"},{"key":"2026060304550819000_2025.12.27.696668v2.2","doi-asserted-by":"crossref","unstructured":"Safiri, S. et al. Alzheimer\u2019s disease: a comprehensive review of epidemiology, risk factors, symptoms diagnosis, management, caregiving, advanced treatments and associated challenges. Front. Med. (Lausanne). 11, (2024).","DOI":"10.3389\/fmed.2024.1474043"},{"key":"2026060304550819000_2025.12.27.696668v2.3","doi-asserted-by":"crossref","first-page":"5143","DOI":"10.1002\/alz.13859","article-title":"Revised criteria for diagnosis and staging of Alzheimer\u2019s disease: Alzheimer\u2019s Association Workgroup","volume":"20","year":"2024","journal-title":"Alzheimer\u2019s & Dementia"},{"key":"2026060304550819000_2025.12.27.696668v2.4","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1007\/s00401-018-1897-9","article-title":"Estimation of amyloid distribution by [18F]flutemetamol PET predicts the neuropathological phase of amyloid \u03b2-protein deposition","volume":"136","year":"2018","journal-title":"Acta Neuropathol"},{"key":"2026060304550819000_2025.12.27.696668v2.5","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.122.265200"},{"key":"2026060304550819000_2025.12.27.696668v2.6","doi-asserted-by":"publisher","DOI":"10.1038\/S41582-024-00942-2"},{"key":"2026060304550819000_2025.12.27.696668v2.7","doi-asserted-by":"publisher","DOI":"10.1038\/s41583-018-0067-3"},{"key":"2026060304550819000_2025.12.27.696668v2.8","doi-asserted-by":"publisher","DOI":"10.1007\/BF00308809"},{"key":"2026060304550819000_2025.12.27.696668v2.9","first-page":"526","article-title":"Biomarker modeling of Alzheimer\u2019s disease using PET-based Braak staging. Nat","volume":"2","year":"2022","journal-title":"Aging"},{"key":"2026060304550819000_2025.12.27.696668v2.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2016.01.028"},{"key":"2026060304550819000_2025.12.27.696668v2.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2017.05.058"},{"key":"2026060304550819000_2025.12.27.696668v2.12","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.118.209650"},{"key":"2026060304550819000_2025.12.27.696668v2.13","doi-asserted-by":"crossref","first-page":"255","DOI":"10.1001\/jamaneurol.2023.5319","article-title":"Diagnostic Accuracy of a Plasma Phosphorylated Tau 217 Immunoassay for Alzheimer Disease Pathology","volume":"81","year":"2024","journal-title":"JAMA Neurol"},{"key":"2026060304550819000_2025.12.27.696668v2.14","doi-asserted-by":"publisher","DOI":"10.1038\/s41380-021-01263-2"},{"key":"2026060304550819000_2025.12.27.696668v2.15","doi-asserted-by":"crossref","unstructured":"Karlsson, L. et al. Machine learning prediction of tau-PET in Alzheimer\u2019s disease using plasma, MRI, and clinical data. Alzheimer\u2019s and Dementia 21, (2025).","DOI":"10.1002\/alz.70433"},{"key":"2026060304550819000_2025.12.27.696668v2.16","doi-asserted-by":"crossref","unstructured":"Lew, C. O. , Zhou, L. , Mazurowski, M. A. , Murali Doraiswamy, P. & Petrella, J. R . MRI-based Deep Learning Assessment of Amyloid, Tau, and Neurodegeneration Biomarker Status across the Alzheimer Disease Spectrum. Radiology 309, (2023).","DOI":"10.1148\/radiol.222441"},{"key":"2026060304550819000_2025.12.27.696668v2.17","doi-asserted-by":"crossref","unstructured":"Giorgio, J. et al. A robust and interpretable machine learning approach using multimodal biological data to predict future pathological tau accumulation. Nat. Commun. 13, (2022).","DOI":"10.1038\/s41467-022-28795-7"},{"key":"2026060304550819000_2025.12.27.696668v2.18","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awad346"},{"key":"2026060304550819000_2025.12.27.696668v2.19","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa2305032"},{"key":"2026060304550819000_2025.12.27.696668v2.20","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2026060304550819000_2025.12.27.696668v2.21","unstructured":"Williams, C. et al. A Unified Framework for U-Net Design and Analysis. http:\/\/arxiv.org\/abs\/2305.19638 (2023)."},{"key":"2026060304550819000_2025.12.27.696668v2.22","unstructured":"Oktay, O. et al. Attention U-Net: Learning Where to Look for the Pancreas. http:\/\/arxiv.org\/abs\/1804.03999 (2018)."},{"key":"2026060304550819000_2025.12.27.696668v2.23","doi-asserted-by":"crossref","unstructured":"Hatamizadeh, A. et al. UNETR: Transformers for 3D Medical Image Segmentation. http:\/\/arxiv.org\/abs\/2103.10504 (2021).","DOI":"10.1109\/WACV51458.2022.00181"},{"key":"2026060304550819000_2025.12.27.696668v2.24","doi-asserted-by":"crossref","unstructured":"Cai, Y. et al. Swin Unet3D: a three-dimensional medical image segmentation network combining vision transformer and convolution. BMC Med. Inform. Decis. Mak. 23, (2023).","DOI":"10.1186\/s12911-023-02129-z"},{"key":"2026060304550819000_2025.12.27.696668v2.25","unstructured":"Yang, Y.-Q. et al. Swin3D: A Pretrained Transformer Backbone for 3D Indoor Scene Understanding. http:\/\/arxiv.org\/abs\/2304.06906 (2023)."},{"key":"2026060304550819000_2025.12.27.696668v2.26","unstructured":"Lai, X. , et al. Stratified Transformer for 3D Point Cloud Segmentation. arXiv preprint arXiv:2203.14508 https:\/\/arxiv.org\/abs\/2203.14508 (2022)."},{"key":"2026060304550819000_2025.12.27.696668v2.27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.632"},{"key":"2026060304550819000_2025.12.27.696668v2.28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.244"},{"key":"2026060304550819000_2025.12.27.696668v2.29","doi-asserted-by":"crossref","unstructured":"Zhang, L. , Chen, X. , Zhang, J. , Dong, R. & Ma, K. Contrastive Deep Supervision. http:\/\/arxiv.org\/abs\/2207.05306 (2022).","DOI":"10.1007\/978-3-031-19809-0_1"},{"key":"2026060304550819000_2025.12.27.696668v2.30","doi-asserted-by":"publisher","DOI":"10.1101\/2025.04.23.25326302"},{"key":"2026060304550819000_2025.12.27.696668v2.31","doi-asserted-by":"crossref","unstructured":"Zheng, X. et al. Generating synthetic brain PET images of synaptic density based on MR T1 images using deep learning. EJNMMI Phys. 12, 30 (2025).","DOI":"10.1186\/s40658-025-00744-5"},{"key":"2026060304550819000_2025.12.27.696668v2.32","unstructured":"Dosovitskiy, A. et al. An Image is Worth 16\u00d716 Words: Transformers for Image Recognition at Scale. http:\/\/arxiv.org\/abs\/2010.11929 (2020)."},{"key":"2026060304550819000_2025.12.27.696668v2.33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01074"},{"key":"2026060304550819000_2025.12.27.696668v2.34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2006.11239"},{"key":"2026060304550819000_2025.12.27.696668v2.35","unstructured":"Saharia, C. , et al. Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding. arXiv preprint arXiv:2205.11487 https:\/\/arxiv.org\/abs\/2205.11487 (2022)."},{"key":"2026060304550819000_2025.12.27.696668v2.36","doi-asserted-by":"crossref","unstructured":"Lee, J. et al. Synthesizing images of tau pathology from cross-modal neuroimaging using deep learning. Brain 147, (2024).","DOI":"10.1093\/brain\/awad346"},{"key":"2026060304550819000_2025.12.27.696668v2.37","doi-asserted-by":"publisher","DOI":"10.1007\/s13311-021-01030-9"},{"key":"2026060304550819000_2025.12.27.696668v2.38","doi-asserted-by":"publisher","DOI":"10.1212\/CON.0000000000001133"},{"key":"2026060304550819000_2025.12.27.696668v2.39","doi-asserted-by":"crossref","unstructured":"Guedj, E. et al. EANM procedure guidelines for brain PET imaging using [18F]FDG, version 3. Eur. J. Nucl. Med. Mol. Imaging 49, (2022).","DOI":"10.1007\/s00259-022-05755-3"},{"key":"2026060304550819000_2025.12.27.696668v2.40","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awaa276"},{"key":"2026060304550819000_2025.12.27.696668v2.41","unstructured":"Leuzy, A. et al. Comparison of Group-Level and Individualized Brain Regions for Measuring Change in Longitudinal Tau Positron Emission Tomography in Alzheimer Disease. JAMA Neurol. 80, 614 (2023)."},{"key":"2026060304550819000_2025.12.27.696668v2.42","doi-asserted-by":"crossref","first-page":"587","DOI":"10.1038\/s41380-024-02672-9","article-title":"Tau PET positivity predicts clinically relevant cognitive decline driven by Alzheimer\u2019s disease compared to comorbid cases; proof of concept in the ADNI study","volume":"30","year":"2025","journal-title":"Mol. Psychiatry"},{"key":"2026060304550819000_2025.12.27.696668v2.43","doi-asserted-by":"crossref","unstructured":"Ossenkoppele, R. et al. Associations between tau, A\u03b2, and cortical thickness with cognition in Alzheimer disease. Neurology 92, (2019).","DOI":"10.1212\/WNL.0000000000006875"},{"key":"2026060304550819000_2025.12.27.696668v2.44","doi-asserted-by":"crossref","unstructured":"Groot, C. et al. Tau Positron Emission Tomography for Predicting Dementia in Individuals With Mild Cognitive Impairment. JAMA Neurol. 81, 845 (2024).","DOI":"10.1001\/jamaneurol.2024.1612"},{"key":"2026060304550819000_2025.12.27.696668v2.45","doi-asserted-by":"crossref","first-page":"e70435","DOI":"10.1002\/alz.70435","article-title":"Longitudinal tau aggregation, atrophy, and cognitive decline in Alzheimer\u2019s disease","volume":"21","year":"2025","journal-title":"Alzheimers Dement"},{"key":"2026060304550819000_2025.12.27.696668v2.46","doi-asserted-by":"crossref","unstructured":"St-Onge, F. , Chapleau, M. , Breitner, J. C. S. , Villeneuve, S. & Pichet Binette, A . Tau accumulation and its spatial progression across the Alzheimer\u2019s disease spectrum. Brain Commun. 6, (2023).","DOI":"10.1093\/braincomms\/fcae031"},{"key":"2026060304550819000_2025.12.27.696668v2.47","doi-asserted-by":"crossref","first-page":"802","DOI":"10.14283\/jpad.2024.122","article-title":"Amyloid and Tau Prediction of Cognitive and Functional Decline in Unimpaired Older Individuals: Longitudinal Data from the A4 and LEARN Studies","volume":"11","year":"2024","journal-title":"J. Prev. Alzheimers Dis"},{"key":"2026060304550819000_2025.12.27.696668v2.48","doi-asserted-by":"crossref","unstructured":"Sims, J. R. et al. Donanemab in Early Symptomatic Alzheimer Disease. JAMA 330, 512 (2023).","DOI":"10.1001\/jama.2023.21109"},{"key":"2026060304550819000_2025.12.27.696668v2.49","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awab114"},{"key":"2026060304550819000_2025.12.27.696668v2.50","doi-asserted-by":"crossref","unstructured":"DeTure, M. A. & Dickson, D. W . The neuropathological diagnosis of Alzheimer\u2019s disease. Mol. Neurodegener. 14, 32 (2019).","DOI":"10.1186\/s13024-019-0333-5"},{"key":"2026060304550819000_2025.12.27.696668v2.51","doi-asserted-by":"publisher","DOI":"10.1093\/brain\/awaa327"},{"key":"2026060304550819000_2025.12.27.696668v2.52","doi-asserted-by":"crossref","unstructured":"Phillips, J. S. , et al. Rates of longitudinal change in 18F-flortaucipir PET vary by brain region, cognitive impairment, and age in atypical Alzheimer\u2019s disease. Alzheimer\u2019s and Dementia 18, (2022).","DOI":"10.1002\/alz.12456"},{"key":"2026060304550819000_2025.12.27.696668v2.53","doi-asserted-by":"publisher","DOI":"10.1016\/S1474-4422(20)30440-3"},{"key":"2026060304550819000_2025.12.27.696668v2.54","doi-asserted-by":"crossref","unstructured":"He, K. , Zhang, X. , Ren, S. & Sun, J. Deep Residual Learning for Image Recognition. (2015).","DOI":"10.1109\/CVPR.2016.90"},{"key":"2026060304550819000_2025.12.27.696668v2.55","unstructured":"Tan, M. & Le, Q. V. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. (2020)."},{"key":"2026060304550819000_2025.12.27.696668v2.56","unstructured":"Dehghani, M. , et al. Scaling Vision Transformers to 22 Billion Parameters. (2023)."},{"key":"2026060304550819000_2025.12.27.696668v2.57","doi-asserted-by":"crossref","unstructured":"Li, X. L. & Liang, P. Prefix-Tuning: Optimizing Continuous Prompts for Generation. (2021).","DOI":"10.18653\/v1\/2021.acl-long.353"},{"key":"2026060304550819000_2025.12.27.696668v2.58","doi-asserted-by":"crossref","unstructured":"Jia, M. , et al. Visual Prompt Tuning. (2022).","DOI":"10.1007\/978-3-031-19827-4_41"},{"key":"2026060304550819000_2025.12.27.696668v2.59","doi-asserted-by":"publisher","DOI":"10.1038\/s41380-021-01249-0)"},{"key":"2026060304550819000_2025.12.27.696668v2.60","doi-asserted-by":"publisher","DOI":"10.1016\/S1474-4422(20)30440-3"},{"key":"2026060304550819000_2025.12.27.696668v2.61","doi-asserted-by":"crossref","first-page":"1235","DOI":"10.1002\/alz.12456","article-title":"Rates of longitudinal change in 18 F-flortaucipir PET vary by brain region, cognitive impairment, and age in atypical Alzheimer\u2019s disease","volume":"18","year":"2022","journal-title":"Alzheimer\u2019s & Dementia"},{"key":"2026060304550819000_2025.12.27.696668v2.62","doi-asserted-by":"crossref","first-page":"5143","DOI":"10.1002\/alz.13859","article-title":"Revised criteria for diagnosis and staging of Alzheimer\u2019s disease: Alzheimer\u2019s Association Workgroup","volume":"20","year":"2024","journal-title":"Alzheimer\u2019s & Dementia"},{"key":"2026060304550819000_2025.12.27.696668v2.63","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.121.263196"},{"key":"2026060304550819000_2025.12.27.696668v2.64","doi-asserted-by":"publisher","DOI":"10.2967\/jnumed.120.254961"},{"key":"2026060304550819000_2025.12.27.696668v2.65","doi-asserted-by":"publisher","DOI":"10.1007\/s40336-018-0290-y"},{"key":"2026060304550819000_2025.12.27.696668v2.66","unstructured":"Yang, B. , Bender, G. , Le, Q. V. & Ngiam, J . CondConv: Conditionally Parameterized Convolutions for Efficient Inference. http:\/\/arxiv.org\/abs\/1904.04971 (2019)."},{"key":"2026060304550819000_2025.12.27.696668v2.67","unstructured":"Prokhorenkova, L. , Gusev, G. , Vorobev, A. , Dorogush, A. V. & Gulin, A. CatBoost: Unbiased boosting with categorical features. in Advances in Neural Information Processing Systems (NeurIPS) 6638\u20136648 (2018)."},{"key":"2026060304550819000_2025.12.27.696668v2.68","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2012.01.021"},{"key":"2026060304550819000_2025.12.27.696668v2.69","doi-asserted-by":"crossref","first-page":"e14378","DOI":"10.1002\/alz.14378","article-title":"Positron emission tomography harmonization in the Alzheimer\u2019s Disease Neuroimaging Initiative: A scalable and rigorous approach to multisite amyloid and tau quantification","volume":"21","year":"2025","journal-title":"Alzheimers Dement"},{"key":"2026060304550819000_2025.12.27.696668v2.70","unstructured":"Paszke, A. , et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library. (2019)."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2025.12.27.696668","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T11:55:33Z","timestamp":1780487733000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2025.12.27.696668"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,27]]},"references-count":70,"URL":"https:\/\/doi.org\/10.64898\/2025.12.27.696668","relation":{},"subject":[],"published":{"date-parts":[[2025,12,27]]},"subtype":"preprint"}}