{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T05:59:49Z","timestamp":1790315989196,"version":"4.1.0"},"reference-count":30,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T00:00:00Z","timestamp":1790294400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Aging Neurosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Early differentiation between Parkinson\u2019s disease (PD) and multiple system atrophy-parkinsonian type (MSA-P) remains clinically challenging because of overlapping motor symptoms and subtle imaging abnormalities.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We developed a fully automated multimodal PET\/MRI framework for early PD\u2013MSA-P differentiation that integrates atlas-guided subject-specific label generation, nnU-Net segmentation of the caudate nuclei and putamina, multimodal image fusion, and Vision Transformer classification with multi-slice voting (ViT+MSV). A total of 155 early-stage patients, including 121 with PD and 34 with MSA-P, were evaluated using patient-level stratified 4-fold cross-validation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The segmentation model achieved an overall Dice score of 0.853 (95% CI, 0.830\u20130.877) and an IoU of 0.747 (95% CI, 0.725\u20130.770). ViT+MSV achieved a cross-validated AUC of 0.947 (95% CI, 0.905\u20130.981), an accuracy of 0.929 (95% CI, 0.890\u20130.968), a sensitivity of 0.942 (95% CI, 0.900\u20130.982), and a specificity of 0.882 (95% CI, 0.765\u20130.974). The combination of [\n                      <jats:sup>18<\/jats:sup>\n                      F]fluorodeoxyglucose (FDG) PET, [\n                      <jats:sup>11<\/jats:sup>\n                      C]CFT dopamine-transporter PET, and T2-weighted imaging (T2WI) showed the best overall classification performance. Grad-CAM visualizations showed that class-discriminative gradient-based saliency was mainly concentrated on the bilateral putamen.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These findings suggest that anatomically guided multimodal PET\/MRI analysis with transformer-based classification may support early PD\u2013MSA-P differentiation while providing spatially interpretable evidence for model decisions.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnagi.2026.1928959","type":"journal-article","created":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T05:31:46Z","timestamp":1790314306000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Differentiating early Parkinson\u2019s disease from multiple system atrophy on multimodal PET\/MRI using atlas-guided segmentation and vision transformer classification"],"prefix":"10.3389","volume":"18","author":[{"given":"Chao","family":"Cong","sequence":"first","affiliation":[{"name":"School of Electrical and Electronic Engineering, Chongqing University of Technology","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuewei","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Electrical and Electronic Engineering, Chongqing University of Technology","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shams Tabrez","family":"Rifat","sequence":"additional","affiliation":[{"name":"School of Electrical and Electronic Engineering, Chongqing University of Technology","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongyun","family":"Tan","sequence":"additional","affiliation":[{"name":"School of Electrical and Electronic Engineering, Chongqing University of Technology","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinju","family":"Sun","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Daping Hospital, Army Medical University","place":["Chongqing, China"]},{"name":"Chongqing Clinical Research Center for Imaging and Nuclear Medicine","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenping","family":"Peng","sequence":"additional","affiliation":[{"name":"School of Electrical and Electronic Engineering, Chongqing University of Technology","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiao","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Daping Hospital, Army Medical University","place":["Chongqing, China"]},{"name":"Chongqing Clinical Research Center for Imaging and Nuclear Medicine","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,25]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"260","DOI":"10.1148\/radiol.2021203341","article-title":"Imaging the substantia nigra in Parkinson disease and other parkinsonian syndromes.","volume":"300","author":"Bae","year":"2021","journal-title":"Radiology"},{"key":"B2","doi-asserted-by":"publisher","first-page":"103859","DOI":"10.1016\/j.nicl.2025.103859","article-title":"Multi-Center 3D CNN for Parkinson\u2019s disease diagnosis and prognosis using clinical and T1-weighted MRI data.","volume":"48","author":"Basaia","year":"2025","journal-title":"Neuroimage Clin."},{"key":"B3","doi-asserted-by":"publisher","first-page":"103405","DOI":"10.1016\/j.nicl.2023.103405","article-title":"Explainable classification of Parkinson\u2019s disease using deep learning trained on a large multi-center database of T1-weighted MRI datasets.","volume":"38","author":"Camacho","year":"2023","journal-title":"Neuroimage Clin."},{"key":"B4","doi-asserted-by":"publisher","first-page":"55","DOI":"10.1186\/s13550-025-01245-3","article-title":"Deep learning for Parkinson\u2019s disease classification using multimodal and multi-sequences PET\/MR images.","volume":"15","author":"Chang","year":"2025","journal-title":"EJNMMI Res."},{"key":"B5","doi-asserted-by":"publisher","first-page":"110735","DOI":"10.1016\/j.ejrad.2023.110735","article-title":"Automatic classification of MSA subtypes using Whole-brain gray matter function and Structure-Based radiomics approach.","volume":"161","author":"Chen","year":"2023","journal-title":"Eur. J. Radiol."},{"key":"B6","doi-asserted-by":"publisher","first-page":"102997","DOI":"10.1016\/j.nicl.2022.102997","article-title":"Advanced brain aging in multiple system atrophy compared to Parkinson\u2019s disease.","volume":"34","author":"Chen","year":"2022","journal-title":"Neuroimage Clin."},{"key":"B7","doi-asserted-by":"publisher","first-page":"112272","DOI":"10.1016\/j.ejrad.2025.112272","article-title":"Explainable classification of Parkinson\u2019s disease with different motor subtypes by analyzing the synthetic MRI quantitative parameters of subcortical nuclei.","volume":"190","author":"Cheng","year":"2025","journal-title":"Eur. J. Radiol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"e2672","DOI":"10.1212\/WNL.0000000000201269","article-title":"Association between white matter networks and the pattern of striatal dopamine depletion in patients with Parkinson disease","volume":"99","author":"Chung","year":"2022","journal-title":"Neurology"},{"key":"B9","doi-asserted-by":"publisher","first-page":"107880","DOI":"10.1016\/j.cmpb.2023.107880","article-title":"Deep learning techniques in PET\/CT imaging: A comprehensive review from sinogram to image space.","volume":"243","author":"Fallahpoor","year":"2024","journal-title":"Comput. Methods Programs Biomed."},{"key":"B10","doi-asserted-by":"publisher","first-page":"e1529","DOI":"10.1212\/WNL.0000000000206772","article-title":"Cutaneous \u03b1-Synuclein signatures in patients with multiple system atrophy and Parkinson disease.","volume":"100","author":"Gibbons","year":"2023","journal-title":"Neurology"},{"key":"B11","doi-asserted-by":"publisher","first-page":"17562864251318865","DOI":"10.1177\/17562864251318865","article-title":"Differentiating idiopathic Parkinson\u2019s disease from multiple system atrophy-P using brain MRI-based radiomics: A multicenter study.","volume":"18","author":"Huang","year":"2025","journal-title":"Ther. Adv. Neurol. Disord."},{"key":"B12","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation.","volume":"18","author":"Isensee","year":"2021","journal-title":"Nat. Methods"},{"key":"B13","doi-asserted-by":"publisher","first-page":"891","DOI":"10.1007\/s00702-025-02921-3","article-title":"Neuroimaging in multiple system atrophy: Clinical implications and novel developments.","volume":"133","author":"Jost","year":"2025","journal-title":"J. Neural Transm."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1781","DOI":"10.1002\/mds.28632","article-title":"Dopamine receptors in Parkinson\u2019s disease: A meta-analysis of imaging studies","volume":"36","author":"Kaasinen","year":"2021","journal-title":"Mov. Disord."},{"key":"B15","doi-asserted-by":"publisher","first-page":"1252","DOI":"10.1016\/S1474-4422(24)00396-X","article-title":"Multiple system atrophy: Advances in pathophysiology, diagnosis, and treatment.","volume":"23","author":"Krismer","year":"2024","journal-title":"Lancet Neurol."},{"key":"B16","doi-asserted-by":"publisher","first-page":"1962","DOI":"10.1007\/s00259-025-07478-7","article-title":"Enhancing interpretability of AI with radiomics-based deep neural network: Proof of concept in the classification of Parkinsonian syndromes with 18F-FDG PET imaging.","volume":"53","author":"Li","year":"2026","journal-title":"Eur. J. Nucl. Med. Mol. Imaging"},{"key":"B17","doi-asserted-by":"publisher","first-page":"103807","DOI":"10.1016\/j.nicl.2025.103807","article-title":"Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes.","volume":"47","author":"Li","year":"2025","journal-title":"Neuroimage Clin."},{"key":"B18","doi-asserted-by":"publisher","first-page":"645","DOI":"10.1038\/s41582-023-00874-3","article-title":"Linking the cerebellum to Parkinson disease: An update.","volume":"19","author":"Li","year":"2023","journal-title":"Nat. Rev. Neurol."},{"key":"B19","doi-asserted-by":"publisher","first-page":"106791","DOI":"10.1016\/j.compbiomed.2023.106791","article-title":"MedViT: A robust vision transformer for generalized medical image classification.","volume":"157","author":"Manzari","year":"2023","journal-title":"Comput. Biol. Med."},{"key":"B20","doi-asserted-by":"publisher","first-page":"2139","DOI":"10.1002\/mds.30300","article-title":"Deep learning to differentiate parkinsonian syndromes using multimodal magnetic resonance imaging: A proof-of-concept study.","volume":"40","author":"Mattia","year":"2025","journal-title":"Mov. Disord."},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1101\/2025.11.17.688918","article-title":"Explainable 3D Cnns link regional and network level disruption in early Parkinson\u2019s Mris to symptom progression.","author":"Moroze","year":"2025","journal-title":"bioRxiv"},{"key":"B22","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s41572-022-00382-6","article-title":"Multiple system atrophy.","volume":"8","author":"Poewe","year":"2022","journal-title":"Nat. Rev. Dis. Primers"},{"key":"B23","article-title":"Do Vision Transformers see Like Convolutional Neural Networks?","volume-title":"Proceedings of the 35th International Conference on Neural Information Processing System","author":"Raghu","year":"2021"},{"key":"B24","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2017.74","article-title":"Grad-Cam: Visual Explanations from Deep Networks via Gradient-Based Localization","volume-title":"In proceedings of the IEEE International Conference on Computer Vision (ICCV)","author":"Selvaraju","year":"2017"},{"key":"B25","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1038\/s41531-025-01025-9","article-title":"Machine learning for Parkinson\u2019s disease: A comprehensive review of datasets, algorithms, and challenges.","volume":"11","author":"Shokrpour","year":"2025","journal-title":"NPJ Parkinsons Dis."},{"key":"B26","doi-asserted-by":"publisher","first-page":"6750","DOI":"10.1038\/s41467-023-42305-3","article-title":"The \u03b1-synuclein PET tracer [18F] ACI-12589 distinguishes multiple system atrophy from other neurodegenerative diseases.","volume":"14","author":"Smith","year":"2023","journal-title":"Nat. Commun."},{"key":"B27","doi-asserted-by":"publisher","first-page":"662","DOI":"10.1007\/s00330-023-10003-9","article-title":"Identification of Parkinson\u2019s disease and multiple system atrophy using multimodal PET\/MRI radiomics.","volume":"34","author":"Sun","year":"2024","journal-title":"Eur. Radiol."},{"key":"B28","doi-asserted-by":"publisher","first-page":"8218","DOI":"10.1007\/s00330-021-07979-7","article-title":"Radiomics on routine T1-weighted MRI can delineate Parkinson\u2019s disease from multiple system atrophy and progressive supranuclear palsy.","volume":"31","author":"Tupe-Waghmare","year":"2021","journal-title":"Eur. Radiol."},{"key":"B29","doi-asserted-by":"publisher","first-page":"495","DOI":"10.1001\/jamaneurol.2025.0112","article-title":"Automated imaging differentiation for parkinsonism.","volume":"82","author":"Vaillancourt","year":"2025","journal-title":"JAMA Neurol."},{"key":"B30","doi-asserted-by":"publisher","first-page":"5661","DOI":"10.1038\/s41467-024-49949-9","article-title":"Neuroimaging and fluid biomarkers in Parkinson\u2019s disease in an era of targeted interventions.","volume":"15","author":"Zarkali","year":"2024","journal-title":"Nat. Commun."}],"container-title":["Frontiers in Aging Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnagi.2026.1928959\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T05:31:47Z","timestamp":1790314307000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnagi.2026.1928959\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,25]]},"references-count":30,"alternative-id":["10.3389\/fnagi.2026.1928959"],"URL":"https:\/\/doi.org\/10.3389\/fnagi.2026.1928959","relation":{},"ISSN":["1663-4365"],"issn-type":[{"value":"1663-4365","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,25]]},"article-number":"1928959"}}