{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T10:52:50Z","timestamp":1781866370023,"version":"3.54.5"},"reference-count":24,"publisher":"Springer Science and Business Media LLC","issue":"7","license":[{"start":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T00:00:00Z","timestamp":1772150400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T00:00:00Z","timestamp":1772150400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100009394","name":"Faculty of Medicine, Munich University of Technology","doi-asserted-by":"publisher","award":["TUM KKF Clinician Scientist Program, project reference E-19 and H-27"],"award-info":[{"award-number":["TUM KKF Clinician Scientist Program, project reference E-19 and H-27"]}],"id":[{"id":"10.13039\/501100009394","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Assessing regional brain atrophy on 3D-T1w imaging is crucial for evaluating neurodegenerative disorders. However, high-quality volumetric imaging is not always available. Thus, AI-based algorithms were developed to generate \u201csynthetic\u201d 3D-T1w sequences using various clinical sequences as input. This retrospective study aims to investigate whether regional atrophy patterns are preserved in deep learning-based synthetic 3D-T1w sequences from different inputs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>The study included patients with Alzheimer\u2019s disease (AD), Frontotemporal dementia (FTD), and healthy controls (HC). Probands were scanned at 3\u2009T, and deep learning-based synthetic 3D-T1w images were generated from various inputs (3D FLAIR, 4\u2009mm axial FLAIR, 4\u2009mm coronal T2) using FreeSurfer-based SynthSR. Real 3D-T1w images served as the reference standard. Brain volumetry was performed using SynthSeg+ in FreeSurfer and the AssemblyNet-AD-FTD pipeline in VolBrain.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Global and regional volumes differed significantly between deep learning-based synthetized sequences and the reference standard 3D T1 for all subgroups and inputs (total white matter volume AD\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.0002, FTD\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.0001, HC\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.0116; total gray matter volume for AD, FTD, and HC\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.0001), except for hippocampal volumes. This systematic error in overestimating volumes affected automated disease probability prediction in FTD for all inputs (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.0001) and in HC for coronal T2 input (adj.\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.0054).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Deep learning-based synthetic 3D-T1w sequences introduce systematic errors in assessing global and regional brain volumetric measures, leading to overestimated volumes in controls and patients. Resulting synthetic images should be used cautiously, especially for volumetric analyses.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Question<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>It remains unclear whether deep learning-based synthetic 3D-T1w images from various inputs preserve regional atrophy patterns sufficiently to serve as input for automated volumetry<\/jats:italic>\n                      .\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Findings<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>Deep learning-based synthetic T1w images overestimate regional and global brain volumes in neurodegenerative diseases and controls, increasing with lower quality inputs<\/jats:italic>\n                      .\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Clinical relevance<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>Deep learning-based synthetic images should only be used with caution for volumetric evaluation of brain MRI scans. If possible, 3D scans should be used as input<\/jats:italic>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00330-025-12302-9","type":"journal-article","created":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T15:12:51Z","timestamp":1772205171000},"page":"6025-6037","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Deep learning-based synthetic brain MRI for the assessment of regional atrophy patterns in neurodegenerative diseases"],"prefix":"10.1007","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2259-9221","authenticated-orcid":false,"given":"Evamaria O.","family":"Riedel","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Severin","family":"Schramm","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fabian","family":"Bongratz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin J.","family":"Gruber","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dominik","family":"Sepp","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Karolin J.","family":"Paprottka","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Claus","family":"Zimmer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benedikt","family":"Wiestler","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dennis M.","family":"Hedderich","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,2,27]]},"reference":[{"key":"12302_CR1","doi-asserted-by":"crossref","unstructured":"Maresova P, Hruska J, Klimova B, Barakovic S, Krejcar O (2020) Activities of daily living and associated costs in the most widespread neurodegenerative diseases: a systematic review. Clin Interv Aging 15:1841\u20131862","DOI":"10.2147\/CIA.S264688"},{"key":"12302_CR2","doi-asserted-by":"publisher","first-page":"e105","DOI":"10.1016\/S2468-2667(21)00249-8","volume":"7","author":"E Nichols","year":"2022","unstructured":"Nichols E, Steinmetz JD, Vollset SE et al (2022) Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health 7:e105\u2013e125","journal-title":"Lancet Public Health"},{"key":"12302_CR3","doi-asserted-by":"publisher","first-page":"1761","DOI":"10.1007\/s00330-024-11047-1","volume":"35","author":"R Graf","year":"2025","unstructured":"Graf R, Platzek PS, Riedel EO et al (2025) Generating synthetic high-resolution spinal STIR and T1w images from T2w FSE and low-resolution axial Dixon. Eur Radiol 35:1761\u20131771","journal-title":"Eur Radiol"},{"key":"12302_CR4","doi-asserted-by":"publisher","first-page":"5882","DOI":"10.1007\/s00330-023-09512-4","volume":"33","author":"S Schlaeger","year":"2023","unstructured":"Schlaeger S, Drummer K, El Husseini M et al (2023) Synthetic T2-weighted fat sat based on a generative adversarial network shows potential for scan time reduction in spine imaging in a multicenter test dataset. Eur Radiol 33:5882\u20135893","journal-title":"Eur Radiol"},{"key":"12302_CR5","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.add3607","volume":"9","author":"JE Iglesias","year":"2023","unstructured":"Iglesias JE, Billot B, Balbastre Y et al (2023) SynthSR: a public AI tool to turn heterogeneous clinical brain scans into high-resolution T1-weighted images for 3D morphometry. Sci Adv 9:eadd3607","journal-title":"Sci Adv"},{"key":"12302_CR6","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1016\/j.jneumeth.2016.03.001","volume":"264","author":"X Li","year":"2016","unstructured":"Li X, Morgan PS, Ashburner J, Smith J, Rorden C (2016) The first step for neuroimaging data analysis: DICOM to NIfTI conversion. J Neurosci Methods 264:47\u201356","journal-title":"J Neurosci Methods"},{"key":"12302_CR7","doi-asserted-by":"publisher","unstructured":"Gulban OF, Nielson D, J Lee (2022) poldracklab\/pydeface: PyDeface v2.0.2. Zenodo https:\/\/doi.org\/10.5281\/zenodo.6856482","DOI":"10.5281\/zenodo.6856482"},{"key":"12302_CR8","doi-asserted-by":"publisher","first-page":"54","DOI":"10.1186\/s13244-022-01195-7","volume":"13","author":"C Rubbert","year":"2022","unstructured":"Rubbert C, Wolf L, Turowski B et al (2022) Impact of defacing on automated brain atrophy estimation. Insights Imaging 13:54","journal-title":"Insights Imaging"},{"key":"12302_CR9","doi-asserted-by":"publisher","first-page":"774","DOI":"10.1016\/j.neuroimage.2012.01.021","volume":"62","author":"B Fischl","year":"2012","unstructured":"Fischl B (2012) FreeSurfer. Neuroimage 62:774\u2013781","journal-title":"Neuroimage"},{"key":"12302_CR10","doi-asserted-by":"publisher","first-page":"30","DOI":"10.3389\/fninf.2016.00030","volume":"10","author":"JV Manj\u00f3n","year":"2016","unstructured":"Manj\u00f3n JV, Coup\u00e9 P (2016) volBrain: an online MRI brain volumetry system. Front Neuroinform 10:30","journal-title":"Front Neuroinform"},{"key":"12302_CR11","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2020.117026","volume":"219","author":"P Coup\u00e9","year":"2020","unstructured":"Coup\u00e9 P, Mansencal B, Cl\u00e9ment M et al (2020) AssemblyNet: a large ensemble of CNNs for 3D whole brain MRI segmentation. Neuroimage 219:117026","journal-title":"Neuroimage"},{"key":"12302_CR12","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2023.102636","volume":"144","author":"HD Nguyen","year":"2023","unstructured":"Nguyen HD, Cl\u00e9ment M, Planche V, Mansencal B, Coup\u00e9 P (2023) Deep grading for MRI-based differential diagnosis of Alzheimer\u2019s disease and frontotemporal dementia. Artif Intell Med 144:102636","journal-title":"Artif Intell Med"},{"key":"12302_CR13","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2216399120","volume":"120","author":"B Billot","year":"2023","unstructured":"Billot B, Magdamo C, Cheng Y, Arnold SE, Das S, Iglesias JE (2023) Robust machine learning segmentation for large-scale analysis of heterogeneous clinical brain MRI datasets. Proc Natl Acad Sci USA 120:e2216399120","journal-title":"Proc Natl Acad Sci USA"},{"key":"12302_CR14","unstructured":"Team RC (2022) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna"},{"key":"12302_CR15","doi-asserted-by":"publisher","first-page":"855","DOI":"10.4329\/wjr.v6.i11.855","volume":"6","author":"J Tohka","year":"2014","unstructured":"Tohka J (2014) Partial volume effect modeling for segmentation and tissue classification of brain magnetic resonance images: a review. World J Radiol 6:855\u2013864","journal-title":"World J Radiol"},{"key":"12302_CR16","first-page":"2053","volume":"40","author":"CC Meltzer","year":"1999","unstructured":"Meltzer CC, Kinahan PE, Greer PJ et al (1999) Comparative evaluation of MR-based partial-volume correction schemes for PET. J Nucl Med 40:2053\u20132065","journal-title":"J Nucl Med"},{"key":"12302_CR17","doi-asserted-by":"crossref","unstructured":"Kondrateva E, Pominova M, Popova E, Sharaev M, Bernstein A, Burnaev E (2021) Domain shift in computer vision models for MRI data analysis: an overview. SPIE","DOI":"10.1117\/12.2587872"},{"key":"12302_CR18","doi-asserted-by":"crossref","unstructured":"Zalevskyi V, Sanchez T, Roulet M et al (2024) Improving cross-domain brain tissue segmentation in fetal MRI WITH SYNTHETIC DAta. Springer Nature Switzerland, Cham, pp 437\u2013447","DOI":"10.1007\/978-3-031-72378-0_41"},{"key":"12302_CR19","doi-asserted-by":"crossref","unstructured":"Full PM, Isensee F, J\u00e4ger PF, Maier-Hein K (2021) Studying robustness of semantic segmentation under domain shift in cardiac MRI. Springer International Publishing, Cham, pp 238\u2013249","DOI":"10.1007\/978-3-030-68107-4_24"},{"key":"12302_CR20","doi-asserted-by":"publisher","DOI":"10.1016\/j.nicl.2018.101645","volume":"21","author":"S Basaia","year":"2019","unstructured":"Basaia S, Agosta F, Wagner L et al (2019) Automated classification of Alzheimer\u2019s disease and mild cognitive impairment using a single MRI and deep neural networks. Neuroimage Clin 21:101645","journal-title":"Neuroimage Clin"},{"key":"12302_CR21","doi-asserted-by":"publisher","first-page":"681","DOI":"10.1093\/brain\/awm319","volume":"131","author":"S Kl\u00f6ppel","year":"2008","unstructured":"Kl\u00f6ppel S, Stonnington CM, Chu C et al (2008) Automatic classification of MR scans in Alzheimer\u2019s disease. Brain 131:681\u2013689","journal-title":"Brain"},{"key":"12302_CR22","doi-asserted-by":"publisher","DOI":"10.1016\/j.clinimag.2024.110301","volume":"115","author":"A Zarei","year":"2024","unstructured":"Zarei A, Keshavarz A, Jafari E et al (2024) Automated classification of Alzheimer\u2019s disease, mild cognitive impairment, and cognitively normal patients using 3D convolutional neural network and radiomic features from T1-weighted brain MRI: a comparative study on detection accuracy. Clin Imaging 115:110301","journal-title":"Clin Imaging"},{"key":"12302_CR23","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-53733-6","volume":"14","author":"AM El-Assy","year":"2024","unstructured":"El-Assy AM, Amer HM, Ibrahim HM, Mohamed MA (2024) A novel CNN architecture for accurate early detection and classification of Alzheimer\u2019s disease using MRI data. Sci Rep 14:3463","journal-title":"Sci Rep"},{"key":"12302_CR24","doi-asserted-by":"publisher","first-page":"1920","DOI":"10.1093\/brain\/awaa137","volume":"143","author":"S Qiu","year":"2020","unstructured":"Qiu S, Joshi PS, Miller MI et al (2020) Development and validation of an interpretable deep learning framework for Alzheimer\u2019s disease classification. Brain 143:1920\u20131933","journal-title":"Brain"}],"container-title":["European Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-025-12302-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00330-025-12302-9","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-025-12302-9.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T10:17:33Z","timestamp":1781864253000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00330-025-12302-9"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,27]]},"references-count":24,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["12302"],"URL":"https:\/\/doi.org\/10.1007\/s00330-025-12302-9","relation":{},"ISSN":["1432-1084"],"issn-type":[{"value":"1432-1084","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,27]]},"assertion":[{"value":"20 May 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 November 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 December 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 February 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with ethical standards"}},{"value":"The scientific guarantor of this publication is Dennis M. Hedderich.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Guarantor"}},{"value":"B. Wiestler has disclosed no relevant relationships pertaining to activities associated with this article but has received speaker honoraria from Philips. C. Zimmer has disclosed no relevant relationships concerning this article but has served on scientific advisory boards for Philips and Bayer Schering. C. Zimmer is also a co-editor on the Advisory Board of Clinical Neuroradiology and has received speaker honoraria from Bayer-Schering and Philips. Additionally, C. Zimmer's institution has received research support and investigator fees for clinical studies from multiple companies, including Biogen Idec, Quintiles, MSD Sharp & Dohme, and Boehringer Ingelheim. D.M. Hedderich is Section Editor for Advanced Imaging of Neuroradiology and has received speakers\u2019 honoraria from Eisai GmbH. All other authors declare no conflicts of interest.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"No complex statistical methods exceeding the statistical expertise of the authors were necessary for this paper.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Statistics and biometry"}},{"value":"Written informed consent was waived by the Institutional Review Board.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed consent"}},{"value":"Institutional Review Board approval was obtained (2024-71-S-CB).","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"The patients of the cohort used here were all part of a larger cohort in: Hedderich et al, normative brain volume reports may improve differential diagnosis of dementing neurodegenerative diseases in clinical practice,\n                      European\n                      Radiology\n                      , 2020 and to a small part in Hedderich et al, Impact of normative brain volume reports on the diagnosis of neurodegenerative dementia disorders in neuroradiology: a real-world, clinical practice study, Frontiers in Aging Neuroscience, 2022. The current study investigates the preservation of atrophy patterns in neurodegenerative diseases and the impact on automated disease classification, and therefore has a completely different research focus than the other two studies, which investigated the effect of Normative brain volume reports on the diagnosis made by human radiologists.","order":7,"name":"Ethics","group":{"name":"EthicsHeading","label":"Study subjects or cohorts overlap"}},{"order":8,"name":"Ethics","group":{"name":"EthicsHeading","label":"Methodology"}}]}}