AI‐enhanced Centiloid quantification of amyloid PET images
Novel artificial intelligence (AI)-method that penalizes biologically implausible longitudinal trajectories, enabling the model to learn standardized uptake value ratios (SUVR) correction factors without requiring longitudinal data at inference time. Improves Centiloid consistency across tracers and studies, significantly enhancing cross-sectional and longitudinal amyloid positron emission tomography (PET) quantification. DeepSUVR-derived Centiloids show stronger associations with cognition, visual reads, and neuropathology. Longitudinal variability is reduced three- to five-fold, enabling more reliable tracking of amyloid accumulation and better detection of treatment effects. Novel reference and target masks derived from DeepSUVR replicate most of the model's performance, offering a practical alternative for integration into existing pipelines.
Authors
- Georgios Zisis (ORCID: https://orcid.org/0000-0003-2967-0318)
- Michelle K. Lupton (ORCID: https://orcid.org/0000-0002-7274-7299)
- Robert Adam (ORCID: https://orcid.org/0000-0003-0053-7919)
- Ashley Gillman (ORCID: https://orcid.org/0000-0001-9130-1092)
- Azadeh Feizpour
- Léo Lebrat (ORCID: https://orcid.org/0000-0003-1427-4523)
- Timothy Cox (ORCID: https://orcid.org/0000-0002-7232-5096)
- Joanne S. Robertson
- Gill Farrar (ORCID: https://orcid.org/0000-0002-0726-723X)
- The Health and Aging Brain Study (HABS‐HD) Study Team
- Michael J Breakspear
- Christopher C. Rowe
- OASIS3
- ADNI‐DOD
- Michael W. Weiner
- ADNeT
- Clifford R. Jack Jr
- Tammie LS. Benzinger
- Colin L. Masters
- Robert Koeppe
- Sid E. O'Bryant
- Pamela LaMontagne
- John C. Morris
- Cathy Short
- Ariane Bollack
- Jurgen Fripp
- Victor L. Villemagne
- PISA
- Manu S. Goyal
- Susan Landau
- Lyduine E. Collij
- Frederik Barkhof
- WRAP
- Reisa Sperling
- the A4/LEARN Study Team
- Ying Xia
- Duygu Tosun
- Sterling C. Johnson
- Christopher G. Schwarz
- for the Alzheimer's Disease Neuroimaging Initiative
- Pierrick Bourgeat
Institutions
- University of North Texas (US)
- Brigham and Women's Hospital (US)
- Netherlands Institute for Neuroscience (NL)
- San Francisco VA Medical Center (US)
- Mayo Clinic (US)
- University of Wisconsin–Madison (US)
- Queensland University of Technology (AU)
- University of North Texas Health Science Center (US)
- The University of Queensland (AU)
- University of Pittsburgh (US)
- University of California, San Francisco (US)
- General Electric (Spain) (ES)
- Washington University in St. Louis (US)
- QIMR Berghofer Medical Research Institute (AU)
- Florey Institute of Neuroscience and Mental Health (AU)
- University of Michigan (US)
- University of California System (US)
- Women's and Children's Health Network (AU)
- Mallinckrodt (United States) (US)
- Amsterdam Neuroscience (NL)
- Australian e-Health Research Centre (AU)
- CSIRO Health and Biosecurity (AU)
- National Hospital for Neurology and Neurosurgery (GB)
- Amsterdam University Medical Centers (NL)
- Austin Health (AU)
- University College London (GB)
- University of Newcastle Australia (AU)
- Vrije Universiteit Amsterdam (NL)
- University of California, Berkeley (US)
Publication Details
- Journal
- Alzheimer s & Dementia
- Published
- 2026-02-01
- DOI
- https://doi.org/10.1002/alz.71162
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00