EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI‐enhanced Centiloid quantification of amyloid PET images

Pierrick Bourgeat iD, Jurgen Fripp, Leo Lebrat, Ying Xia, Azadeh Feizpour, Timothy Cox, Georgios Zisis, Ashley Gillman, Manu S Goyal, Duygu Tosun, Tammie Ls Benzinger, Pamela LaMontagne und 33 weitere

Alzheimer's & Dementia · 2026-02 · 22 (2)

DOI: 10.1002/alz.71162 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Alzheimer's & Dementia, online erschienen am 11.02.2026. Verfasst von Pierrick Bourgeat und 44 weiteren Autor:innen. Beteiligte Einrichtung: Australian eHealth Research Centre, CSIRO Health and Biosecurity, Brisbane, QLD, Australia. Hauptthemen laut MEDLINE: Positron-Emission Tomography, Alzheimer Disease, Deep Learning, Amyloid beta-Peptides. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIA NIH HHS, NIBIB NIH HHS, NCATS NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION The Centiloid scale is the standard for amyloid (Aβ) PET quantification in research and clinical settings. However, variability between tracers and scanners remains a challenge. METHODS This study introduces DeepSUVR, a deep learning method to correct Centiloid quantification, by penalizing implausible longitudinal trajectories during training. The model was trained using data from 2,129 participants (7,149 Aβ positron emission tomography [PET] scans) in the Australian Imaging, Biomarkers and Lifestyle Study of ageing (AIBL)/Alzheimer's Disease Neuroimaging Initiative (ADNI) and validated using 15,807 Aβ PET scans from 10,543 participants across 10 external datasets. RESULTS DeepSUVR increased correlation between tracers, and reduced variability in the Aß‐negatives. It showed significantly stronger association with cognition, visual reads, neuropathology, and increased longitudinal consistency between studies. DeepSUVR also increased the effect size for detecting small treatment related slowing of amyloid accumulation in the A4 study. DISCUSSION DeepSUVR substantially advances Aβ PET quantification, outperforming all standard approaches, which is particularly important for clinical decision making and to detect subtle or early changes in Aβ. Highlights 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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Positron-Emission TomographyAlzheimer DiseaseDeep LearningAmyloid beta-PeptidesBrainHumansFemaleMaleAgedNeuroimagingAged, 80 and over

Stichwörter der Autor:innen

Alzheimer's diseaseCentiloidamyloid positron emission tomographydeep Learninglongitudinal analysis

Substanzen

Amyloid beta-Peptides

Autor:innen und Institutionen

  • Australian eHealth Research Centre, CSIRO Health and Biosecurity, Brisbane, QLD, Australia.
  • School of Electrical Engineering & Robotics, Faculty of Engineering, Queensland University of Technology, Brisbane, QLD, Australia.
  • Department of Molecular Imaging & Therapy, Austin Health, Melbourne, VIC, Australia.
  • The Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, Melbourne, VIC, Australia.
  • Mallinckrodt Institute of Radiology, Washington University School of Medicine, St Louis, Missouri, USA.
  • Department of Neurology, Washington University School of Medicine, St. Louis, Missouri, USA.

ORCID-Profile: Pierrick Bourgeat

Förderung

  • NIA NIH HHS
  • NIBIB NIH HHS
  • NCATS NIH HHS
  • NIH HHS
  • NINDS NIH HHS
  • National Health and Medical Research Council
  • National Institutes of Health

Zitieren

APA

Bourgeat, P., Fripp, J., Lebrat, L., Xia, Y., Feizpour, A., Cox, T., Zisis, G., Gillman, A., Goyal, M. S., Tosun, D., Benzinger, T. L., LaMontagne, P., Breakspear, M., Lupton, M. K., Short, C., Adam, R., Robertson, J. S., Sperling, R., O'Bryant, S. E., …, & Dore, V. (2026). AI‐enhanced Centiloid quantification of amyloid PET images. Alzheimer's & Dementia, 22(2). https://doi.org/10.1002/alz.71162

Vancouver

Bourgeat P, Fripp J, Lebrat L, Xia Y, Feizpour A, Cox T, et al. AI‐enhanced Centiloid quantification of amyloid PET images. Alzheimer's & Dementia. 2026;22(2). doi:10.1002/alz.71162

BibTeX
@article{Bourgeat2026,
  author = {Bourgeat, Pierrick and Fripp, Jurgen and Lebrat, Leo and Xia, Ying and Feizpour, Azadeh and Cox, Timothy and Zisis, Georgios and Gillman, Ashley and Goyal, Manu S and Tosun, Duygu and Benzinger, Tammie Ls and LaMontagne, Pamela and Breakspear, Michael and Lupton, Michelle K and Short, Cathy and Adam, Robert and Robertson, Joanne S and Sperling, Reisa and O'Bryant, Sid E and Johnson, Sterling C and Jr, Clifford R Jack and Schwarz, Christopher G and Barkhof, Frederik and Farrar, Gill and Bollack, Ariane and Collij, Lyduine E and Landau, Susan and Koeppe, Robert and Alzheimer's Disease Neuroimaging Initiative, Alzheimer's Disease Neuroimaging Initiative and OASIS3, OASIS3 and A4/LEARN Study Team, A4/LEARN Study Team and AMYPAD consortium, AMYPAD consortium and Health and Aging Brain Study (HABS‐HD) Study Team, Health and Aging Brain Study (HABS‐HD) Study Team and Mayo Clinic Study of Aging, Mayo Clinic Study of Aging and WRAP, WRAP and ADNI‐DOD, ADNI‐DOD and PISA, PISA and ADNeT, ADNeT and AIBL research group, AIBL research group and Morris, John C and Weiner, Michael W and Villemagne, Victor L and Masters, Colin L and Rowe, Christopher C and Dore, Vincent},
  title = {AI‐enhanced Centiloid quantification of amyloid PET images},
  journal = {Alzheimer's & Dementia},
  year = {2026},
  volume = {22},
  number = {2},
  doi = {10.1002/alz.71162}
}
Kostenlos starten →