A Bayesian classification model for differential diagnosis of Alzheimer's disease and frontotemporal dementia using plasma biomarkers
produced the highest-ranked triplet (AUC = 0.95) with modest, consistent gains. Posterior coefficients were biologically coherent (AD-specific pTau217 effects; severity-linked NfL), and calibration was satisfactory, with minor overconfidence only at extreme probabilities.ConclusionsA parsimonious pTau217 + NfL panel captures most diagnostic information in the full plasma profile, providing an accurate probabilistic classifier with interpretable uncertainty to support differential diagnosis and clinical triage in precision neurology.
Authors
- Tommaso Costa (ORCID: https://orcid.org/0000-0002-0822-862X)
- K. Blennow
- Donato Liloia (ORCID: https://orcid.org/0000-0002-9481-8510)
- Alessandro Padovani (ORCID: https://orcid.org/0000-0002-0119-3639)
- Henrik Zettemberg
- Enrico Premi
- Nicholas Ashton
- Barbara Borroni
Institutions
- University of Science and Technology of China (CN)
- University of Wisconsin–Madison (US)
- King's College London (GB)
- South London and Maudsley NHS Foundation Trust (GB)
- Stavanger University Hospital (NO)
- Sahlgrenska University Hospital (SE)
- Brescia University (US)
- Sorbonne Université (FR)
- Azienda Socio Sanitaria Territoriale degli Spedali Civili di Brescia (IT)
- Neuroscience Institute (IT)
- Ospedale Koelliker (IT)
- Centro San Giovanni di Dio Fatebenefratelli (IT)
- Pitié-Salpêtrière Hospital (FR)
- UK Dementia Research Institute (GB)
- Banner Sun Health Research Institute (US)
- Institut du Cerveau (FR)
- University College London (GB)
- University of Turin (IT)
- University of Brescia (IT)
- University of Gothenburg (SE)
Publication Details
- Journal
- Journal of Alzheimer s Disease
- Published
- 2026-04-06
- DOI
- https://doi.org/10.1177/13872877261439950
- Primary Topic
- Dementia and Cognitive Impairment Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00