Machine Learning Approach for Predicting Amyloid and Tau Positivity in Alzheimer’s Disease Using Clinically Accessible Features

Sep 30, 2026

Prediction of Alzheimer’s disease (AD) biomarkers can strengthen public health strategies when biomarker-relevant data are collected easily in a single clinical encounter. This poster evaluates a machine learning model that predicts amyloid and tau positivity using clinically accessible features rather than complex workflows. Samples were built from ADNI, TRIAD, and PPMI and validated on NACC, using features such as age, sex, education, diagnosis, MoCA scores, and BMI. Amyloid/tau positivity was defined by PET or CSF measures, achieving high AUC performance in both phases.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages3

Reading Time8 min

AccessFree to download

Biomarkers

Background

Method

Result

Conclusion

Correspondence

DOI: 10. 1002/alz70856_105937 BIOMARKERS POSTER PRESENTATION BIOMARKERS (NON-NEUROIMAGING) Machine Learning Approach for Predicting Amyloid and Tau Positivity in Alzheimer’s Disease Using Clinically Accessible Features Daniel Arnold1 Luiza Santos Machado2 Nesrine Rahmouni3 Joseph Therriault4 Stijn Servaes4 Jenna Stevenson4 Arthur C. Macedo4 Artur Francisco Schumacher-Schuh5 Christian Mattjie6 Firoza Z Lussier7 Mira Chamoun4 Gleb Bezgin4 Andrea L. Benedet8 Tharick A Pascoal7 Rodrigo C. Barros6 Marco De Bastiani 1 Pedro Rosa-Neto4 Eduardo R. Zimmer 1, 9, 10 Wyllians Vendramini Borelli 1, 11, 12 Alzheimer’s Disease Neuroimaging Initiative 13 1 Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil 2 University of Gothenburg, Gothenburg, VG, Sweden 3 Montreal Neurological Institute, Montreal, QC, Canada 4 McGill University, Montreal, QC, Canada 5 Hospital de Clínicas de Porto Alegre, Porto Alegre, RS, Brazil 6 PUCRS, Porto Alegre, Rio Grande do Sul, Brazil 7 University of Pittsburgh, Pittsburgh, PA, USA 8 Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy, University of Gothenburg, Mölndal, Sweden 9 Brain Institute of Rio Grande do Sul (InsCer), PUCRS, Porto Alegre, Rio Grande do Sul, Brazil 10 McGill Centre for Studies in Aging, Montreal, QC, Canada 11 Centro de Memória, Hospital Moinhos de Vento, Porto Alegre, RS, Brazil 12 Clinical Hospital of Porto Alegre, Porto Alegre, Rio Grande do Sul, Brazil 13 Laboratory of Neuro Imaging (LONI), University of Southern California, Los Angeles, CA, USA Abstract Background: Prediction of Alzheimer’s disease (AD) biomarkers can improve public health strategies, especially if achieved with easily collectable data in a single consultation. Machine learning (ML) offers versatile tools for clinical and research applications. This study investigated a ML model’s ability to predict amyloid and tau positivity using easily obtainable features. Method: Individuals with amyloid and tau status were selected from ADNI, TRIAD, and PPMI datasets (Model Building) and from the NACC dataset (Validation). Shared clinical features included age, sex, education, clinical diagnosis, MoCA scores, and BMI. Amyloid positivity was defined by amyloid-PET (PIB-PET, FBB-PET, or AZD4694-PET) or CSF AB42, and tau positivity by Tau-PET (MK6240-PET, AV1451-PET) or CSF ptau181. For NACC, positivity derived from fields AMYLPET and TAUPETAD. Data processing is summarized in Figure 1. Result: The Model Building sample included 1593 individuals (mean MoCA 25.3 ± 4. 3). The Validation cohort included 861 individuals (mean MoCA 22.2 ± 2.8). The model achieved high performance for predicting amyloid and tau positivity, with mean AUCs of 0.89 and 0.83 for Model Building and Validation, respectively (Figure 2a, 2b). In Validation, high sensitivity (0.93) came at the expense of specificity (0.65), while Model Building showed balanced sensitivity and specificity (0.81 and 0.83). Higher age, lower This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2025 The Alzheimer’s Association. Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association. Alzheimer’s Dement. 2025;21(Suppl. 2):e105937. [https://doi.org/10.1002/alz70856_105937](https://doi.org/10.1002/alz70856_105937) wi[leyonlinelibrary.com/journal/alz](leyonlinelibrary.com/journal/alz) 1of3 Correspondence Daniel Arnold, Universidade Federal do Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil. [Email:](Email: contatodanielarnold@gmail.com)[ contatodanielarnold@gmail.com](Email: contatodanielarnold@gmail.com) MoCA, cognitive impairment, female sex, and lower BMI increased the probability of positivity (Figure 2c, 2d). Co

FAQs

What clinical features were used to predict amyloid and tau positivity?

The model used age, sex, education, clinical diagnosis, MoCA scores, and BMI as shared clinical features.

How was amyloid and tau positivity defined in the study?

Amyloid positivity was defined using amyloid-PET (PIB-PET, FBB-PET, or AZD4694-PET) or CSF AB42, and tau positivity used Tau-PET (MK6240-PET, AV1451-PET) or CSF ptau181.

What performance did the model achieve in validation?

In validation, the model showed mean AUCs of 0.89 for model building and 0.83 for validation, with high sensitivity (0.93) and specificity (0.65) reported for validation.

Sep 30, 2026

Uploaded by anakgang17