AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment

Abstract Alzheimer’s disease (AD) diagnosis hinges on detecting amyloid beta (Aβ) plaques and neurofibrillary tau (τ) tangles, typically assessed using PET imaging. While accurate, these modalities are expensive and not widely accessible, limiting their utility in routine clinical practice. Here, we...

Full description

Saved in:
Bibliographic Details
Main Authors: Varuna H. Jasodanand, Sahana S. Kowshik, Shreyas Puducheri, Michael F. Romano, Lingyi Xu, Rhoda Au, Vijaya B. Kolachalama
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-62590-4
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849764241637113856
author Varuna H. Jasodanand
Sahana S. Kowshik
Shreyas Puducheri
Michael F. Romano
Lingyi Xu
Rhoda Au
Vijaya B. Kolachalama
author_facet Varuna H. Jasodanand
Sahana S. Kowshik
Shreyas Puducheri
Michael F. Romano
Lingyi Xu
Rhoda Au
Vijaya B. Kolachalama
author_sort Varuna H. Jasodanand
collection DOAJ
description Abstract Alzheimer’s disease (AD) diagnosis hinges on detecting amyloid beta (Aβ) plaques and neurofibrillary tau (τ) tangles, typically assessed using PET imaging. While accurate, these modalities are expensive and not widely accessible, limiting their utility in routine clinical practice. Here, we present a multimodal computational framework that integrates data from seven distinct cohorts comprising 12, 185 participants to estimate individual PET profiles using more readily available neurological assessments. Our approach achieved an AUROC of 0.79 and 0.84 in classifying Aβ and τ status, respectively. Predicted PET status was consistent with various biomarker profiles and postmortem pathology, and model-identified regional brain volumes aligned with known spatial patterns of tau deposition. This approach can support scalable pre-screening of candidates for anti-amyloid therapies and clinical trials targeting Aβ and τ, offering a practical alternative to direct PET imaging.
format Article
id doaj-art-edbe0b4f76a7426d87d2dfa4502830bc
institution DOAJ
issn 2041-1723
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-edbe0b4f76a7426d87d2dfa4502830bc2025-08-20T03:05:10ZengNature PortfolioNature Communications2041-17232025-08-0116111910.1038/s41467-025-62590-4AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessmentVaruna H. Jasodanand0Sahana S. Kowshik1Shreyas Puducheri2Michael F. Romano3Lingyi Xu4Rhoda Au5Vijaya B. Kolachalama6Department of Medicine, Boston University Chobanian & Avedisian School of MedicineDepartment of Medicine, Boston University Chobanian & Avedisian School of MedicineDepartment of Medicine, Boston University Chobanian & Avedisian School of MedicineDepartment of Radiology & Biomedical Imaging, University of California San FranciscoDepartment of Medicine, Boston University Chobanian & Avedisian School of MedicineDepartment of Medicine, Boston University Chobanian & Avedisian School of MedicineDepartment of Medicine, Boston University Chobanian & Avedisian School of MedicineAbstract Alzheimer’s disease (AD) diagnosis hinges on detecting amyloid beta (Aβ) plaques and neurofibrillary tau (τ) tangles, typically assessed using PET imaging. While accurate, these modalities are expensive and not widely accessible, limiting their utility in routine clinical practice. Here, we present a multimodal computational framework that integrates data from seven distinct cohorts comprising 12, 185 participants to estimate individual PET profiles using more readily available neurological assessments. Our approach achieved an AUROC of 0.79 and 0.84 in classifying Aβ and τ status, respectively. Predicted PET status was consistent with various biomarker profiles and postmortem pathology, and model-identified regional brain volumes aligned with known spatial patterns of tau deposition. This approach can support scalable pre-screening of candidates for anti-amyloid therapies and clinical trials targeting Aβ and τ, offering a practical alternative to direct PET imaging.https://doi.org/10.1038/s41467-025-62590-4
spellingShingle Varuna H. Jasodanand
Sahana S. Kowshik
Shreyas Puducheri
Michael F. Romano
Lingyi Xu
Rhoda Au
Vijaya B. Kolachalama
AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
Nature Communications
title AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
title_full AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
title_fullStr AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
title_full_unstemmed AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
title_short AI-driven fusion of multimodal data for Alzheimer’s disease biomarker assessment
title_sort ai driven fusion of multimodal data for alzheimer s disease biomarker assessment
url https://doi.org/10.1038/s41467-025-62590-4
work_keys_str_mv AT varunahjasodanand aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT sahanaskowshik aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT shreyaspuducheri aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT michaelfromano aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT lingyixu aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT rhodaau aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment
AT vijayabkolachalama aidrivenfusionofmultimodaldataforalzheimersdiseasebiomarkerassessment