An integrated multimodal neural network framework for early diagnosis of Alzheimer’s disease and mild cognitive impairment combining genomic, neuroimaging, and behavioral modalities

In the field of biomedical informatics and the diagnosis of Alzheimer’s disease, the swift and precise detection of Alzheimer’s disease (AD) and its early stage, mild cognitive impairment (MCI), presents a considerable obstacle in precision medicine. Traditional diagnostic techniques generally rely on singular data sources, thus failing to adequately capture the complex biological and behavioral interrelations that influence disease advancement. This study presents a novel multimodal deep learning architecture designed to amalgamate genetic information, neuroimaging techniques, and behavioral assessments to improve early detection proficiency. The suggested framework consists of three simultaneous analytical pathways: a three-dimensional convolutional neural network (3D-CNN) designed for the extraction of morphological characteristics from MRI data, a densely interconnected neural network focused on the analysis of genetic variants, and a non-diagnostic longitudinal assessment of behavioral trajectories and cognitive measurements. To efficiently integrate these diverse data streams, an attention-driven adaptive modality gating mechanism is utilized, which dynamically adjusts the impact of each modality. Validation employing harmonized multimodal data from ADNI and UK Biobank demonstrates that the integrated multimodal model exceeds unimodal alternatives regarding predictive efficacy and resilience. Furthermore, the modality gating mechanism provides comprehensible results by pinpointing significant multimodal biomarkers that signify early stages of Alzheimer’s disease. This framework collectively highlights the potential of integrative multimodal data fusion techniques in improving early diagnostic accuracy and enabling personalized risk assessment for Alzheimer’s disease and mild cognitive impairment.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-21
DOI
https://doi.org/10.1038/s41598-026-81578-8
Primary Topic
Dementia and Cognitive Impairment Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

An integrated multimodal neural network framework for early diagnosis of Alzheimer’s disease and mild cognitive impairment combining genomic, neuroimaging, and behavioral modalities

Yongjian Cui, Yan Li
Scientific Reports
Dementia and Cognitive Impairment Research
article

An integrated multimodal neural network framework for early diagnosis of Alzheimer’s disease and mild cognitive impairment combining genomic, neuroimaging, and behavioral modalities

Yongjian Cui, Yan Li
article en

Abstract

In the field of biomedical informatics and the diagnosis of Alzheimer’s disease, the swift and precise detection of Alzheimer’s disease (AD) and its early stage, mild cognitive impairment (MCI), presents a considerable obstacle in precision medicine. Traditional diagnostic techniques generally rely on singular data sources, thus failing to adequately capture the complex biological and behavioral interrelations that influence disease advancement. This study presents a novel multimodal deep learning architecture designed to amalgamate genetic information, neuroimaging techniques, and behavioral assessments to improve early detection proficiency. The suggested framework consists of three simultaneous analytical pathways: a three-dimensional convolutional neural network (3D-CNN) designed for the extraction of morphological characteristics from MRI data, a densely interconnected neural network focused on the analysis of genetic variants, and a non-diagnostic longitudinal assessment of behavioral trajectories and cognitive measurements. To efficiently integrate these diverse data streams, an attention-driven adaptive modality gating mechanism is utilized, which dynamically adjusts the impact of each modality. Validation employing harmonized multimodal data from ADNI and UK Biobank demonstrates that the integrated multimodal model exceeds unimodal alternatives regarding predictive efficacy and resilience. Furthermore, the modality gating mechanism provides comprehensible results by pinpointing significant multimodal biomarkers that signify early stages of Alzheimer’s disease. This framework collectively highlights the potential of integrative multimodal data fusion techniques in improving early diagnostic accuracy and enabling personalized risk assessment for Alzheimer’s disease and mild cognitive impairment.

Scientific Reports
Xinjiang Medical University (CN), Sixth Affiliated Hospital of Xinjiang Medical University (CN)
Dementia and Cognitive Impairment Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon