Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification

The unknown pathogenic mechanisms of Alzheimer’s disease (AD) make treatment challenging. Neuroimaging genetics offers a method for identifying disease biomarkers for early diagnosis, but traditional approaches struggle with complex non-linear, multimodal and multi-expression data. However, traditio...

Full description

Saved in:
Bibliographic Details
Main Authors: Shan Huang, Yixin Liu, Jingyu Zhang, Yiming Wang
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Artificial Cells, Nanomedicine, and Biotechnology
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/21691401.2025.2506591
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849309971205849088
author Shan Huang
Yixin Liu
Jingyu Zhang
Yiming Wang
author_facet Shan Huang
Yixin Liu
Jingyu Zhang
Yiming Wang
author_sort Shan Huang
collection DOAJ
description The unknown pathogenic mechanisms of Alzheimer’s disease (AD) make treatment challenging. Neuroimaging genetics offers a method for identifying disease biomarkers for early diagnosis, but traditional approaches struggle with complex non-linear, multimodal and multi-expression data. However, traditional association analysis methods face challenges in handling nonlinear, multimodal and multi-expression data. Therefore, a multimodal attention fusion deep self-restructuring presentation (MAFDSRP) model is proposed to solve the above problem. First, multimodal brain imaging data are processed through a novel histogram-matching multiple attention mechanisms to dynamically adjust the weight of each input brain image data. Simultaneous, the genetic data are preprocessed to remove low-quality samples. Subsequently, the genetic data and fused neuroimaging data are separately input into the self-reconstruction network to learn the nonlinear relationships and perform subspace clustering at the top layer of the network. Finally, the learned genetic data and fused neuroimaging data are analysed through expression association analysis to identify AD-related biomarkers. The identified biomarkers underwent systematic multi-level analysis, revealing biomarker roles at molecular, tissue and functional levels, highlighting processes like inflammation, lipid metabolism, memory and emotional processing linked to AD. The experimental results show that MAFDSRP achieved 0.58 in association analysis, demonstrating its great potential in accurately identifying AD-related biomarkers.
format Article
id doaj-art-b76dc85131264ea399a2294dcbae8d52
institution Kabale University
issn 2169-1401
2169-141X
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Artificial Cells, Nanomedicine, and Biotechnology
spelling doaj-art-b76dc85131264ea399a2294dcbae8d522025-08-20T03:53:56ZengTaylor & Francis GroupArtificial Cells, Nanomedicine, and Biotechnology2169-14012169-141X2025-12-0153123124310.1080/21691401.2025.2506591Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identificationShan Huang0Yixin Liu1Jingyu Zhang2Yiming Wang3The Second Affiliated Hospital, Harbin Medical University, Harbin, ChinaModern Education Technology Center, Harbin Medical University, Harbin, ChinaThe Fourth Affiliated Hospital of Harbin Medical University, Harbin, ChinaSchool of Computer and Control Engineering, Northeast Forestry University, Harbin, ChinaThe unknown pathogenic mechanisms of Alzheimer’s disease (AD) make treatment challenging. Neuroimaging genetics offers a method for identifying disease biomarkers for early diagnosis, but traditional approaches struggle with complex non-linear, multimodal and multi-expression data. However, traditional association analysis methods face challenges in handling nonlinear, multimodal and multi-expression data. Therefore, a multimodal attention fusion deep self-restructuring presentation (MAFDSRP) model is proposed to solve the above problem. First, multimodal brain imaging data are processed through a novel histogram-matching multiple attention mechanisms to dynamically adjust the weight of each input brain image data. Simultaneous, the genetic data are preprocessed to remove low-quality samples. Subsequently, the genetic data and fused neuroimaging data are separately input into the self-reconstruction network to learn the nonlinear relationships and perform subspace clustering at the top layer of the network. Finally, the learned genetic data and fused neuroimaging data are analysed through expression association analysis to identify AD-related biomarkers. The identified biomarkers underwent systematic multi-level analysis, revealing biomarker roles at molecular, tissue and functional levels, highlighting processes like inflammation, lipid metabolism, memory and emotional processing linked to AD. The experimental results show that MAFDSRP achieved 0.58 in association analysis, demonstrating its great potential in accurately identifying AD-related biomarkers.https://www.tandfonline.com/doi/10.1080/21691401.2025.2506591Alzheimer’s diseasedeep learningmultimodal fusionneuroimaging genetics
spellingShingle Shan Huang
Yixin Liu
Jingyu Zhang
Yiming Wang
Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
Artificial Cells, Nanomedicine, and Biotechnology
Alzheimer’s disease
deep learning
multimodal fusion
neuroimaging genetics
title Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
title_full Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
title_fullStr Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
title_full_unstemmed Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
title_short Multimodal attention fusion deep self-reconstruction presentation model for Alzheimer’s disease diagnosis and biomarker identification
title_sort multimodal attention fusion deep self reconstruction presentation model for alzheimer s disease diagnosis and biomarker identification
topic Alzheimer’s disease
deep learning
multimodal fusion
neuroimaging genetics
url https://www.tandfonline.com/doi/10.1080/21691401.2025.2506591
work_keys_str_mv AT shanhuang multimodalattentionfusiondeepselfreconstructionpresentationmodelforalzheimersdiseasediagnosisandbiomarkeridentification
AT yixinliu multimodalattentionfusiondeepselfreconstructionpresentationmodelforalzheimersdiseasediagnosisandbiomarkeridentification
AT jingyuzhang multimodalattentionfusiondeepselfreconstructionpresentationmodelforalzheimersdiseasediagnosisandbiomarkeridentification
AT yimingwang multimodalattentionfusiondeepselfreconstructionpresentationmodelforalzheimersdiseasediagnosisandbiomarkeridentification