View in EDS

NEUROFUSION-AD: A HYBRID 3D CNN-TRANSFORMER-BILSTM FRAMEWORK FOR FIVE-STAGE DEMENTIA PREDICTION AND DETECTION FROM STRUCTURAL MRI.

Saved in:
Bibliographic Details
Title: NEUROFUSION-AD: A HYBRID 3D CNN-TRANSFORMER-BILSTM FRAMEWORK FOR FIVE-STAGE DEMENTIA PREDICTION AND DETECTION FROM STRUCTURAL MRI.
Authors: Bhatt, Kaushal Kishor, Sehgal, Parveen
Source: Lex Localis: Journal of Local Self-Government; 2025, Vol. 23 Issue 11, p182-194, 13p
Subject Terms: DEMENTIA, DEEP learning, LONG short-term memory, CLASSIFICATION, MAGNETIC resonance imaging, TRANSFORMER models
Abstract: Background: Early and accurate detection of dementia progression remains a critical challenge in neuroimaging, with most existing approaches limited to binary or ternary classification schemes that inadequately capture the gradual cognitive decline characteristic of Alzheimer's disease (AD). Methods: We propose NeuroFusion-AD, a novel hybrid deep learning framework that integrates three complementary processing streams: (1) 3D convolutional neural networks for whole-brain volumetric feature extraction, (2) 2D CNN coupled with bidirectional LSTM for slicesequential temporal modeling, and (3) vision transformer with anatomical position encoding for multi-region fusion. The architecture incorporates ordinal classification constraints, self-supervised pretraining via masked volume modeling, and multi-task learning for joint stage prediction and cognitive score regression. We evaluated the framework on the ADNI dataset (2,847 subjects, 7,234 scans) with external validation on OASIS (755 subjects, 2,168 sessions) across five clinical stages: Cognitively Normal (CN), Significant Memory Concern (SMC), Early Mild Cognitive Impairment (EMCI), Late Mild Cognitive Impairment (LMCI), and Alzheimer's Disease (AD). Results: NeuroFusion-AD achieved superior performance with 97.5% accuracy, 97.2% precision, 96.9% recall, and 97.0% F1-score, significantly outperforming state-of-the-art methods including LGG-NeXt (95.81% accuracy), AD-Diff (90.78% accuracy), and traditional CNNs (72-78% accuracy). Stage-wise analysis demonstrated robust sensitivity (>94%) and specificity (>96%) across all dementia stages, with particularly strong performance in challenging SMC (95.8% sensitivity) and EMCI (94.5% sensitivity) classifications. Statistical significance testing confirmed improvements over baselines (p < 0.05 for all comparisons), with large effect sizes (Cohen's d: 0.42-2.87). Conclusions: The proposed NeuroFusion-AD framework addresses critical limitations in automated dementia classification by providing accurate five-stage categorization with clinical interpretability through attention visualization and ordinal constraint enforcement. The multi-stream architecture's balanced performance across all stages supports its potential for clinical deployment in early dementia detection and monitoring. [ABSTRACT FROM AUTHOR]
Copyright of Lex Localis: Journal of Local Self-Government is the property of Institute for Local Self-Government & Public Procurement Maribor and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
Description
Abstract:Background: Early and accurate detection of dementia progression remains a critical challenge in neuroimaging, with most existing approaches limited to binary or ternary classification schemes that inadequately capture the gradual cognitive decline characteristic of Alzheimer's disease (AD). Methods: We propose NeuroFusion-AD, a novel hybrid deep learning framework that integrates three complementary processing streams: (1) 3D convolutional neural networks for whole-brain volumetric feature extraction, (2) 2D CNN coupled with bidirectional LSTM for slicesequential temporal modeling, and (3) vision transformer with anatomical position encoding for multi-region fusion. The architecture incorporates ordinal classification constraints, self-supervised pretraining via masked volume modeling, and multi-task learning for joint stage prediction and cognitive score regression. We evaluated the framework on the ADNI dataset (2,847 subjects, 7,234 scans) with external validation on OASIS (755 subjects, 2,168 sessions) across five clinical stages: Cognitively Normal (CN), Significant Memory Concern (SMC), Early Mild Cognitive Impairment (EMCI), Late Mild Cognitive Impairment (LMCI), and Alzheimer's Disease (AD). Results: NeuroFusion-AD achieved superior performance with 97.5% accuracy, 97.2% precision, 96.9% recall, and 97.0% F1-score, significantly outperforming state-of-the-art methods including LGG-NeXt (95.81% accuracy), AD-Diff (90.78% accuracy), and traditional CNNs (72-78% accuracy). Stage-wise analysis demonstrated robust sensitivity (>94%) and specificity (>96%) across all dementia stages, with particularly strong performance in challenging SMC (95.8% sensitivity) and EMCI (94.5% sensitivity) classifications. Statistical significance testing confirmed improvements over baselines (p < 0.05 for all comparisons), with large effect sizes (Cohen's d: 0.42-2.87). Conclusions: The proposed NeuroFusion-AD framework addresses critical limitations in automated dementia classification by providing accurate five-stage categorization with clinical interpretability through attention visualization and ordinal constraint enforcement. The multi-stream architecture's balanced performance across all stages supports its potential for clinical deployment in early dementia detection and monitoring. [ABSTRACT FROM AUTHOR]
ISSN:15815374
DOI:10.52152/801720