Alzheimer's Disease Prediction Using EffNetViTLoRA and BiLSTM with Multimodal Longitudinal MRI Data

2025-11-27

AI Summary

Early prediction of Alzheimer’s disease (AD) hinges on accurately distinguishing stable mild cognitive impairment (sMCI) from progressive MCI (pMCI), yet this classification remains challenging. To address this, we propose EffNetViTLoRA—a novel hybrid architecture integrating EfficientNet for local spatial feature extraction, Vision Transformer (ViT) for global contextual modeling, and Low-Rank Adaptation (LoRA) for parameter-efficient fine-tuning. Coupled with a bidirectional LSTM (BiLSTM) to capture temporal dynamics across longitudinal MRI scans (four timepoints), our framework enables end-to-end spatiotemporal feature learning for the first time. Furthermore, we incorporate non-imaging biomarkers (e.g., CSF, genetic, and clinical data) into a unified multimodal longitudinal prediction pipeline. Evaluated on the ADNI dataset, the model achieves a mean accuracy of 95.05% on sMCI/pMCI classification—surpassing state-of-the-art methods—and establishes a new paradigm for early AD progression prediction with current best-in-class performance.

Technology Category

Computer Vision: Multi-modal VisionMachine Learning: Multimodal LearningCognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

Search and Retrieval-Augmented AI: Vertical and domain-specific searchSystems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applicationsGraph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Alzheimer's disease (AD) is a prevalent neurodegenerative disorder that progressively impairs memory, decision-making, and overall cognitive function. As AD is irreversible, early prediction is critical for timely intervention and management. Mild Cognitive Impairment (MCI), a transitional stage between cognitively normal (CN) aging and AD, plays a significant role in early AD diagnosis. However, predicting MCI progression remains a significant challenge, as not all individuals with MCI convert to AD. MCI subjects are categorized into stable MCI (sMCI) and progressive MCI (pMCI) based on conversion status. In this study, we propose a generalized, end-to-end deep learning model for AD prediction using MCI cases from the Alzheimer's Disease Neuroimaging Initiative (ADNI). Our hybrid architecture integrates Convolutional Neural Networks and Vision Transformers to capture both local spatial features and global contextual dependencies from Magnetic Resonance Imaging (MRI) scans. To incorporate temporal progression, we further employ Bidirectional Long Short-Term Memory (BiLSTM) networks to process features extracted from four consecutive MRI timepoints along with some other non-image biomarkers, predicting each subject's cognitive status at month 48. Our multimodal model achieved an average progression prediction accuracy of 95.05% between sMCI and pMCI, outperforming existing studies in AD prediction. This work demonstrates state-of-the-art performance in longitudinal AD prediction and highlights the effectiveness of combining spatial and temporal modeling for the early detection of Alzheimer's disease.

Problem

• Predicting MCI progression to Alzheimer's using multimodal MRI data
• Integrating spatial and temporal features for early AD detection
• Distinguishing stable from progressive MCI with deep learning models

Innovation

• Hybrid CNN-Vision Transformer captures spatial features
• BiLSTM processes longitudinal MRI data over time
• Multimodal model integrates imaging and non-image biomarkers