Early structural changes in the hippocampus serve as critical biomarkers for the diagnosis of Alzheimer's disease (AD). However, traditional methods primarily concentrate on intra-regional features, overlooking the informative contributions of edge voxel regions. To address this limitation, we propose an Edge Voxel Feature Fusion Model (EVFFM) that integrates both internal and edge voxel features to enhance AD classification performance. Firstly, a multiscale edge mask is generated from the segmented hippocampal region using a stepwise binarization algorithm. Radiomics and deep learning features are then independently extracted from both the ROI and its edge regions. Sub-sequently, feature selection is conducted using a t-test, Spearman correlation analysis, and LASSO regression. Finally, the retained features of both areas are fused to construct the classification model. Evaluated on the ADNI dataset, EVFFM achieves an accuracy of 93.6 % ($\text{AUC}=0.973$) for AD vs. NC classification and 83.3 % ($\text{AUC}=0.822$) for MCI vs. NC. Furthermore, SHAP-based interpretability analysis confirmed the critical contribution of edge voxel features to the classification outcomes. The proposed method offers a novel approach to auxiliary diagnosis of AD.

In diagnosing Alzheimer’s Disease (AD), structural Magnetic Resonance Imaging (sMRI) is crucial, providing axial, coronal, and sagittal views for a comprehensive analysis of brain structures and pathological changes. Region of Interest (ROI) techniques in neuroimaging further allow detailed examination of structural and functional changes linked to AD. Additionally, certain SNPs have been identified as significant genetic markers for AD risk. Given the capabilities of Convolutional Neural Networks (CNNs) and Transformer models in handling multimodal data, this study introduces a novel hybrid network, DAResNet-ViT, for AD diagnosis. The model utilizes 24 key sMRI slices from each view per participant, combined with ROI features and genetic data, for classifying subjects. Data from two Alzheimer’s Disease Neuroimaging Initiative (ADNI) datasets were preprocessed and used to train and validate DAResNet-ViT. The model was evaluated through experiments focusing on AD diagnosis and the prediction of Mild Cognitive Impairment (MCI) conversion, with its effectiveness demonstrated in ablation studies and comparative experiments.

Magnetic resonance imaging (MRI) technology shows significant potential in predicting early pathological changes associated with Alzheimer's disease (AD). However, the complexity of MRI and their multidimensional characteristics pose challenges in model classification tasks, and single-scale features often fail to capture subtle changes in lesion areas effectively. To improve the accuracy of AD predictions, we propose a network model based on multi-scale features and a dual self-attention mechanism (MSDA). This model integrates depthwise separable convolutions with an improved self-attention mechanism, thereby enhancing the classification ability for AD. First, MRI undergo head motion correction, image alignment, and skull stripping to enhance the model's capability to extract features related to AD lesions. Second, we designed a multi-scale convolutional network structure that utilizes depthwise separable convolution kernels of varying sizes, allowing the network to effectively capture multi-scale feature information from MRI and accurately identify lesion areas. Finally, we introduced a dual self-attention module, which includes channel self-attention and spatial self-attention, further augmenting the model's ability to extract lesion features by learning the differences in features across different categories of MRI in both channel and spatial dimensions. Experimental results indicate that the MSDA network model demonstrates exceptional performance in classifying normal controls (NC) and AD within the ADNI dataset, outperforming existing models in classification accuracy, performance, and generalization capability. The accuracy, sensitivity, and specificity of the model reached 97.8%, 96.3%, and 99.4%, respectively.

Alzheimer’s disease (AD) is a common neurodegenerative disease in the elderly, characterized by progressive memory decline, language impairment, and behavioral abnormalities. MRI has been widely utilized as a non-invasive technique for diagnosing AD. However, traditional cnn have limitations in extracting multi-scale lesion features, making it challenging to fully capture the local information and global dependencies of the lesion area. To overcome this issue, we propose a multi-scale feature attention network (MSFAN), which integrates multi-scale feature extraction (MSF), channel attention mechanism (CAM) and non-local block for AD classification. Specifically, the MSF module extracts local features at different scales through cascaded multi-scale convolution kernels; the CAM module adaptively reweights channel responses to emphasize critical lesion areas; and the non-local block captures long-range dependencies across different brain regions, thereby enhancing global feature representation. Experimental results show that MSFAN performs well compared to other cnn models in AD classification tasks, with accuracy, precision, recall and F1 score all significantly improved.

Alzheimer’s disease (AD) is a neurodegenerative disorder that primarily affects the elderly, characterized by cognitive decline and memory loss. Magnetic resonance imaging (MRI) plays a crucial role in the auxiliary diagnosis of AD. Recent studies have demonstrated that deep learning algorithms have achieved remarkable success in predicting AD. As an innovative feature extraction technology, ResNet has considerably advanced the application of deep learning in medical imaging. 3D convolutional neural networks (CNNs) can process three-dimensional MRI data directly, capturing the intricate three-dimensional characteristics of the brain’s structure. This study proposes a novel framework based on 3D ResNet, which integrates spatial and channel attention mechanisms to significantly enhance the sensitivity and accuracy of feature representation. Furthermore, we have incorporated non-local blocks within the residual blocks to enable the model to capture long-range dependencies. Experimental analysis indicates that our 3D ResNet model exhibits outstanding classification performance and can effectively support the clinical diagnosis of AD.

Alzheimer's disease (AD) is a neurodegenerative disorder that progresses in a slow and irreversible manner. Although many computer-aided methods have been used to diagnose AD, the issue of underutilization of detailed information and features persists. In this study, we propose a new AD diagnostic network (FPRes-Net) that can fully learn the rich information of 3D MRI slices by extracting multi-scale features and feature fusion. Firstly, in order to fully extract multi-scale information, a network structure combining ResNet-50 with feature pyramids was designed. Next, a feature fusion method was designed to reduce noise and increase the importance of important features. Finally, a visually interpretable method called Gradient-weighted Class Activation Mapping (Grad-CAM) was introduced to visualize important feature regions in AD diagnosis. Experimental analysis was conducted on the publicly accessible ADNI-1 dataset, and our proposed FPRes-Net model performed better than other advanced research methods, with an accuracy rate of 99.5%. Our proposed model can be effectively used for clinical diagnosis of AD.