School of Computer Science and Technology, Henan Polytechnic University, Jiaozuo, Henan 454000, PR China.
School of Computer Science and Technology, Henan Polytechnic University, Jiaozuo, Henan 454000, PR China.
Comput Methods Programs Biomed. 2024 Sep;254:108259. doi: 10.1016/j.cmpb.2024.108259. Epub 2024 Jun 6.
Alzheimer's disease (AD) is a dreaded degenerative disease that results in a profound decline in human cognition and memory. Due to its intricate pathogenesis and the lack of effective therapeutic interventions, early diagnosis plays a paramount role in AD. Recent research based on neuroimaging has shown that the application of deep learning methods by multimodal neural images can effectively detect AD. However, these methods only concatenate and fuse the high-level features extracted from different modalities, ignoring the fusion and interaction of low-level features across modalities. It consequently leads to unsatisfactory classification performance.
In this paper, we propose a novel multi-scale attention and cross-enhanced fusion network, MACFNet, which enables the interaction of multi-stage low-level features between inputs to learn shared feature representations. We first construct a novel Cross-Enhanced Fusion Module (CEFM), which fuses low-level features from different modalities through a multi-stage cross-structure. In addition, an Efficient Spatial Channel Attention (ECSA) module is proposed, which is able to focus on important AD-related features in images more efficiently and achieve feature enhancement from different modalities through two-stage residual concatenation. Finally, we also propose a multiscale attention guiding block (MSAG) based on dilated convolution, which can obtain rich receptive fields without increasing model parameters and computation, and effectively improve the efficiency of multiscale feature extraction.
Experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate that our MACFNet has better classification performance than existing multimodal methods, with classification accuracies of 99.59 %, 98.85 %, 99.61 %, and 98.23 % for AD vs. CN, AD vs. MCI, CN vs. MCI and AD vs. CN vs. MCI, respectively, and specificity of 98.92 %, 97.07 %, 99.58 % and 99.04 %, and sensitivity of 99.91 %, 99.89 %, 99.63 % and 97.75 %, respectively.
The proposed MACFNet is a high-accuracy multimodal AD diagnostic framework. Through the cross mechanism and efficient attention, MACFNet can make full use of the low-level features of different modal medical images and effectively pay attention to the local and global information of the images. This work provides a valuable reference for multi-mode AD diagnosis.
阿尔茨海默病（AD）是一种可怕的退行性疾病，会导致人类认知和记忆能力的严重下降。由于其复杂的发病机制和缺乏有效的治疗干预措施，早期诊断在 AD 中起着至关重要的作用。最近基于神经影像学的研究表明，通过多模态神经影像应用深度学习方法可以有效检测 AD。然而，这些方法仅对来自不同模态的高级特征进行串联和融合，忽略了跨模态的低级特征的融合和交互。因此导致分类性能不佳。
本文提出了一种新的多尺度注意力和跨增强融合网络 MACFNet，该网络能够在输入之间进行多阶段的低级特征交互，以学习共享特征表示。我们首先构建了一个新颖的 Cross-Enhanced Fusion Module (CEFM)，该模块通过多阶段交叉结构融合来自不同模态的低级特征。此外，还提出了一种高效空间通道注意力（ECSA）模块，该模块能够更有效地关注图像中与 AD 相关的重要特征，并通过两级残差串联从不同模态实现特征增强。最后，我们还提出了一种基于扩张卷积的多尺度注意力引导块（MSAG），该块可以在不增加模型参数和计算量的情况下获得丰富的感受野，并有效地提高多尺度特征提取的效率。
在阿尔茨海默病神经影像学倡议（ADNI）数据集上的实验表明，我们的 MACFNet 比现有的多模态方法具有更好的分类性能，对 AD 与 CN、AD 与 MCI、CN 与 MCI 以及 AD 与 CN 与 MCI 的分类准确率分别为 99.59%、98.85%、99.61%和 98.23%，特异性分别为 98.92%、97.07%、99.58%和 99.04%，敏感性分别为 99.91%、99.89%、99.63%和 97.75%。
所提出的 MACFNet 是一种高精度的多模态 AD 诊断框架。通过交叉机制和有效的注意力，MACFNet 可以充分利用不同模态医学图像的低级特征，并有效地关注图像的局部和全局信息。这项工作为多模态 AD 诊断提供了有价值的参考。
