本研究旨在探索利用多模态磁共振成像（MRI）技术结合机器学习方法，构建阿尔茨海默病（AD）的早期诊断模型。通过对不同模态的MRI图像进行特征提取与分析，我们试图发现与AD相关的神经影像学标志物，并评估其在疾病早期识别中的效能。研究纳入120名受试者，包括40名AD患者、40名轻度认知障碍（MCI）患者以及40名健康对照者。采用支持向量机（SVM）和随机森林（RF）两种机器学习算法构建分类模型，并通过交叉验证评估模型性能。结果显示，结合多模态MRI特征的模型在区分AD患者与健康对照者时表现出较高的准确率（SVM：87.5%，RF：85.0%），且在识别MCI转化为AD的风险方面也显示出潜在价值。本研究表明，多模态MRI结合机器学习技术为AD的早期诊断提供了新的有力工具，有望为临床干预和疾病管理提供重要支持。

关键字：阿尔茨海默病；多模态MRI；机器学习；早期诊断；支持向量机；随机森林

Abstract：

This study aims to explore the use of multimodal magnetic resonance imaging (MRI) technology combined with machine learning methods to construct an early diagnosis model for Alzheimer's disease (AD). By extracting and analyzing features from different modalities of MRI images, we seek to identify neuroimaging biomarkers associated with AD and evaluate their efficacy in early disease recognition. The study included 120 subjects, comprising 40 AD patients, 40 patients with mild cognitive impairment (MCI), and 40 healthy controls. Two machine learning algorithms, support vector machine (SVM) and random forest (RF), were employed to build classification models, and their performance was evaluated through cross-validation. The results indicated that models incorporating multimodal MRI features demonstrated high accuracy in distinguishing AD patients from healthy controls (SVM: 87.5%, RF: 85.0%) and also showed potential in identifying the risk of MCI converting to AD. This study suggests that multimodal MRI combined with machine learning technology provides a new and powerful tool for the early diagnosis of AD, which is expected to offer significant support for clinical intervention and disease management.

Keywords: Alzheimer's disease; Multimodal MRI; Machine learning; Early diagnosis; Support vector machine; Random forest

正文内容 / Content：

可下载全文PDF查阅，并按照本文版权申明进行使用。

Download the full text PDF for viewing and using it according to the license of this paper.、

参考文献 / References：

[1] 智威覃, 钊刘, 允敏陆, 等. 基于广义卷积神经网络的阿尔茨海默病多模态磁共振图像分类方法研究 [J]. Sheng Wu Yi Xue Gong Cheng Xue Za Zhi = Journal of Biomedical Engineering, 2023, 40(2): 217.

[2] 胡忠婕, 陈楠, 宋海庆, 等. 遗忘型轻度认知障碍和阿尔茨海默病的多模态 MRI 研究进展 [J]. 中华放射学杂志, 2014(6): 4. DOI: 10.3760/cma.j.issn.1005-1201.2014.06.024.

[3] 郭伟凡. 基于多模态磁共振影像的阿尔茨海默病早期分类研究 [J]. [2024-08-12].

[4] 郭春杰. 基于嗅觉 fMRI 及多参数 DTI 的阿尔茨海默病及轻度认知障碍多模态 MRI 研究 [D]. 吉林大学, 2016.

[5] 沈望舒. 阿尔茨海默病的 MRI 智能识别方法 [J]. [2024-08-12].

[6] 韩英妹, 李一杰, 张衡, 等. 基于 MRI 分析阿尔茨海默病大尺度脑网络研究进展 [J]. 实用医学杂志, 2024(004): 040.

[7] 卓奕楠, 杨鹏, 邓云, 等. 基于多模态典型相关特征表达的阿尔兹海默病诊断 [J]. 中国生物医学工程学报, 2018, 37(1): 7. DOI: 10.3969/j.issn.0258-8021.2018.01.001.

[8] 刘学娜, 张新卿. 基于功能磁共振成像的脑网络图论分析及其在阿尔茨海默病中的应用 [J]. 中国现代神经疾病杂志, 2012, 12(4): 483-487. DOI: 10.3969/j.issn.1672-6731.2012.04.023.

本文是一篇遵循创作共用许可证（CC BY）的开放获取文章

Canonical URL: https://creativecommons.org/licenses/by/4.0/