A multi-modal and multi-stage region of interest-based fusion network convolutional neural network model to differentiate progressive mild cognitive impairment from stable mild cognitive impairment.

IF 3.4 3区医学 Q2 NEUROSCIENCES

Journal of Alzheimer's Disease Pub Date : 2024-11-22 DOI:10.1177/13872877241295287

Zhenpeng Chen, Beier Qi, Bin Jing, Ruijuan Dong, Rong Chen, Pujie Feng, Yilu Shou, Haiyun Li

{"title":"A multi-modal and multi-stage region of interest-based fusion network convolutional neural network model to differentiate progressive mild cognitive impairment from stable mild cognitive impairment.","authors":"Zhenpeng Chen, Beier Qi, Bin Jing, Ruijuan Dong, Rong Chen, Pujie Feng, Yilu Shou, Haiyun Li","doi":"10.1177/13872877241295287","DOIUrl":null,"url":null,"abstract":"Background: Accurately differentiating stable mild cognitive impairment (sMCI) from progressive MCI (pMCI) is clinically relevant, and identification of pMCI is crucial for timely treatment before it evolves into Alzheimer's disease (AD).Objective: To construct a convolutional neural network (CNN) model to differentiate pMCI from sMCI integrating features from structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) images.Methods: We proposed a multi-modal and multi-stage region of interest (ROI)-based fusion network (m2ROI-FN) CNN model to differentiate pMCI from sMCI, adopting a multi-stage fusion strategy to integrate deep semantic features and multiple morphological metrics derived from ROIs of sMRI and PET images. Specifically, ten AD-related ROIs of each modality images were selected as patches inputting into 3D hierarchical CNNs. The deep semantic features extracted by the CNNs were fused through the multi-modal integration module and further combined with the multiple morphological metrics extracted by FreeSurfer. Finally, the multilayer perceptron classifier was utilized for subject-level MCI recognition.Results: The proposed model achieved accuracy of 77.4% to differentiate pMCI from sMCI with 5-fold cross validation on the entire ADNI database. Further, ADNI-1&2 were formed into an independent sample for model training and validation, and ADNI-3&GO were formed into another independent sample for multi-center testing. The model achieved 73.2% accuracy in distinguishing pMCI and sMCI on ADNI-1&2 and 75% accuracy on ADNI-3&GO.Conclusions: An effective m2ROI-FN model to distinguish pMCI from sMCI was proposed, which was capable of capturing distinctive features in ROIs of sMRI and PET images. The experimental results demonstrated that the model has the potential to differentiate pMCI from sMCI.","PeriodicalId":14929,"journal":{"name":"Journal of Alzheimer's Disease","volume":" ","pages":"13872877241295287"},"PeriodicalIF":3.4000,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Alzheimer's Disease","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1177/13872877241295287","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"NEUROSCIENCES","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Accurately differentiating stable mild cognitive impairment (sMCI) from progressive MCI (pMCI) is clinically relevant, and identification of pMCI is crucial for timely treatment before it evolves into Alzheimer's disease (AD).

Objective: To construct a convolutional neural network (CNN) model to differentiate pMCI from sMCI integrating features from structural magnetic resonance imaging (sMRI) and positron emission tomography (PET) images.

Methods: We proposed a multi-modal and multi-stage region of interest (ROI)-based fusion network (m2ROI-FN) CNN model to differentiate pMCI from sMCI, adopting a multi-stage fusion strategy to integrate deep semantic features and multiple morphological metrics derived from ROIs of sMRI and PET images. Specifically, ten AD-related ROIs of each modality images were selected as patches inputting into 3D hierarchical CNNs. The deep semantic features extracted by the CNNs were fused through the multi-modal integration module and further combined with the multiple morphological metrics extracted by FreeSurfer. Finally, the multilayer perceptron classifier was utilized for subject-level MCI recognition.

Results: The proposed model achieved accuracy of 77.4% to differentiate pMCI from sMCI with 5-fold cross validation on the entire ADNI database. Further, ADNI-1&2 were formed into an independent sample for model training and validation, and ADNI-3&GO were formed into another independent sample for multi-center testing. The model achieved 73.2% accuracy in distinguishing pMCI and sMCI on ADNI-1&2 and 75% accuracy on ADNI-3&GO.

Conclusions: An effective m2ROI-FN model to distinguish pMCI from sMCI was proposed, which was capable of capturing distinctive features in ROIs of sMRI and PET images. The experimental results demonstrated that the model has the potential to differentiate pMCI from sMCI.

查看原文本刊更多论文

基于多模态和多阶段兴趣区融合网络的卷积神经网络模型，用于区分进行性轻度认知障碍和稳定型轻度认知障碍。

背景：准确区分稳定型轻度认知障碍（sMCI）和进行性MCI（pMCI）具有临床意义，而识别pMCI对于在其演变为阿尔茨海默病（AD）之前及时治疗至关重要：构建一个卷积神经网络（CNN）模型，结合结构性磁共振成像（sMRI）和正电子发射断层扫描（PET）图像的特征来区分 pMCI 和 sMCI：我们提出了一种基于感兴趣区（ROI）的多模态、多阶段融合网络（m2ROI-FN）CNN模型来区分pMCI和sMCI，该模型采用多阶段融合策略来整合sMRI和PET图像ROI的深度语义特征和多种形态学指标。具体来说，在每种模式的图像中选择十个与 AD 相关的 ROI 作为补丁输入三维分层 CNN。CNN 提取的深层语义特征通过多模态整合模块进行融合，并进一步与 FreeSurfer 提取的多种形态指标相结合。最后，利用多层感知器分类器进行受试者级别的 MCI 识别：结果：通过对整个 ADNI 数据库进行 5 倍交叉验证，所提出的模型在区分 pMCI 和 sMCI 方面达到了 77.4% 的准确率。此外，ADNI-1 和 ADNI-2 组成一个独立样本进行模型训练和验证，ADNI-3 和 ADNIGO 组成另一个独立样本进行多中心测试。该模型在ADNI-1&2上区分pMCI和sMCI的准确率为73.2%，在ADNI-3&GO上区分pMCI和sMCI的准确率为75%：本文提出了一种有效的 m2ROI-FN 模型来区分 pMCI 和 sMCI，该模型能够捕捉 sMRI 和 PET 图像 ROI 中的显著特征。实验结果表明，该模型具有区分 pMCI 和 sMCI 的潜力。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Alzheimer's Disease 医学-神经科学

CiteScore

6.40

自引率

7.50%

发文量

1327

审稿时长

2 months

期刊介绍： The Journal of Alzheimer''s Disease (JAD) is an international multidisciplinary journal to facilitate progress in understanding the etiology, pathogenesis, epidemiology, genetics, behavior, treatment and psychology of Alzheimer''s disease. The journal publishes research reports, reviews, short communications, hypotheses, ethics reviews, book reviews, and letters-to-the-editor. The journal is dedicated to providing an open forum for original research that will expedite our fundamental understanding of Alzheimer''s disease.