[18F]基于FDG pet的个体代谢放射组学网络预测多系统萎缩认知功能障碍的新方法

IF 4.5 2区医学 Q1 NEUROIMAGING

NeuroImage Pub Date : 2025-08-25 DOI:10.1016/j.neuroimage.2025.121433

Daoyan Hu , Xiaofeng Dou , Jing Wang , Chentao Jin , Ke Liu , Rui Zhou , Xiaohui Zhang , Congcong Yu , Yan Zhong , Mei Tian , Hong Zhang

{"title":"[18F]基于FDG pet的个体代谢放射组学网络预测多系统萎缩认知功能障碍的新方法","authors":"Daoyan Hu , Xiaofeng Dou , Jing Wang , Chentao Jin , Ke Liu , Rui Zhou , Xiaohui Zhang , Congcong Yu , Yan Zhong , Mei Tian , Hong Zhang","doi":"10.1016/j.neuroimage.2025.121433","DOIUrl":null,"url":null,"abstract":"<div><h3>Purpose</h3><div>Many efforts have been tried to evaluate multiple system atrophy (MSA)-related cognitive impairment, however, there is still lacking of effective approach. In this study, for the first time, we developed the individual metabolic radiomics networks (IMRN) using [<sup>18</sup>F]FDG PET imaging to investigate brain metabolic connectivity patterns of MSA and validated the usefulness of IMRN-based predictive model for MSA-related cognitive impairment.</div></div><div><h3>Methods</h3><div>In this retrospective study, we recruited 115 MSA patients with [<sup>18</sup>F]FDG PET/CT scans. IMRN was constructed by extracting non-redundant radiomics features from each brain region and computing pairwise Pearson correlation coefficients among these features. The validation of IMRN included assessments of small-world properties, test-retest reliability, and metabolic-genetic correlations. Connectome-based predictive modeling (CPM) was implemented to predict Mini Mental State Examination (MMSE) scores, while network-based statistics (NBS) were compared between MSA patients with cognitive impairment (MSA-CI, <em>n</em> = 58; MMSE < 27) and those with normal cognition (MSA-NC, <em>n</em> = 57; MMSE ≥ 27). A support vector machine (SVM) classifier for detecting MSA-CI was developed using discriminative IMRN edges.</div></div><div><h3>Results</h3><div>IMRN showed small-world properties (σ > 1), high reliability (average edge ICC = 0.754), and a significant correlation with gene expression (<em>r</em> = 0.44, <em>P</em> < 0.001). CPM significantly predicted cognitive scores through IMRN edges (positive network: <em>r</em> = 0.27, <em>P</em> = 0.03; negative network: <em>r</em> = 0.28, <em>P</em> = 0.02). NBS revealed decreased cerebellar-cortical connectivity (73 edges) and increased intra-cerebellar/limbic connectivity (24 edges) in MSA-CI compared to MSA-NC. The IMRN-based SVM outperformed SUVR-based SVM in classifying MSA-CI (accuracy: 73.91% vs 62.61%; AUC: 0.80 vs 0.69).</div></div><div><h3>Conclusion</h3><div>This study established a novel approach of IMRN for assessing whole brain metabolic connectivity, uncovering distinct cerebellar connectivity patterns in MSA-CI, which held promise for facilitating personalized cognitive evaluations in MSA.</div></div>","PeriodicalId":19299,"journal":{"name":"NeuroImage","volume":"319 ","pages":"Article 121433"},"PeriodicalIF":4.5000,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A novel approach of [18F]FDG PET-based individual metabolic radiomics network to predict cognitive impairment in multiple system atrophy\",\"authors\":\"Daoyan Hu , Xiaofeng Dou , Jing Wang , Chentao Jin , Ke Liu , Rui Zhou , Xiaohui Zhang , Congcong Yu , Yan Zhong , Mei Tian , Hong Zhang\",\"doi\":\"10.1016/j.neuroimage.2025.121433\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Purpose</h3><div>Many efforts have been tried to evaluate multiple system atrophy (MSA)-related cognitive impairment, however, there is still lacking of effective approach. In this study, for the first time, we developed the individual metabolic radiomics networks (IMRN) using [<sup>18</sup>F]FDG PET imaging to investigate brain metabolic connectivity patterns of MSA and validated the usefulness of IMRN-based predictive model for MSA-related cognitive impairment.</div></div><div><h3>Methods</h3><div>In this retrospective study, we recruited 115 MSA patients with [<sup>18</sup>F]FDG PET/CT scans. IMRN was constructed by extracting non-redundant radiomics features from each brain region and computing pairwise Pearson correlation coefficients among these features. The validation of IMRN included assessments of small-world properties, test-retest reliability, and metabolic-genetic correlations. Connectome-based predictive modeling (CPM) was implemented to predict Mini Mental State Examination (MMSE) scores, while network-based statistics (NBS) were compared between MSA patients with cognitive impairment (MSA-CI, <em>n</em> = 58; MMSE < 27) and those with normal cognition (MSA-NC, <em>n</em> = 57; MMSE ≥ 27). A support vector machine (SVM) classifier for detecting MSA-CI was developed using discriminative IMRN edges.</div></div><div><h3>Results</h3><div>IMRN showed small-world properties (σ > 1), high reliability (average edge ICC = 0.754), and a significant correlation with gene expression (<em>r</em> = 0.44, <em>P</em> < 0.001). CPM significantly predicted cognitive scores through IMRN edges (positive network: <em>r</em> = 0.27, <em>P</em> = 0.03; negative network: <em>r</em> = 0.28, <em>P</em> = 0.02). NBS revealed decreased cerebellar-cortical connectivity (73 edges) and increased intra-cerebellar/limbic connectivity (24 edges) in MSA-CI compared to MSA-NC. The IMRN-based SVM outperformed SUVR-based SVM in classifying MSA-CI (accuracy: 73.91% vs 62.61%; AUC: 0.80 vs 0.69).</div></div><div><h3>Conclusion</h3><div>This study established a novel approach of IMRN for assessing whole brain metabolic connectivity, uncovering distinct cerebellar connectivity patterns in MSA-CI, which held promise for facilitating personalized cognitive evaluations in MSA.</div></div>\",\"PeriodicalId\":19299,\"journal\":{\"name\":\"NeuroImage\",\"volume\":\"319 \",\"pages\":\"Article 121433\"},\"PeriodicalIF\":4.5000,\"publicationDate\":\"2025-08-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"NeuroImage\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1053811925004367\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"NEUROIMAGING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"NeuroImage","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1053811925004367","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"NEUROIMAGING","Score":null,"Total":0}

引用次数: 0

摘要

目的对多系统萎缩（MSA）相关的认知功能障碍进行了评价，但缺乏有效的评价方法。在本研究中，我们首次利用[18F]FDG PET成像建立了个体代谢放射组学网络（IMRN）来研究MSA的脑代谢连接模式，并验证了基于IMRN的MSA相关认知障碍预测模型的有效性。方法在本回顾性研究中，我们招募了115例MSA患者进行[18F]FDG PET/CT扫描。IMRN通过从每个脑区提取非冗余放射组学特征并计算这些特征之间的成对Pearson相关系数来构建。IMRN的验证包括对小世界特性、重测信度和代谢-遗传相关性的评估。采用基于连接体的预测模型（CPM）预测迷你精神状态检查（MMSE）评分，并比较认知功能障碍的MSA患者（MSA- ci, n = 58; MMSE < 27）与认知功能正常的MSA患者（MSA- nc, n = 57; MMSE≥27）之间基于网络的统计（NBS）。提出了一种基于判别IMRN边缘的MSA-CI支持向量机分类器。结果mrn具有小世界特性（σ > 1），高信度（平均边缘ICC = 0.754），与基因表达显著相关（r = 0.44, P < 0.001）。CPM通过IMRN边缘显著预测认知得分（正网络：r = 0.27, P = 0.03；负网络：r = 0.28, P = 0.02）。NBS显示，与MSA-NC相比，MSA-CI的小脑-皮质连通性降低（73条边），小脑内/边缘连通性增加（24条边）。基于imrn的SVM在MSA-CI分类上优于基于sur的SVM（准确率：73.91% vs 62.61%; AUC: 0.80 vs 0.69）。本研究建立了一种新的IMRN方法来评估全脑代谢连通性，揭示了MSA- ci中不同的小脑连通性模式，为MSA的个性化认知评估提供了希望。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A novel approach of [18F]FDG PET-based individual metabolic radiomics network to predict cognitive impairment in multiple system atrophy

Purpose

Many efforts have been tried to evaluate multiple system atrophy (MSA)-related cognitive impairment, however, there is still lacking of effective approach. In this study, for the first time, we developed the individual metabolic radiomics networks (IMRN) using [¹⁸F]FDG PET imaging to investigate brain metabolic connectivity patterns of MSA and validated the usefulness of IMRN-based predictive model for MSA-related cognitive impairment.

Methods

In this retrospective study, we recruited 115 MSA patients with [¹⁸F]FDG PET/CT scans. IMRN was constructed by extracting non-redundant radiomics features from each brain region and computing pairwise Pearson correlation coefficients among these features. The validation of IMRN included assessments of small-world properties, test-retest reliability, and metabolic-genetic correlations. Connectome-based predictive modeling (CPM) was implemented to predict Mini Mental State Examination (MMSE) scores, while network-based statistics (NBS) were compared between MSA patients with cognitive impairment (MSA-CI, n = 58; MMSE < 27) and those with normal cognition (MSA-NC, n = 57; MMSE ≥ 27). A support vector machine (SVM) classifier for detecting MSA-CI was developed using discriminative IMRN edges.

Results

IMRN showed small-world properties (σ > 1), high reliability (average edge ICC = 0.754), and a significant correlation with gene expression (r = 0.44, P < 0.001). CPM significantly predicted cognitive scores through IMRN edges (positive network: r = 0.27, P = 0.03; negative network: r = 0.28, P = 0.02). NBS revealed decreased cerebellar-cortical connectivity (73 edges) and increased intra-cerebellar/limbic connectivity (24 edges) in MSA-CI compared to MSA-NC. The IMRN-based SVM outperformed SUVR-based SVM in classifying MSA-CI (accuracy: 73.91% vs 62.61%; AUC: 0.80 vs 0.69).

Conclusion

This study established a novel approach of IMRN for assessing whole brain metabolic connectivity, uncovering distinct cerebellar connectivity patterns in MSA-CI, which held promise for facilitating personalized cognitive evaluations in MSA.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

NeuroImage 医学-核医学

CiteScore

11.30

自引率

10.50%

发文量

809

审稿时长

63 days

期刊介绍： NeuroImage, a Journal of Brain Function provides a vehicle for communicating important advances in acquiring, analyzing, and modelling neuroimaging data and in applying these techniques to the study of structure-function and brain-behavior relationships. Though the emphasis is on the macroscopic level of human brain organization, meso-and microscopic neuroimaging across all species will be considered if informative for understanding the aforementioned relationships.