A normative model–based assessment framework for large-scale, multi-site EEG data

IF 3.7 3区医学 Q2 NEUROSCIENCES

Brain Research Bulletin Pub Date : 2025-09-12 DOI:10.1016/j.brainresbull.2025.111546

Qiwei Dong , Yuxi Zhou , Xiaoyu Xiong , Pengyu Liu , Jianfu Li , Cheng Luo , Diankun Gong , Li Dong , Dezhong Yao

{"title":"A normative model–based assessment framework for large-scale, multi-site EEG data","authors":"Qiwei Dong , Yuxi Zhou , Xiaoyu Xiong , Pengyu Liu , Jianfu Li , Cheng Luo , Diankun Gong , Li Dong , Dezhong Yao","doi":"10.1016/j.brainresbull.2025.111546","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Electroencephalography (EEG) overcomes the subjectivity inherent in questionnaire-based and observational assessments. However, most existing EEG-based evaluation methods still impose discrete categorical states onto continuously varying neural dynamics, thereby neglecting the continuity of states. With the rise of neuroscience alliances, challenges such as batch-effects across datasets and inconsistencies introduced by diverse EEG electrode montages have become increasingly prominent. Therefore, a robust assessment framework that accommodates large‑scale, multi‑site EEG data is expected.</div></div><div><h3>Methods</h3><div>A normative model-based assessment framework was developed for large-scale, multi-site EEG data, with attention assessments used as illustrative examples. Normative models are first constructed using EEG features from 1212 young individuals, and quantile ranks are computed. Next, feature selection is performed, and elastic net regression and support vector regression are used to model distributed attention (DA) and focused attention (FA). The results from normative model-based features are compared with original features to demonstrate the advantage of quantile rank features. Finally, the model’s test-retest reliability and generalizability are assessed.</div></div><div><h3>Results</h3><div>The framework identifies statistical differences (<em>q</em> < 0.05) in attention performance between the top and bottom 20 % participants on attention scales. EEG features demonstrated specific patterns related to accuracy and reaction time in both DA and FA tasks. The normative model outperformed in predictive tasks, showing enhanced stability and interpretability. Additionally, the framework demonstrates strong test-retest reliability and robust generalizability (ICC > 0.9).</div></div><div><h3>Conclusion</h3><div>In conclusion, we proposed a normative model–based framework that harmonizes large‑scale, multi‑site EEG data, enabling efficient and reliable attention assessment while demonstrating promise for broader EEG‑based applications.</div></div>","PeriodicalId":9302,"journal":{"name":"Brain Research Bulletin","volume":"231 ","pages":"Article 111546"},"PeriodicalIF":3.7000,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Brain Research Bulletin","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0361923025003582","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

Electroencephalography (EEG) overcomes the subjectivity inherent in questionnaire-based and observational assessments. However, most existing EEG-based evaluation methods still impose discrete categorical states onto continuously varying neural dynamics, thereby neglecting the continuity of states. With the rise of neuroscience alliances, challenges such as batch-effects across datasets and inconsistencies introduced by diverse EEG electrode montages have become increasingly prominent. Therefore, a robust assessment framework that accommodates large‑scale, multi‑site EEG data is expected.

Methods

A normative model-based assessment framework was developed for large-scale, multi-site EEG data, with attention assessments used as illustrative examples. Normative models are first constructed using EEG features from 1212 young individuals, and quantile ranks are computed. Next, feature selection is performed, and elastic net regression and support vector regression are used to model distributed attention (DA) and focused attention (FA). The results from normative model-based features are compared with original features to demonstrate the advantage of quantile rank features. Finally, the model’s test-retest reliability and generalizability are assessed.

Results

The framework identifies statistical differences (q < 0.05) in attention performance between the top and bottom 20 % participants on attention scales. EEG features demonstrated specific patterns related to accuracy and reaction time in both DA and FA tasks. The normative model outperformed in predictive tasks, showing enhanced stability and interpretability. Additionally, the framework demonstrates strong test-retest reliability and robust generalizability (ICC > 0.9).

Conclusion

In conclusion, we proposed a normative model–based framework that harmonizes large‑scale, multi‑site EEG data, enabling efficient and reliable attention assessment while demonstrating promise for broader EEG‑based applications.

查看原文本刊更多论文

一种基于规范模型的大规模多点脑电数据评估框架。

背景：脑电图（EEG）克服了基于问卷和观察性评估固有的主观性。然而，大多数现有的基于脑电图的评估方法仍然将离散的分类状态强加于连续变化的神经动力学上，从而忽略了状态的连续性。随着神经科学联盟的兴起，诸如跨数据集的批量效应和不同EEG电极蒙太奇引入的不一致性等挑战变得越来越突出。因此，一个强大的评估框架，以适应大规模，多地点的脑电图数据是预期的。方法：以注意力评价为例，对大规模、多位点脑电数据建立基于规范模型的评价框架。首先利用1212名年轻人的脑电图特征构建规范模型，并计算分位数秩。接下来，进行特征选择，并使用弹性网络回归和支持向量回归对分布式注意（DA）和集中注意（FA）进行建模。将基于规范模型的特征结果与原始特征进行比较，证明了分位数秩特征的优势。最后，对模型的重测信度和泛化性进行了评价。结果：该框架确定了前20%和后20%参与者在注意表现上的统计学差异（q < 0.05）。脑电特征显示了与DA和FA任务的准确性和反应时间相关的特定模式。规范模型在预测任务中表现出色，表现出增强的稳定性和可解释性。此外，该框架具有较强的重测信度和稳健的泛化能力（ICC > 0.9）。结论：总之，我们提出了一个规范的基于模型的框架，该框架协调了大规模、多地点的脑电图数据，实现了高效、可靠的注意力评估，同时展示了基于脑电图的更广泛应用的前景。

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

Brain Research Bulletin 医学-神经科学

CiteScore

6.90

自引率

2.60%

发文量

253

审稿时长

67 days

期刊介绍： The Brain Research Bulletin (BRB) aims to publish novel work that advances our knowledge of molecular and cellular mechanisms that underlie neural network properties associated with behavior, cognition and other brain functions during neurodevelopment and in the adult. Although clinical research is out of the Journal''s scope, the BRB also aims to publish translation research that provides insight into biological mechanisms and processes associated with neurodegeneration mechanisms, neurological diseases and neuropsychiatric disorders. The Journal is especially interested in research using novel methodologies, such as optogenetics, multielectrode array recordings and life imaging in wild-type and genetically-modified animal models, with the goal to advance our understanding of how neurons, glia and networks function in vivo.