基于EEG的机器学习流程用于痴呆分类的防泄漏经验基准测试
Leakage-Safe Empirical Benchmarking of EEG-Based Machine Learning Pipelines for Dementia Classification
查看机构详情
- Curtin University（科廷大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本文提出防泄漏基准测试，评估EEG痴呆分类流程，确定最佳实践为ASR+ICA、10秒时段、多域特征及线性SVM，在AD vs CN分类中达到87.69%准确率。
中文摘要 AI 辅助
脑电图（EEG）是一种低成本、非侵入性的痴呆筛查信号源，然而现有的基于EEG的研究仍难以比较，因为预处理、EEG分段、特征设计、分类器选择以及验证协议在不同研究间存在差异，且常常被孤立评估。这种变异性限制了稳健流程推荐的推导。本文提出了一种针对静息态EEG痴呆分类的防泄漏经验基准，并利用该基准确定了一个实用的最佳实践流程。利用公开的OpenNeuro ds004504数据集，该基准评估了伪迹校正、固定长度EEG分段、仅训练集增强、多域特征提取、折叠内特征选择、经典机器学习（ML）分类器、受试者级聚合以及在留一受试者（LOSO）验证下的解释性。该基准中表现最佳的流程结合了伪迹子空间重建（ASR）后接独立成分分析（ICA）、10秒EEG时段（仅训练集幅度缩放和高斯噪声增强）、频谱、复杂性和成对连接性特征、互信息前100选择、线性支持向量机分类以及平均概率聚合。在阿尔茨海默病（AD）与认知正常对照（CN）分类中，该流程达到了87.69%的准确率、88.89%的F1分数和91.20%的AUC，同时从1596个原始描述符中保留了100个特征。结果支持紧凑的多域EEG特征作为受试者级痴呆分类的可解释且防泄漏的基线。
英文摘要
Electroencephalography (EEG) is a low-cost and non-invasive signal source for dementia screening, yet existing EEG-based studies remain difficult to compare because preprocessing, EEG segmentation, feature design, classifier choice, and validation protocols vary across studies and are often evaluated in isolation. This variability limits the derivation of robust pipeline recommendations. This paper presents a leakage-safe empirical benchmark for resting-state EEG-based dementia classification and uses it to identify a practical best-practice pipeline. Using the public OpenNeuro ds004504 dataset, the benchmark evaluates artifact correction, fixed-length EEG segmentation, training-only augmentation, multi-domain feature extraction, fold-internal feature selection, classical machine-learning (ML) classifiers, subject-level aggregation, and interpretation under leave-one-subject-out (LOSO) validation. The best-performing pipeline in this benchmark combines Artifact Subspace Reconstruction (ASR) followed by Independent Component Analysis (ICA), 10 s EEG epochs with training-only amplitude scaling and Gaussian-noise augmentation, spectral, complexity, and pairwise connectivity features, mutual-information top-100 selection, linear support vector machine classification, and mean-probability aggregation. It achieves 87.69 percent accuracy, 88.89 percent F1-score, and 91.20 percent AUC on Alzheimer disease (AD) versus cognitively normal controls (CN) classification while retaining 100 features from 1596 raw descriptors. The results support compact multi-domain EEG features as an interpretable and leakage-safe baseline for subject-level dementia classification.
