Deep learning models have shown promise in EEG-based outcome prediction for comatose patients after cardiac arrest, but their reliability is often compromised by subtle forms of data leakage. In particular, when long EEG recordings are segmented into short windows and reused across multiple training stages, models may implicitly encode and propagate label information, leading to overly optimistic validation performance and poor generalization.
In this study, we identify a previously overlooked form of data leakage in multi-stage EEG modeling pipelines. We demonstrate that violating strict patient-level separation can significantly inflate validation metrics while causing substantial degradation on independent test data.
To address this issue, we propose a leakage-aware two-stage framework. In the first stage, short EEG segments are transformed into embedding representations using a convolutional neural network with an ArcFace objective. In the second stage, a Transformer-based model aggregates these embeddings to produce patient-level predictions, with strict isolation between training cohorts to eliminate leakage pathways.
Experiments on a large-scale EEG dataset of post-cardiac-arrest patients show that the proposed framework achieves stable and generalizable performance under clinically relevant constraints, particularly in maintaining high sensitivity at stringent specificity thresholds. These results highlight the importance of rigorous data partitioning and provide a practical solution for reliable EEG-based outcome prediction.
Cueless EEG imagined speech for subject identification: dataset and benchmarks
Department of Computer Engineering, Sharif University of Technology(沙斐大学计算机工程系);Center for Cognitive Science, Institute for Convergence Science and Technology (ICST)(融合科学与技术研究所认知科学中心);Sharif University of Technology(沙斐大学)
One Model for All: Universal Pre-training for EEG based Emotion Recognition across Heterogeneous Datasets and Paradigms
Key Laboratory of Computing Power Network and Information Security, Ministry of Education, Shandong Computer Science Center (National Supercomputer Center inJinan), Qilu University of Technology (Shandong Academy of Sciences), Jinan, China and Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing, Shandong Fundamental Research Center for Computer Science, Jinan, China(教育部计算功率网络与信息安全重点实验室,山东计算机科学中心(济南国家超算中心),青岛科技大学(山东省科学院),济南,中国和山东省计算功率互联网与服务计算省级重点实验室,山东省计算机科学基础研究中心,济南,中国);College of Intelligence and Computing, Tianjin University(天津大学智能与计算学院)
AntiLeakBench: Preventing Data Contamination by Automatically Constructing Benchmarks with Updated Real-World Knowledge
Nanyang Technological University(南洋理工大学);University of California, Santa Barbara(加州大学圣芭芭拉分校);National University of Singapore(新加坡国立大学);Shanghai Jiao Tong University(上海交通大学);University of Arizona(亚利桑那大学)
Towards a more realistic evaluation of machine learning models for bearing fault diagnosis
Department of Electrical and Electronic Engineering, Federal University of Santa Catarina(电气与电子工程系,圣卡塔琳娜联邦大学);Department of Mechanical Engineering, Federal University of Santa Catarina(机械工程系,圣卡塔琳娜联邦大学)
The role of data partitioning on the performance of EEG-based deep learning models in supervised cross-subject analysis: a preliminary study
University of Padua(帕多瓦大学);Padova Neuroscience Center, University of Padua(帕多瓦大学帕多瓦神经科学中心);University of Applied Sciences Western Switzerland (HES-SO Valais)(瑞士西部应用科学与艺术大学(瓦莱州校区))
Exploiting Multiple EEG Data Domains with Adversarial Learning
Dr. Ing. h.c. F. Porsche AG(保时捷股份公司);Ludwig-Maximilians University(路德维希-马克西米利安大学);Karlsruhe Institute of Technology(卡尔斯鲁厄理工学院);TU Graz(格拉茨技术大学);Silicon Austria Labs(奥地利硅实验室)
EEG-to-fMRI synthesis of task-evoked and spontaneous brain activity: addressing issues of statistical significance and generalizability
Institute for Systems and Robotics – Lisboa and Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Portugal(里斯本系统与机器人研究所及生物工程系,里斯本技术学院,里斯本大学,葡萄牙);Electrical and Computer Engineering Department, Carnegie Mellon University(电气与计算机工程系,卡内基梅隆大学)
随着如今数据的规模和大小显著增加,用于高性能计算和机器学习的联邦学习(Bonawitz et al.[2019])比以往任何时候都更加重要(Abadi et al.[2016])。人们过去认为,共享梯度似乎是安全的,以隐藏在训练阶段的本地训练数据。然而,Zhu等人[2019]证明了可以通过检测梯度从模型训练数据中恢复原始数据。它们使用生成的随机虚拟数据,并最小化它们与真实数据之间的距离。Zhao et al.[2020]进一步推动了收敛算法。通过用交叉熵损失代替原有的损失函数,获得了更好的保真度阈值。在本文中,我们建议在将数据馈送到分类器之前使用额外的dropout(Srivastava et al.[2014])层。它在防止原始数据泄漏方面非常有效,因为即使在5,800个epoch之后,训练数据也不能收敛到小的RMSE,其中丢弃率设置为0.5。
Mjolnir: Breaking the Shield of Perturbation-Protected Gradients via Adaptive Diffusion
The Hong Kong Polytechnic University(香港理工大学);Wuhan University of Technology(武汉理工大学);Shenzhen University(深圳大学);The Hong Kong University of Science and Technology(香港科技大学)