NeuroPriv：面向可穿戴脑电图系统隐私的对抗性表示学习

中文摘要 AI 辅助

可穿戴脑电图（EEG）系统除了预期的健康功能外，还可能泄露敏感信息，对神经隐私造成重大风险。本研究表明，除支持预期认知任务外，常用的EEG特征还可揭示参与者身份和人口统计属性。可穿戴EEG正越来越多地用于认知监测、神经评估和纵向数字健康应用，但许多系统认为传输紧凑的频谱或空间特征而非原始EEG就足以提供隐私保护。以EEGMAT为案例研究，发现紧凑EEG特征在认知状态分类中达到0.788的平衡准确率，同时实现性别、年龄和受试者身份推理的平衡准确率分别为0.858、0.789和0.692。进一步研究显示，隐私感知表示学习在保持任务性能达0.781的同时，将这些推理准确率降至0.563、0.467和0.206。这些发现推动了可穿戴神经健康系统中目的受限表示和显式隐私审计的发展。

英文摘要

Wearable EEG systems may expose sensitive information beyond their intended health function, creating substantial risks to neuroprivacy. In this work, we show that commonly used EEG features can reveal participant identity and demographic attributes in addition to supporting the intended cognitive task. Wearable EEG is increasingly being explored for cognitive monitoring, neurological assessment, and longitudinal digital-health applications, yet many systems assume that transmitting compact spectral or spatial features instead of raw EEG provides sufficient privacy protection. Using EEGMAT as a motivating case study, we find that compact EEG features achieve a balanced accuracy of 0.788 for cognitive-state classification while enabling gender, age, and subject-identity inference with balanced accuracies of 0.858, 0.789, and 0.692, respectively. We further show that privacy-aware representation learning preserves task performance at 0.781 while reducing these inference accuracies to 0.563, 0.467, and 0.206. These findings motivate purpose-limited representations and explicit privacy auditing in wearable neurohealth systems.

发表机构

• University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校）

机构由 AI 辅助整理，请以论文原文为准。