发表机构
University of Pennsylvania; School of Engineering and Applied Science（宾夕法尼亚大学; 工程与应用科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究利用LoRA轻量适配EEG基础模型，发现健康人群模型无法可靠迁移至脑卒中EEG，仅提升模型容量无益，需目标域适配与受试者级评估以构建脑卒中康复BCI
AI 中文摘要
运动想象（MI）脑电图（EEG）解码可用于脑卒中后康复，但在健康人群队列上开发的模型无法可靠迁移至病理EEG。本研究评估低秩适配（LoRA）是否可高效适配3个预训练EEG基础模型（即LaBraM-base、REVE-base和REVE-large），用于左右手MI的二分类解码。采用受试者级五折交叉验证，在PhysioNet EEG运动/想象数据集（EEGMMIDB）和包含30名脑卒中受试者的UET175数据集二分类子集上，评估仅冻结主干的基线模型与LoRA适配模型。在EEGMMIDB上，LoRA使LaBraM-base的准确率提升至0.822，REVE-base提升至0.957；在UET175上，所有仅用主干的模型表现接近随机水平，而LoRA适配后，LaBraM-base仍接近随机水平（0.499±0.009），REVE-base达到0.847±0.194，且优于REVE-large（0.806±0.178），表明仅提升模型容量无法改善脑卒中域的适配性。最优脑卒中适配配置REVE-base LoRA经队列内留一受试者交叉验证（LOOCV）评估，平均准确率为0.952，但受试者级准确率范围为0.586至1.000，存在少量低表现尾部；从EEGMMIDB到UET175的零样本迁移仍接近随机水平（0.464±0.072）。这些发现表明，健康基准的性能无法确保迁移至脑卒中EEG，因此将EEG基础模型转化为伪在线或实时康复BCI时，需包含目标域适配及时间信息性、空间敏感性和生理可区分性的受试者级评估。
英文摘要
Motor imagery (MI) electroencephalography (EEG) decoding could support post-stroke rehabilitation, but models developed on healthy cohorts may not transfer reliably to pathological EEG. We evaluated whether Low-Rank Adaptation (LoRA) can efficiently adapt three pretrained EEG foundation models (i.e., LaBraM-base, REVE-base, and REVE-large) for binary left- versus right-hand MI decoding. Frozen-backbone head-only baselines and LoRA adaptation were evaluated using subject-wise five-fold cross-validation on the PhysioNet EEG Motor Movement/Imagery Dataset and a binary subset of the UET175 dataset comprising 30 stroke participants. On EEGMMIDB, LoRA increased accuracy to 0.822 for LaBraM-base and 0.957 for REVE-base. On UET175, all head-only models performed near chance. With LoRA, LaBraM-base remained near chance (0.499$\pm$0.009), whereas REVE-base reached 0.847$\pm$0.194 and outperformed REVE-large (0.806$\pm$0.178), indicating that increased model capacity alone did not improve stroke-domain adaptation. The strongest stroke configuration, REVE-base LoRA, was further evaluated using within-cohort leave-one-subject-out cross-validation (LOOCV), showing 0.952 mean accuracy, but subject-wise accuracy ranged from 0.586 to 1.000, revealing a small low-performing tail. Zero-shot transfer from EEGMMIDB to UET175 remained near chance (0.464$\pm$0.072). These findings show that healthy-benchmark performance does not ensure transfer to stroke EEG. Translation of EEG foundation models to pseudo-online or real-time rehabilitation BCIs should therefore include target-domain adaptation and subject-level assessment of temporal informativeness, spatial sensitivity, and physiological discriminability.
