文献类型：期刊文献

英文题名：Hands-free motor imagery EEG classification via LLM multi-agents

作者：Zhao, Ruiyu[1];Li, Shurui[2];He, Xinjie[1];Wang, Xingyu[1];Cichocki, Andrzej[3,4,5];Lu, Yunhe[6];Jin, Jing[1,2]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Sch Math, Shanghai 200237, Peoples R China;[3]RIKEN Brain Sci Inst, Lab Adv Brain Signal Proc, Wako 3510198, Japan;[4]Polish Acad Sci, Syst Res Inst, PL-01447 Warsaw, Poland;[5]Nicolaus Copernicus Univ, Dept Informat, PL-87100 Torun, Poland;[6]Shanghai Lansheng Brain Hosp Investment Co Ltd, Shanghai 200336, Peoples R China

年份：2026

卷号：436

外文期刊名：JOURNAL OF NEUROSCIENCE METHODS

收录：;WOS:【SCI-EXPANDED(收录号:WOS:001848606200001)】；

基金：This work was supported by Brain Science and Brain-like Intelligence Technology-National Science and Technology Major Project 2022ZD0208900, in part by key core technologies under Grant BE202 2064-1; This research is also supported by the Lingang Laboratory under Grant No. LGL8998.

语种：英文

外文关键词：Motor imagery; Brain-computer interface; MI-EEG classification; Multi-agent system; Automatic model optimization

摘要：Background: Motor imagery (MI) brain-computer interfaces (BCI) rely on precise electroencephalogram (EEG) classification. However, issues such as the reliance on extensive manual experience for MI-EEG model design, parameter tuning, and optimization directions, along with the poor task flexibility of foundation models and state degradation during long-term multi-agent iterations, severely restrict the state-of-the-art (SOTA) efficiency of MI-EEG. New method: To address these challenges, we propose AutoMI, a novel framework that uses multi-agent automated rapid iterations to construct SOTA MI-EEG models. AutoMI introduces a hybrid decision mechanism that tightly couples Q-learning strategies with deterministic rules. By integrating planning, execution, and output agents with predefined tools, AutoMI ensures broad general applicability across various hyperparameter optimizations and structural improvements. Furthermore, AutoMI integrates experience tracking and rollback mechanisms to prevent ambiguous optimization. Results: In evaluations on the IV2a, OpenBMI, and ECUST-MI datasets, the SOTA models finally constructed through AutoMI iterations achieve accuracies of 77.62%, 78.08%, and 83.02%, with maximum improvement reaching 24.69%, 23.35%, and 23.28% respectively. Furthermore, the average time per iteration for a single subject on the OpenBMI dataset is approximately 500 s. Comparison with existing methods: Compared with automated optimization algorithms, the accuracies increase by 18.42%, 9.27%, and 19.25% respectively, demonstrating the effectiveness of the proposed AutoMI framework and proving that its optimization capability reaches SOTA. Conclusion: Experimental results indicate that AutoMI provides a novel perspective and framework design reference for future BCI model optimization.