Publications by authors named "Huanzi Liu"
Article Synopsis
- Traditional methods struggle with classifying short-time window steady-state visual evoked potential (SSVEP) signals in brain-computer interface (BCI) systems.
- The proposed CBAM-CNN method utilizes multi-subfrequency bands and a convolutional block attention module for better feature extraction and fusion.
- Experimental results demonstrate that CBAM-CNN achieves a peak accuracy of 0.9813 percentage points and outperforms other methods, especially in the short-time window, significantly enhancing information transmission rates.
Article Synopsis
- The paper discusses advancements in Brain-Computer Interface (BCI) technology, specifically focusing on improving accuracy in motor imagery (MI) classification using EEG signals by extracting meaningful brain network features.
- It introduces a new method combining directed transfer function (DTF) and graph theory with traditional common spatial pattern (CSP) techniques, filtering out redundant features using the Lasso method for better classification outcomes.
- The results show that the proposed method, termed CDGL, significantly outperformed traditional CSP methods with higher accuracy, sensitivity, and specificity, especially using 8 EEG electrodes in the Beta frequency band.
A Fused Multidimensional EEG Classification Method Based on an Extreme Tree Feature Selection.
Comput Intell Neurosci
August 2022
When a brain-computer interface (BCI) is designed, high classification accuracy is difficult to obtain for motor imagery (MI) electroencephalogram (EEG) signals in view of their relatively low signal-to-noise ratio. In this paper, a fused multidimensional classification method based on extreme tree feature selection (FMCM-ETFS) is proposed for discerning motor imagery EEG tasks. First, the EEG signal was filtered by a Butterworth filter for preprocessing.
View Article and Find Full Text PDF