WCWeidong Chen

Weidong Chen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Weidong Chen has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Cortical and Subcortical Articulatory Representations of Two Languages Revealed by Speech Decoding Based on sEEG

    by Yunying Wu, Chenyang Li, Haiyan Zhang, Feiyan Chen, Weidong Chen, Shaomin Zhang

    Numerous studies have suggested that the sensorimotor area may play a role in speech decoding due to its involvement in controlling the movement of vocal organs and its close link to speech. However, due to limitations in surface neural recording methods such as EEG and ECoG, it remains uncertain whether deep brain regions are capable of speech decoding. In this study, we utilized two sets of stereo electroencephalography (sEEG) data to investigate how speech is represented in both cortical and subcortical areas when people speak two different languages. The first sEEG dataset involved 10 subjects speaking Dutch, while the second dataset involved 5 subjects speaking Chinese. Consistent with previous studies, specific brain gyrus locations were closely linked to speech. We also discovered that high decoding performance could be achieved in many deep regions when subjects speak two languages. Furthermore, despite the two languages sharing similar articulation representation, they exhibited differences in neural representation, particularly in terms of high and low-frequency differences of audio. The differences of Chinese speakers were greater than those of Dutch speakers. We anticipate that future studies analyzing more data will uncover more precise articulatory representations of cortical and subcortical structures.

  • Open Access

    Transfer Learning based Efficient Schizophrenia Classification on Electroencephalogram (EEG) Signals: A Cross-Dataset Study

    by Tengjun Liu, Yaoyun Zhang, Yunying Wu, Tuoru Li, Zihao Li, Weidong Chen, Shaomin Zhang

    Although EEG classification for schizophrenia has shown promising results on individual datasets, the cross-dataset generalizability of such classification remains unknown. This study aimed to assess this generalizability through transfer learning at segment and individual levels, by employing the spectral convolutional neural network (CNN-S) on two distinct EEG datasets for schizophrenia classification. While direct cross-decoding only obtained baseline transfer accuracies of 54.72% ± 2.77% and 47.78% ± 3.62% at the segment and individual levels, the fine-tuned CNN-S achieved average cross-dataset transfer accuracies of 76.46% ± 1.17% and 56.71% ± 5.76, respectively. To improve the limited generalizability at the individual level, we applied transfer component analysis (TCA), a domain adaptation approach, to the two datasets, leading to an average cross-dataset transfer accuracy of 62.12% ± 4.86%. By combining fine-tuning and TCA, the study obtained an average cross-dataset transfer accuracy of 69.15% ± 4.09% at the individual level. Overall, transfer learning proves useful for cross-dataset EEG schizophrenia classification.