Transfer Learning from ImageNet for MEG-Based Decoding of Imagined Speech
This addresses the problem of limited labeled data and weak signals in non-invasive brain-computer interfaces for imagined speech, representing an incremental advance by adapting existing vision models to a new domain.
The paper tackled the challenge of decoding imagined speech from MEG signals by transforming them into image-like inputs for pretrained vision models, achieving up to 90.4% balanced accuracy for imagery vs. silence and 60.6% for vowel decoding.
Non-invasive decoding of imagined speech remains challenging due to weak, distributed signals and limited labeled data. Our paper introduces an image-based approach that transforms magnetoencephalography (MEG) signals into time-frequency representations compatible with pretrained vision models. MEG data from 21 participants performing imagined speech tasks were projected into three spatial scalogram mixtures via a learnable sensor-space convolution, producing compact image-like inputs for ImageNet-pretrained vision architectures. These models outperformed classical and non-pretrained models, achieving up to 90.4% balanced accuracy for imagery vs. silence, 81.0% vs. silent reading, and 60.6% for vowel decoding. Cross-subject evaluation confirmed that pretrained models capture shared neural representations, and temporal analyses localized discriminative information to imagery-locked intervals. These findings show that pretrained vision models applied to image-based MEG representations can effectively capture the structure of imagined speech in non-invasive neural signals.