# Transfer Learning from ImageNet for MEG-Based Decoding of Imagined Speech

arXiv: https://arxiv.org/abs/2601.15909 · HTML: https://mlsanity.com/paper/2601.15909/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 17 references that are in the map.
- First posted on arXiv 22 Jan 2026.

## Builds on (most cited first, 6 of 6)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [EEGNet: A Compact Convolutional Network for EEG-based Brain-Computer Interfaces](https://mlsanity.com/paper/1611.08024/) (2016)
- [Patches Are All You Need?](https://mlsanity.com/paper/2201.09792/) (2022)
- [Decoding speech perception from non-invasive brain recordings](https://mlsanity.com/paper/2208.12266/) (2022)
- [Brain-to-Text Decoding: A Non-invasive Approach via Typing](https://mlsanity.com/paper/2502.17480/) (2025)

## Built on by (most cited first, 1 of 1)

- [MEG-to-MEG Transfer Learning and Cross-Task Speech/Silence Detection with Limited Data](https://mlsanity.com/paper/2602.18253/) (2026)

## Abstract (arXiv)

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.