Eeg diffuse source imaging method based on autoencoder and spatial difference sparse constraint
By employing an EEG diffusion source imaging method based on autoencoders and spatial difference sparse constraints, the problems of real-time EEG source imaging and brain source size reconstruction are solved, achieving more efficient brain source imaging.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-10-30
- Publication Date
- 2026-07-24
AI Technical Summary
Existing EEG source imaging methods have poor real-time performance, traditional methods require complex prior assumptions and are computationally complex, and deep learning methods cannot accurately reconstruct brain source size information.
An EEG diffusion source imaging method based on autoencoder and spatial difference sparse constraints is adopted. Through autoencoder training and loss function optimization, combined with the temporal and spatial information of EEG data and brain source signals, brain source data is generated using boundary element method and region growing method, and the loss function is optimized to reconstruct brain source signals.
It improves the real-time performance and accuracy of EEG source imaging, enabling better reconstruction of the size and location information of brain sources, and is suitable for brain science research and diagnosis of neurological diseases.
Smart Images

Figure CN117243614B_ABST