MDavid19
@mdavid19.bsky.social
M. Sc. in Electrical Engineering
Biomedical Signal Processing, EEG & BCI
ML/DL, Computer Vision & AI in Healthcare & Medical Imaging
📍 Yazd, Iran
Biomedical Signal Processing, EEG & BCI
ML/DL, Computer Vision & AI in Healthcare & Medical Imaging
📍 Yazd, Iran
🔗 (7/7)
In summary:
🧠 Interpretable brain graphs from EEG
🌐 SWSF-inspired brain network structure
⚙️ Better decoding & identification
#brainGSP
Special thanks to all the co-authors and honored to work with @aitchbi.bsky.social and @andreasantoro.bsky.social.
📄 Read more: biorxiv.org/content/10.1...
In summary:
🧠 Interpretable brain graphs from EEG
🌐 SWSF-inspired brain network structure
⚙️ Better decoding & identification
#brainGSP
Special thanks to all the co-authors and honored to work with @aitchbi.bsky.social and @andreasantoro.bsky.social.
📄 Read more: biorxiv.org/content/10.1...
Small-world scale-free brain graphs from EEG
Developing individualized spatial models that capture the complex dynamics of multi-electrode EEG data is essential for accurately decoding global neural activity. A widely used approach is network mo...
biorxiv.org
May 18, 2025 at 7:36 AM
🔗 (7/7)
In summary:
🧠 Interpretable brain graphs from EEG
🌐 SWSF-inspired brain network structure
⚙️ Better decoding & identification
#brainGSP
Special thanks to all the co-authors and honored to work with @aitchbi.bsky.social and @andreasantoro.bsky.social.
📄 Read more: biorxiv.org/content/10.1...
In summary:
🧠 Interpretable brain graphs from EEG
🌐 SWSF-inspired brain network structure
⚙️ Better decoding & identification
#brainGSP
Special thanks to all the co-authors and honored to work with @aitchbi.bsky.social and @andreasantoro.bsky.social.
📄 Read more: biorxiv.org/content/10.1...
🔗 (6/7)
Our graph models (SWSF and SWSFKron) remain robust across:
✔️ Different graph rewiring settings
✔️ Different EEG time segments
They outperform classic PLV-based graphs in all frequency bands.
Our graph models (SWSF and SWSFKron) remain robust across:
✔️ Different graph rewiring settings
✔️ Different EEG time segments
They outperform classic PLV-based graphs in all frequency bands.
May 18, 2025 at 7:36 AM
🔗 (6/7)
Our graph models (SWSF and SWSFKron) remain robust across:
✔️ Different graph rewiring settings
✔️ Different EEG time segments
They outperform classic PLV-based graphs in all frequency bands.
Our graph models (SWSF and SWSFKron) remain robust across:
✔️ Different graph rewiring settings
✔️ Different EEG time segments
They outperform classic PLV-based graphs in all frequency bands.
🔗 (5/7)
These brain graphs capture consistent patterns across EEG rhythms — helping us identify individuals from their brain activity across 109 subjects.
These brain graphs capture consistent patterns across EEG rhythms — helping us identify individuals from their brain activity across 109 subjects.
May 18, 2025 at 7:36 AM
🔗 (5/7)
These brain graphs capture consistent patterns across EEG rhythms — helping us identify individuals from their brain activity across 109 subjects.
These brain graphs capture consistent patterns across EEG rhythms — helping us identify individuals from their brain activity across 109 subjects.
🔗 (4/7)
We test our approach in two tasks:
🧠 Motor imagery (Brain-Computer Interface via EEG)
🧠 Brain fingerprinting (who is who from EEG)
Our SWSF graphs help distinguish individuals and mental tasks better than other graph models!
We test our approach in two tasks:
🧠 Motor imagery (Brain-Computer Interface via EEG)
🧠 Brain fingerprinting (who is who from EEG)
Our SWSF graphs help distinguish individuals and mental tasks better than other graph models!
May 18, 2025 at 7:36 AM
🔗 (4/7)
We test our approach in two tasks:
🧠 Motor imagery (Brain-Computer Interface via EEG)
🧠 Brain fingerprinting (who is who from EEG)
Our SWSF graphs help distinguish individuals and mental tasks better than other graph models!
We test our approach in two tasks:
🧠 Motor imagery (Brain-Computer Interface via EEG)
🧠 Brain fingerprinting (who is who from EEG)
Our SWSF graphs help distinguish individuals and mental tasks better than other graph models!
🔗 (3/7)
To make these brain networks compact but still efficient, we apply Kron reduction.
It shrinks the network while preserving important features — producing subject-specific graphs ready for learning tasks.
To make these brain networks compact but still efficient, we apply Kron reduction.
It shrinks the network while preserving important features — producing subject-specific graphs ready for learning tasks.
May 18, 2025 at 7:36 AM
🔗 (3/7)
To make these brain networks compact but still efficient, we apply Kron reduction.
It shrinks the network while preserving important features — producing subject-specific graphs ready for learning tasks.
To make these brain networks compact but still efficient, we apply Kron reduction.
It shrinks the network while preserving important features — producing subject-specific graphs ready for learning tasks.
🔗 (2/7)
We turn raw EEG signals into meaningful brain networks using phase-based connectivity.
Then, we refine them to reflect small-world and scale-free properties — features observed in real brain networks. 🧠🌐
We turn raw EEG signals into meaningful brain networks using phase-based connectivity.
Then, we refine them to reflect small-world and scale-free properties — features observed in real brain networks. 🧠🌐
May 18, 2025 at 7:36 AM
🔗 (2/7)
We turn raw EEG signals into meaningful brain networks using phase-based connectivity.
Then, we refine them to reflect small-world and scale-free properties — features observed in real brain networks. 🧠🌐
We turn raw EEG signals into meaningful brain networks using phase-based connectivity.
Then, we refine them to reflect small-world and scale-free properties — features observed in real brain networks. 🧠🌐