The End Of Knowledge - Vault 3 - GTEC BCI & Neurotechnology Spring School 2024 - Day 1

graph LR classDef school fill:#f9d4d4, font-weight:bold, font-size:14px; classDef gtec fill:#d4f9d4, font-weight:bold, font-size:14px; classDef eeg fill:#d4d4f9, font-weight:bold, font-size:14px; classDef bci fill:#f9f9d4, font-weight:bold, font-size:14px; classDef aging fill:#f9d4f9, font-weight:bold, font-size:14px; A[Tomasz M. Rutkowski] --> B[Spring School: BCI topics,
lectures, demos, hackathon. 1] A --> C[G.tec: EEG hardware, software
for medicine, research. 2] C --> D[Quality EEG needs proper
setup, maintenance, placement, verification. 3] C --> E[EEG evolved: large analog
to compact digital, wireless. 4] C --> F[Dry electrodes: good EEG,
no gel, P300 BCI. 5] C --> G[Demos: passive, active wet,
dry electrode comparison. 6] C --> H[G.tec software: g.Recorder, g.HIsys,
g.BSanalyze, real-time, LSL integration. 7] A --> I[BCIs combined: FNIRS, eye tracking,
VR, AI like ChatGPT, DALL-E. 8] I --> J[Screen dress: visualize EEG
concentration on wearable display. 9] I --> K[Unicorn Hybrid Black: 8-channel
wireless consumer EEG, under 1000. 10] K --> L[Unicorn Hybrid Black: dry electrodes,
2-second setup. 11] I --> M[BCI applications: spatial auditory, tactile,
visual for avatar, robot control. 12] A --> N[Reactive BCIs: decode EEG intentions
real-time for interaction. 13] A --> O[Passive BCIs: monitor mental states. 13] A --> P[Multimodal BCIs: EEG + FNIRS,
eye tracking, VR, improve performance. 14] A --> Q[Dementia: growing elderly population, need
objective cognitive monitoring. 15] Q --> R[P300 oddball: cognitive differences
in healthy vs MCI. 16] Q --> S[Emotion recognition: features discriminate
healthy aging vs MCI. 17] Q --> T[EEG graphs: lower network
complexity in MCI subjects. 18] T --> U[EEG network topology: 90%
MCI vs healthy accuracy. 19] Q --> V[FNIRS: different oxygenation in
healthy vs MCI during tasks. 20] Q --> W[Olfactory bulb: 40-100Hz EEG,
measurable above nose. 21] W --> X[Olfactory oddball: different EEG/EBG
in MCI vs healthy. 22] X --> Y[EEG+EBG improved single-trial
olfactory target/non-target classification. 23] W --> Z[FNIRS: different prefrontal responses
to odors in MCI vs healthy. 24] W --> AA[Pupil dynamics: more complex
networks in healthy vs MCI. 25] Q --> AB[Summary: multimodal BCIs promising
as early dementia biomarkers. 26] AB --> AC[ML enables complex BCI
for cognitive assessment in aging. 27] Q --> AD[Collaborative research invited: develop,
validate BCI biomarkers, larger cohorts. 28] Q --> AE[PhD positions: Nicolaus Copernicus University,
Poland, BCI biomarker research. 29] Q --> AF[No current research: BCI biomarkers
for migraine, aphasia: related work pursued. 30] class A,B school; class C,D,E,F,G,H,J,K,L gtec; class I,M,N,O,P bci; class Q,R,S,T,U,V,W,X,Y,Z,AA,AB,AC,AD,AE,AF aging;

Resume:

1.-Spring School covers topics related to brain-computer interfaces (BCIs), with lectures, demos, and a hackathon.

2.-G.tec develops EEG hardware and software for medical and research applications. Their history and products were reviewed.

3.-High-quality EEG recordings require proper setup, equipment maintenance, electrode placement, and real-time signal verification.

4.-EEG devices evolved from large analog systems to compact digital and wireless amplifiers.

5.-Dry electrodes can provide good EEG quality without gel for some BCI applications like P300.

6.-Live demos compared EEG quality and setup time for passive, active wet, and dry electrodes.

7.-G.tec's software like g.Recorder, g.HIsys, g.BSanalyze enable real-time processing and integration with other programs via LSL.

8.-BCIs were combined with FNIRS, eye tracking, virtual reality, and AI like ChatGPT and DALL-E.

9.-G.tec developed the screen dress that visualizes EEG concentration levels on a wearable display embedded in a dress.

10.-The Unicorn Hybrid Black is a 8-channel wireless consumer EEG headset priced under 1000 euros for home use.

11.-The Unicorn Hybrid Black with dry electrodes enables extremely fast 2-second setup times.

12.-BCI applications included using spatial auditory, tactile, and visual paradigms for virtual avatar and robot control.

13.-Reactive BCIs decode user intentions from EEG in real-time for interaction, while passive BCIs monitor mental states.

14.-Multimodal BCIs combine EEG with FNIRS, eye tracking, virtual reality, etc. to improve performance.

15.-Dementia affects a growing elderly population. Lifestyle interventions may help but need objective cognitive monitoring.

16.-Classical P300 oddball paradigms show cognitive differences between healthy and mild cognitive impairment (MCI) groups.

17.-Learning and memory tasks like emotion recognition provide features to discriminate between healthy aging and MCI.

18.-Network neuroscience represents EEG as graphs. MCI subjects have lower network complexity than healthy elderly.

19.-Topological data analysis of EEG networks enables MCI vs healthy classification around 90% accuracy.

20.-FNIRS brain oxygenation patterns differ between healthy and MCI groups during learning/memory tasks.

21.-The olfactory bulb generates high frequency 40-100Hz EEG activity measurable with electrodes above the nose.

22.-Olfactory oddball paradigms presenting target and non-target odors elicit different EEG/EBG responses in MCI vs healthy.

23.-Combining EEG and olfactory EEG (EBG) improved single-trial classification of olfactory targets/non-targets.

24.-FNIRS shows different hemodynamic responses over prefrontal cortex to olfactory stimuli in MCI vs healthy aging.

25.-Pupil size dynamics modeled as networks were more complex in healthy vs MCI during olfactory tasks.

26.-In summary, reactive and passive multimodal BCIs (EEG/FNIRS/EBG/eye tracking) show promise as early dementia biomarkers.

27.-ML makes it feasible to utilize complex BCI paradigms in real-time for cognitive assessment in aging populations.

28.-Collaborative research is invited to develop and validate BCI biomarkers on larger elderly cohorts.

29.-PhD positions are available at Nicolaus Copernicus University in Poland for BCI biomarker research.

30.-No current research on using these BCI biomarkers for migraine or aphasia, but related work is being pursued.

Knowledge Vault built byDavid Vivancos 2024