Knowledge Vault 3/6 - GTEC BCI & Neurotechnology Spring School 2024 - Day 1
g.tec Suite to control Virtual Reality and Augmented Reality
Bernard Wong, g.tec neurotechnology Hong Kong (HK)
<Resume Image >

Concept Graph & Resume using Claude 3 Opus | Chat GPT4 | Llama 3:

graph LR classDef gtech fill:#f9d4d4, font-weight:bold, font-size:14px; classDef unity fill:#d4f9d4, font-weight:bold, font-size:14px; classDef bci fill:#d4d4f9, font-weight:bold, font-size:14px; classDef games fill:#f9f9d4, font-weight:bold, font-size:14px; classDef future fill:#f9d4f9, font-weight:bold, font-size:14px; A[Bernard Wong] --> B[G-Tech: VR/AR games with BCI. 1] B --> C[G-Tech BCI products: Unity Speller,
Simulink model, SDK. 2] C --> D[Unity Speller: P300 protocol,
controls game via UDP. 3] D --> E[P300: oddball paradigm, target vs
non-target stimuli. 4] D --> F[Unity Speller: selectable character grid,
flashing rows/columns. 5] D --> G[Counting flashes trains classifier,
detects P300 response. 6] D --> H[Spelled characters control separate
Unity game via UDP. 7] H --> I[Demo: Unity Speller removes
obstacles in 2D game. 8] C --> J[Simulink model: BCI stimuli,
signal processing, 3D game. 9] J --> K[Example: focus on shape colors,
guess target color. 10] J --> L[Simulink handles data acquisition,
filtering, classification, game control. 11] C --> M[Unity package SDK: integrates BCI
flexibly into apps. 12] M --> N[Unity SDK supports ERP,
cVEP BCI protocols. 13] M --> O[SSVEP support planned for
continuous BCI control. 14] M --> P[Unity SDK enables VR games
with Oculus Quest 2. 15] M --> Q[ADHD game in development,
focuses on puzzle pieces. 16] Q --> R[Focusing reveals image,
may improve concentration. 17] B --> S[BCI as 'superpower', not replacing
standard game controls. 18] B --> T[Considerations: shorter training,
fewer separate apps. 19] B --> U[Bright lighting crucial for
visual BCI stimuli effectiveness. 20] B --> V[Stable frame rate critical for
precise stimuli timing, classification. 21] B --> W[Evolution: better BCI integration
into game environments. 22] B --> X[Bluetooth: Unicorn headset directly
connects to Unity. 23] X --> Y[Unicorn Unity integration: Oculus/Meta
Quest 2/3, Pico 4 support. 24] B --> Z[BCI introduces 3-5s delay,
suits occasional interactions. 25] B --> AA[Longer classifier training may
reduce false positives. 26] B --> AB[Unicorn Unity asset: easy
drag-and-drop game integration. 27] B --> AC[Linux, SteamDeck support for
Unicorn Unity SDK explored. 28] B --> AD[Unicorn simulator enables development
without physical headset. 29] B --> AE[G-Tech's tools: more user-friendly,
integrated for easier adoption. 30] class A,B,W,X,Y,Z,AA,AB,AC,AD,AE gtech; class C,D,F,G,H,M,N,O,P,Q unity; class E,I,J,K,L,S,T,U,V bci; class R games;

Resume:

1.-Bernard, a software developer at G-Tech, presented on developing VR/AR games using brain-computer interface (BCI) technology and Unity.

2.-G-Tech offers different products for developing VR/AR games with BCI: a Unity Speller app, Simulink model, and Unity package.

3.-The Unity Speller app allows selecting commands using the P300 protocol to control a standalone gaming interface via UDP.

4.-The P300 protocol uses an oddball paradigm with infrequent target stimuli among frequent non-target stimuli to elicit a brain response.

5.-The Unity Speller interface has a selectable grid of characters. Users focus on a target character while rows/columns flash.

6.-Counting target flashes trains a classifier to detect the P300 brain response, enabling the user to spell characters.

7.-The spelled characters/commands can control aspects of a separate Unity game application connected via UDP.

8.-In one demo, the Unity Speller selected colors to remove obstacles in a Mario-style 2D game running separately.

9.-The Simulink model integrates the BCI stimuli and signal processing with the actual 3D game environment in one app.

10.-An example Simulink game has the user focus on colors of 3D shapes to guess a target shape's hidden color.

11.-The Simulink model handles data acquisition, filtering, and classification of the BCI data to control the game.

12.-A Unity package SDK allows integrating BCI stimuli, signal processing, and game logic flexibly into one cohesive app.

13.-The Unity package currently supports event-related potential (ERP) and code-based visual evoked potential (cVEP) BCI protocols.

14.-SSVEP support is planned for the Unity SDK in the future. It allows continuous BCI control.

15.-The Unity SDK enables creating VR games with integrated BCI control using the Oculus Quest 2 headset.

16.-An ADHD game is in development using the Unity BCI SDK to have users focus on puzzle pieces.

17.-Focusing on the puzzles to reveal an underlying image may help improve concentration skills in children with ADHD.

18.-Unity BCI games aim to make the BCI control an added "superpower" rather than replacing standard controls completely.

19.-Important considerations for Unity BCI games include using shorter training times and minimizing the number of separate applications needed.

20.-Environmental lighting should be bright enough for the visual BCI stimuli to be effective in evoking the brain responses.

21.-A stable frame rate is critical in Unity BCI games to enable precisely timed stimuli and accurate brain response classification.

22.-G-Tech's BCI game development has evolved to better integrate stimuli into the actual game environment within one application.

23.-Bluetooth allows directly connecting G-Tech's Unicorn BCI headset to Unity without streaming data through a separate interface.

24.-The Unicorn Unity integration will soon support the Oculus/Meta Quest 2/3 and Pico 4 virtual reality headsets.

25.-The BCI signal processing introduces a roughly 3-5 second delay for commands, suitable for occasional game interactions but not continuous control.

26.-Lengthening classifier training from 30 seconds to 1 minute may reduce false positives when shifting focus between selectable objects.

27.-The Unicorn Unity asset will be available in the Unity asset store to enable easy drag-and-drop integration into games.

28.-LINUX and SteamDeck support for the Unicorn Unity SDK are being explored but are not yet fully implemented.

29.-A Unicorn simulator allows developing BCI games without a physical headset to facilitate collaboration between remote developers.

30.-G-Tech's VR/AR BCI game development tools have evolved to be more user-friendly and cohesively integrated to facilitate easier adoption.

Knowledge Vault built byDavid Vivancos 2024