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

Resume:

1.- The BrainIO hackathon was held as part of the GTEC Spring School 2024. A team presented their prototype for controlling character movement in Unity using brain signals.

2.- Another team from Mexico called "Fitstuff set of mind" created a car racing game controlled by EEG and unicorn speller to help with attention issues.

3.- A team called "Mind Raisers" created a single racer car simulator game in Unity controlled by a customized unicorn speller board to accelerate and change direction.

4.- A team from Mexico built a game to train sustained visual focus and concentration while ignoring distractions, using theta/beta ratios from unicorn EEG.

5.- A Mexican team made an interactive geography puzzle game controlled by a P300 speller to match country borders and engage elementary students.

6.- A team created Project Alex, an education puzzle game to explore world geography by matching country silhouettes, aimed at elementary school children.

7.- A Mexico team made a video game character that travels to the dystopian future but is given power to change the past by a Hispanic god.

8.- A team called Time Voyagers made an educational game with a character who travels from a dystopian future to the present to collect plants.

9.- A team used ECoG data to decode attempted finger movements and control a 3D printed robotic hand with 3 degrees of freedom.

10.- A team analyzed ECoG data from patients with unresponsive wakefulness syndrome, finding differences in P300 responses to vibrotactile stimuli between patients.

11.- A team called Eco Girls analyzed an ECoG dataset to classify 3 hand poses - fist, peace sign, open hand, achieving 98.6% accuracy.

12.- An Italian team used ECoG video data to classify when objects appeared on screen vs just background, using machine learning on frequency bands.

13.- A team made an educational game called Tekawani controlled by EEG about a character who travels from a dystopian future to change the past.

14.- A Colombian team used EEG motor imagery data from stroke patients to compare classification methods and found decreased performance in one subject post-therapy.

15.- An Australian team replicated LDA classification of EEG motor imagery in stroke and found similar decreased post-therapy performance in one subject as the Colombians.

16.- A French team tried using EEG dream data to extract emotional states and generate corresponding images using machine learning but ran out of time.

17.- A Mexican team made a relaxation neurofeedback game measuring alpha and theta activity, where users shoot growing flames by concentrating to achieve a flow state.

18.- A Taiwan team made an EEG controlled orthosis for stroke rehab allowing individual finger control via motor imagery, to enable Rock Paper Scissors play.

19.- An Egyptian team built an EEG controlled humanoid robot and VR orthosis for stroke rehabilitation, using hybrid SSVEP and motor imagery BCI.

20.- A Canadian team used EEG, physiological metrics and VR to train surgeons in stress inoculation and optimal performance mindset pre & intra-operatively.

21.- A Polish team made an EMDR & VR therapy for PTSD, integrating alpha/delta neurofeedback, procedurally generated visuals, and ChatGPT to customize treatment.

22.- A Canadian team made a meditation aid integrating EEG concentration monitoring, auditory feedback and ChatGPT generated guidance to help refocus wandering minds.

23.- A Polish team made a biometric identity authentication system called Proof of Mind based on unique memory EEG signatures as an alternative to iris scanning.

24.- A Colombian team made a VR arcade game controlled by EEG measuring attention to train focus and treat ADHD by problem solving without distractions.

25.- A Romanian team made a 3D scooter game in LabView controlled by a P300 speller moving along paths of increasing difficulty to train neurorehabilitation.

26.- An Italian team made a competitive 2-player EEG controlled dancing robot game using motor imagery to trigger dance moves, developed in ROS.

27.- A Dutch team analyzed ECoG data to determine optimal music for focus in an app, by measuring responses during reading comprehension tests.

28.- A Mexican team made a P300 mind-controlled wheelchair with intelligent navigation for accessibility using a Geodesic Sensor Net and virtual environment.

29.- A Korean team used ECoG to decode attempted hand poses, train an LSTM to generate trajectories, and control a soft robotic glove for rehab.

30.- Three Australian teams replicated a study using EEG motor imagery in stroke rehab with similar results - CNN outperformed traditional methods.

31.- A Serbian team tried using auto-encoders to reconstruct full EEG from partial channels to reduce hardware costs while preserving accuracy, with limited success.

32.- A Romanian team made a BCI VR app for disabled users to virtually travel the world in Google Street View using a P300 speller.

33.- An Italian team made "BrainWave Arcade", a mind-controlled Pac-Man game using the unicorn speller in Unity to enable gaming for motor impaired users.

34.- A Canadian team used EEG to detect optimal music for focus based on responses during reading tests, envisioning medical & educational uses.

35.- A Taiwanese team created an EEG drone swarm light show inspired by a brain-to-brain interface study, to celebrate human connection and collaboration.

36.- An international team analyzed P300 data to optimize averaging trials and channels to improve classification accuracy of target vs non-target stimuli.

37.- An international team created a mental well-being app monitoring EEG to give users insight into their emotional states and guide meditation.

38.- An Italian team made a BCI musical jam system letting a performer control guitar effects and audience visuals with EEG for enhanced connection.

39.- A Mexican team made a P300 mind-controlled smartphone app to open accessibility for the motor impaired, achieving high accuracy and usability.

40.- An Italian team worked on improving P300 speller performance using power spectral density features and random forest classification, achieving 98% accuracy.

41.- A Romanian team made a BCI VR Google Street View world exploration app for the disabled using a P300 speller and speech recognition.

42.- A Dutch team made MindMelody, an app that finds optimal focus music for each user based on EEG responses during tasks.

43.- An international team analyzed P300 data, finding that channel reduction and resampling non-target trials 5:1 improved random forest classification to 98% accuracy.

44.- An international team made an EEG music recommendation app and generated a personalized visual artwork of the user's emotional experience of the song.

45.- A Canadian team created an EEG-controlled VR Google Street View exploration app for the disabled, integrating voice control and ChatGPT place descriptions.

46.- A Canadian team used EEG to determine optimal background music for focus during reading, envisioning applications for ADHD, fatigue and Alzheimer's monitoring.

47.- A Romanian team made an EEG-controlled VR maze game to entertain the disabled, with coins to collect and mazes regenerating to sustain flow states.

48.- An Italian team made an EEG-controlled Pac-Man VR game for cognitive training and motor impairment accessibility, using the unicorn speller in Unity.

49.- An Italian team made a BCI musical jam system, letting a guitarist control effects and visuals with EEG, and the audience's EEG influence the sound.

50.- A Mexican team made an EEG smartphone controller app for accessibility, using the unicorn P300 speller, achieving high accuracy and adding many functions.

51.- An international team used machine learning to reconstruct missing EEG channels from partial data to reduce hardware costs while preserving accuracy.

52.- A Romanian team made a VR mind-controlled app for the disabled to virtually tour the world in Google Street View, with ChatGPT place descriptions.

53.- A Taiwanese team decoded ECoG data to control a personalized AI coach in VR for engaging post-stroke hand rehabilitation through games and encouragement.

54.- A Canadian team analyzed P300 data, optimized trial averaging and channel selection, and achieved 98% target classification accuracy with random forests.

55.- A Romanian team made an EEG-controlled VR travel app for the disabled to explore Google Street View with voice commands and AI guidance.

56.- An Indian team optimized P300 speller trial averaging and features to achieve 98% classification accuracy of target stimuli using random forests.

57.- A Canadian team used EEG to detect P300 responses to visual stimuli of letters and numbers, achieving 91.5% classification accuracy with CNN.

58.- An Italian team made an EEG-controlled MATLAB & Unity VR soccer goalie game for mental stimulation, with randomized saves for added challenge.

59.- An international team analyzed ECoG responses to different visual categories, finding unique activation patterns and decoding objects vs faces with reasonable accuracy.

60.- An international team explored using EEG auto-encoders to compress data from high density to low density electrode montages while preserving signal.

61.- A Romanian team built a VR street view exploration app controlled by a P300 speller for accessibility, with AI guides and Wikipedia info.

62.- A Canadian team used EEG to detect which music keeps a user most focused during reading, for personalized playlists in education and medicine.

63.- An Italian team made a multiplayer EEG-controlled robot dancing game using motor imagery, inverse kinematics, and a scoring system to compete.

64.- A Mexican team created a P300 mind-controlled mobile app to open accessibility of smart devices, adding many control functions with high accuracy.

65.- A British team made an EEG-based personalized visual artwork generated from a user's brain activity while listening to music, as an artistic experience.

66.- A Romanian team made a VR app for virtually touring landmarks using Google Street View, EEG and speech control, to improve disabled accessibility.

67.- A Canadian team decoded EEG to an AI recommender system for optimal focus music during work, validated with improved reading comprehension.

68.- An Italian team made a 2-player EEG robot dance battle game using motor imagery, musically-synched animations and a scoring system.

69.- The hackathon concluded with an awards ceremony recognizing the top three projects in each category - gaming, artistic/programming, and data analysis.

70.- Here are the winners in Gaming Projects:

71.- 1st Place: Mind-Controlled Pac-Man by The Shout of the Carpathians (Romania)

72.- 2nd Place: Stroke Rehabilitation Game by NYCU_AS Rehab Group (Taiwan)

73.- 3rd Place: Not announced

74.- Here are the winners in Programming & Artistic Projects:

75.- 1st Place: EEG-Controlled Musical Jam System by Chimera Underground (Italy)

76.- 2nd Place: EMDR VR Therapy for PTSD by Reminder (Poland)

77.- 3rd Place: EEG-Controlled Orthosis for Stroke Rehab by MindMap (Romania)

78.- Here are the winners in Data Analysis Projects:

79.- 1st Place: ECoG Hand Pose Decoding by ECoggers (International)

80.- 2nd Place: EEG Stroke Rehab Motor Imagery Analysis by BOLIMAC (Serbia)

81.- 3rd Place: Variational Mode Decomposition for SSVEP Detection by BCI Arians (International)

82.- The winning teams represented a diverse set of countries and approaches, highlighting the creativity and innovation in applying BCI technology to real-world challenges in healthcare, accessibility, gaming, arts and more. The hackathon provided a platform for students and researchers to collaborate across disciplines and time zones in advancing the field of neurotechnology.

Knowledge Vault built byDavid Vivancos 2024