Our Mission
- Brain is linked to unresolved mysteries. Neural signals are not reproduced across individuals. AI and machine learning (ML) are still superficially or not employed at all, without any large scale proven generalization across subjects. This generalization is kind of a "holy grail", to support the democratization and applicability of such systems at scale.
- At the same time affective, mental, cognitive states are challenging to monitor, classify or detect, whereas Human Computer Interaction is still in the "stone age" since interfaces are far away from being natural and aware of the mental/affective states.
- Our mission is to tackle within this context, the generalization across individuals blending the areas of brain neural signals, neuroinformatics, BCIs, eye-tracking, virtual reality while leveraging machine/deep learning superpowers.
Use Cases / Solutions
We introduce novel ML/DL-powered interfaces that employ non-invasive brain computer sensors to function across users requiring minimal calibration data.
Our aim is to interface as seamlessly and efficiently with systems/devices via AI solution that generalize. This is downstream linked to novel multimodal affective aware applications within an VR/AR setting in well-being related use cases. Up to now developed cases include the following:
- Motor imagery for brain-controlled applications; Assistive Technology
- Accessibility, actual or virtual mobility,
- Cognitive rehabilitation
- Monitor cognition, predict training outcomes, BCI-triggered interventions for cognitive functions e.g., Attention, Memory, Visuospatial abilities, Enhanced BCI-based Neurofeedback, e.g. Attention deficits, Learning, Stress
- Mental state monitoring*
- Detect mental/physical workload, prevent overload.
Such pipelines can be employed for Motor/Cognitive Rehabilitation, Relaxation Therapies, Pain Reduction.
Some of these approaches can be also employed in “milder” well-able cases out of the rehabilitation context such as personal mental state, emotion recognition, work-cycle fatigue and so on and so forth.
The above use cases are co-developed with Panagiotis Kourtesis, Serafeim Perdikis; *+Ioanna Zioga.
More use cases available upon request.
Stay tuned or talk to us for further custom development; upcoming focused use-cases are on the way.
Technology
> State- of-the- art ML/DL
>Transfer Learning, Self-supervised learning, ++
>Explainable AI
> Leverage massive datasets
Applications
> Brain-controlled devices / BCI for Assistive Technology
> Neurorehabilitation - Motor Rehabilitation
>Cognitive Rehabilitation/Training
> Mental State Monitoring/ Relaxation/ Pain reduction
EEG integrated with other modalities such as eye tracking all together embedded into AR/VR headsets empowered by state-of the art deep learning and transfer learning approaches can open new directions towards novel affect-aware human computer interfaces, and assistive techologies.
The Team
Internal Team
Nikos Antonopoulos
ML Research Engineer
Ioannis Gkinis
ML Engineer/SW developer
Christina Sartzetaki
ML Engineer
Panos Antoniadis
ML Engineer
External collaborators/advisors
Panagiotis Kourtesis
Experienced VR & HCI, Researcher, and Cognitive Neuroscientist, at Inria
Ioanna Zioga
Postdoctoral Researcher at the Donders Institute for Brain, Cognition and Behaviour, the Netherlands
Manousos Klados
Experienced BCI, Aff. Comp. Research Scientist, Prof. at CITY College/York Uni.
Stavros Dimitriadis
Publication
C. Sartzetaki, P. Antoniadis, N. Antonopoulos, A. Krasoulis, S. Perdikis, V. Pitsikalis, “Beyond Within-Subject Performance: A Multi-Dataset Study of Fine-Tuning in the EEG Domain”, IEEE SMC 2023