Cogitat won an award for technology that could make it possible to control machines using the brain – and enable more realistic virtual reality.
The Imperial startup won the top prize after outperforming its competitors in an international competition to find the best machine learning technologies to identify brain states and behavioral intentions using brain waves.
Medical applications of brain-computer interfaces include restoring movement after an accident. A new emerging application increases the realism of virtual reality.
The BEETL AI Challenge for EEG Transfer Learning competition, sponsored by Facebook Reality Labs, was held as part of the prestigious NeurIPS 2021 machine learning conference. It aimed to identify teams and global companies that have the best technology in place. ‘machine learning to accurately identify brain and mental states from electroencephalogram (EEG) data.
Technologies effective in this space could be used to improve medical diagnostics or to power brain-computer interfaces, which allow users to control machines without physical controllers. Ultimately, they could be a key technology behind the “metaverse,” a hypothetical future version of the Internet that consists of 3D virtual reality environments.
Exploitation of EEG data
EEG devices have long been used in medical and research settings to non-invasively measure electrical activity in the brain. However, it is very difficult to accurately identify brain or mental states using EEG data, because the data is very noisy and the relationship between electrical activity and brain or mental state varies both between individuals and within individuals at any time.
Advances in machine learning, especially deep learning, are starting to identify complex relationships that consistently develop between individuals between EEG measurements and brain or mental states such as motor intentions. . Software based on these advances could be used for various applications, including driving brain-computer interfaces.
Cogitat demonstrated a clear advantage over its competitors in competitive tests, which involved interpretation of EEG data on sleep and motor intentions. In a field of more than 110 academic and industrial teams, Cogitat obtained 1st place in the motor image decoding task with a significant margin over the competition – a difference of 5% compared to second place and 16% by compared to third place – and 1st place overall in the competition.
The team attributes their success to their focus on machine learning and software, which they focus on while remaining hardware independent.
It is very difficult to accurately identify brain or mental states using EEG data, but machine learning is starting to make it possible.
Dr Dimitrios Adamos, Co-Founder and CTO and Imperial Researcher of Cogitat said: “Our team combines advanced expertise in the field of machine learning and neuroinformatics. Our technology focuses on decoding motor imagery, robust to anatomical and neurological differences between users and different EEG recording setups. These sources of variability undermine the reliability of EEG decoding and have hampered the widespread use of EEG outside of laboratories for years ”.
Stefanos Zafeiriou, Co-Founder and Chief Scientist of Cogitat, and Professor of Machine Learning and Computer Vision in the Department of Computer Science at Imperial Oil, said: “We use very specialized deep learning technology, specific to this application and suitable for the interpretation of EEG signals. Our success reflects the world-class machine learning expertise of my lab at Imperial Oil and the best neuroinformaticians we have in the business. “
Dr Rebeca Santamaria-Fernandez, Director of Industrial Partnerships and Commercialization (Engineering) at Imperial, said: “I wish Cogitat big congratulations on this victory, which demonstrates the great potential of their technology to power next-generation brain-computer interfaces. The team’s academic and business expertise, and track record in successfully commercializing technology and data developed at Imperial, places them in an excellent position to launch impactful technology that seamlessly merges the mind. and the machine.
The future of virtual reality
Brain-computer interfaces have a variety of potential applications, including medical applications such as restoring movement and control in those who have had accidents, and assisting in stroke rehabilitation by translating motor signals into the brain. ‘a patient under artificial muscle stimulation. A new emerging application increases the realism of virtual reality by allowing users to navigate virtual environments by thinking rather than using a physical controller.
We could have VR games for rehabilitation, or games that could improve reaction times and athletic performance. Allan Ponniah CEO of Cogitat
Allan Ponniah, CEO of Cogitat, said the company is interested in a variety of use cases, especially within VR: “We’ve been hearing about VR for decades, but the computing power for create these environments and the level of investment, suggests that the time has finally come. We are also interested in entertainment and medicine. There is also a space where the two come together. We could have VR games for rehabilitation, or performance enhancing games that could improve reaction times and athletic performance. For me, this interaction is the most interesting thing.
The company currently continues to research applications of the technology in the real world and seeks partnerships with other technology companies in the virtual reality and games industries to take full advantage of the opportunities created by the new era of portable devices.