Wearable Deep Spinal Interfacing (WDSI)

require augmenting the capacity of humans to control and coordinate actions beyond the bodily limitations of having e.g. only a pair of hands. Augmentation beyond our present human limitations would have on the one hand, a revolutionary impact in our understanding of the working principles and limitations of the central nervous system, and, on the other hand, a breakthrough impact in a variety of technological scenarios. Applications where human augmentation would be game-changing include robotic surgery where surgeons may perform more complex procedures by using extra pair of hands, the ability to work in extreme environments by robotic teleoperation (e.g., in deep space, underwater or in dangerous plants), or next generation manufacturing where workers may use multiple limbs such as the confined spaces of aircraft assemble, and so on. The question whether or not human augmentation is at all possible, however, has no positive answer yet. The challenge is that existing approaches to interface human and machine, require us to repurpose the body's signals, e.g., signals that control the motion of our hand can be used to control a prosthetic hand. This is important for amputee patients that want to restore hand function but means that able-bodied people can either choose to control a robotic hand or their own hand but use both independently at the same time. Achieving the latter would be true human augmentation. Enabling human augmentation to operate technology while at the same time keeping our natural ability to act, talk and walk naturally intact would allow us to expand the capabilities and competitiveness of our species to novel technology domains enabled by artificial intelligence.
Human augmentation has currently not been realised with any existing technologies, partly because of the limitations of direct linking with the human brain. In this ambitious and high-risk project, we propose that some brain signals not directly related to a limb's movement "bleed-through" all the way to its muscles electrical activity. Recent work has shown that muscle electrical signals, which can be easily measured with wristwatch-like sensors, may indeed contain hidden brain information that could be read out to operate technology, while the user engages the muscles for normal physical activities.
Here, we propose a unique approach to realise human augmentation by combining wrist-watch-like sensors and artificial intelligence. We aim to show that we can enable a human user to control their arm movements, while at the same time using the signals from the arm muscles to operate an independent technological device. This would constitute a breakthrough achievement in human interfacing and a decisive demonstration of true human motor augmentation. The resulting augmentation technology will substantially differ from all current mainstream approaches in that it will not require surgery or alterations of the human body, as it will be applicable in the form of wearable sensors. As such, our work will have an impact not only for movement impaired individuals, such as in the control of prosthetic limbs, but also for everyday consumers, as a new way to interfacing with computers and other technology. To realise this goal we bring together two international leading teams of AI and medical engineers and propose a work plan of 18 months to build, a practical demonstration that our technology vision works.