Whole brain emulation (WBE) aims at replicating the functions of a whole brain in some other operating substrate. The engineering practice of System Identification (SID) can be applied in a way that makes this big problem a feasible collection of connected smaller system identification problems to solve. The most efficient technology currently used for achieving WBE via SID involves closed-loop neuroprosthetics. While this seems to be a feasible approach to look at input-output computations in rodents and monkeys so far, there is still plenty of room for improvement in order to make this technology available for human trials.
A number of clinical trials and commercially available implants for humans have become available. Most of them use deep brain stimulation (DBS), slowly moving towards closed-loop stimulation, in that they are able to record and stimulate at the same time, in order to prevent epileptic seizures. There are also a number of clinical trials with brain-machine interfaces (BMI) either recording from pre-motors areas and having an output in sensory areas, allowing the patients to feel objects with their prosthetic hands and enable better control of artificial limbs.
There is a conceptual and technological gap between clinical research using DBS electrodes and basic science using single units in rodents and monkeys. The conceptual gap is given by the fact that no causal explanation of how DBS stimulation can prevent tremor or depression down to neuronal circuits, while basic research with single units is evolving in small incremental steps, bound by the rules of academic research. The technological gap comes from the fact that DBS involves the use of large electrodes and fixed stimulation patterns, while closed-loop neuroprosthetics currently tested in rodents and primates uses single unit recordings, in order to achieve optimal spatial resolution, while generating a model of recorded activity in order to derive unique stimulation patterns. Moreover, further testing of related emerging technologies in rodents, monkeys and clinical trials is required in order to make this technology feasible.
In order to bridge this gap, we welcome original contributions as well as reviews which pave the path towards connecting basic research that performs system identification for closed-loop neuroprosthetics and clinical applications. This is not only the next logical step in clinical research in neurotechnology, but would also provide invaluable neuroscientific data on input-output computations in the human brain that would build up on the project of WBE. There are a number of technological challenges that need to be addressed in the process, including reduced invasiveness, more efficient surgery and stereotaxic procedures, improvement of the model used for the recorded data, smaller, more stable and biocompatible electrodes for the implant, more coverage of neurons, better data acquisition and storage for a common database, and improved behavioral testing and monitoring.
We therefore welcome contributions showing improvements of any of these aspects with the final aim of performing WBE-aimed system identification in humans. This also includes early-stage research on novel approaches, such as nanotechnology, laser surgery, and advanced modeling systems.
Whole brain emulation (WBE) aims at replicating the functions of a whole brain in some other operating substrate. The engineering practice of System Identification (SID) can be applied in a way that makes this big problem a feasible collection of connected smaller system identification problems to solve. The most efficient technology currently used for achieving WBE via SID involves closed-loop neuroprosthetics. While this seems to be a feasible approach to look at input-output computations in rodents and monkeys so far, there is still plenty of room for improvement in order to make this technology available for human trials.
A number of clinical trials and commercially available implants for humans have become available. Most of them use deep brain stimulation (DBS), slowly moving towards closed-loop stimulation, in that they are able to record and stimulate at the same time, in order to prevent epileptic seizures. There are also a number of clinical trials with brain-machine interfaces (BMI) either recording from pre-motors areas and having an output in sensory areas, allowing the patients to feel objects with their prosthetic hands and enable better control of artificial limbs.
There is a conceptual and technological gap between clinical research using DBS electrodes and basic science using single units in rodents and monkeys. The conceptual gap is given by the fact that no causal explanation of how DBS stimulation can prevent tremor or depression down to neuronal circuits, while basic research with single units is evolving in small incremental steps, bound by the rules of academic research. The technological gap comes from the fact that DBS involves the use of large electrodes and fixed stimulation patterns, while closed-loop neuroprosthetics currently tested in rodents and primates uses single unit recordings, in order to achieve optimal spatial resolution, while generating a model of recorded activity in order to derive unique stimulation patterns. Moreover, further testing of related emerging technologies in rodents, monkeys and clinical trials is required in order to make this technology feasible.
In order to bridge this gap, we welcome original contributions as well as reviews which pave the path towards connecting basic research that performs system identification for closed-loop neuroprosthetics and clinical applications. This is not only the next logical step in clinical research in neurotechnology, but would also provide invaluable neuroscientific data on input-output computations in the human brain that would build up on the project of WBE. There are a number of technological challenges that need to be addressed in the process, including reduced invasiveness, more efficient surgery and stereotaxic procedures, improvement of the model used for the recorded data, smaller, more stable and biocompatible electrodes for the implant, more coverage of neurons, better data acquisition and storage for a common database, and improved behavioral testing and monitoring.
We therefore welcome contributions showing improvements of any of these aspects with the final aim of performing WBE-aimed system identification in humans. This also includes early-stage research on novel approaches, such as nanotechnology, laser surgery, and advanced modeling systems.
