Commentary From Novel Technology to Novel Applications: Comment on "An Integrated Brain-Machine Interface Platform With Thousands of Channels" by Elon Musk and Neuralink Pisarchik et al JOURNAL OF MEDICAL INTERNET RESEARCH XSL • FO RenderX (original) (raw)

The first attempts to translate neuronal activity into commands to control external devices were made in monkeys yet in 1960s. After that, during 1960-1970, the biological feedback was realized in monkeys, to provide voluntary control of the firing rate of cortical neurons. The term "brain-computer interface" appeared only in earlier 1970s. The brain-computer interface is usually referred to as a "brain-machine interface" in invasive studies. Nowadays, the brain-computer interface and brain-machine interface research and applications are considered one of the most exciting interdisciplinary areas of science and technology. In particular, brain-computer interfaces are very promising for neurorehabilitation of sensory and motor disabilities, neurocommunication, exoskeletons, cognitive state evaluation, etc. Advanced mathematical methods for extraction and classification of neuronal activity features hold out hope for the future use of brain-computer interfaces in everyday life. At the same time, the lack of effective invasive neuroimaging techniques providing a high-resolution neural activity recording for medical purposes limits the brain-machine interface implementation in clinics. In their paper, Elon Musk and Neuralink have successfully addressed the major issues hampering the next generation of invasive brain-computer interface (or brain-machine interface) development by introducing a novel integrated platform enabling a high-quality registration of thousands of channels. Their device contains arrays of flexible electrode threads with up to 3072 electrodes per array, distributed across 96 threads. To overcome a surgical limitation, the authors have built a neurosurgical robot that inserts 6 threads per minute with a micrometer spatial precision. To increase the biocompatibility, they created a neurosurgical robot, which implants polymer probes much faster and more safely than existing surgical approaches. Using this platform in freely moving rats, the authors report a spiking yield of up to 85.5%. Although the developed system is considered an effective platform for research in rodents, it can serve as an invasive neurointerface prototype for clinical applications. Specifically, multielectrode neurointerfaces may become the basis for new communication systems and advanced assistive technologies for paralyzed people as well as control external devices and interact with the entire environment, eg, by integrating into new fast developed technologies, such as Smart Home and Internet of Things. Moreover, the brain-computer interface applications are very promising for detecting hidden information in the user's brain, which cannot be revealed by conventional communication channels.