The object of this research is to develop a new platform concept, BioBolt, as a distributed wireless neural interface for minimally-invasive, fully-implantable epidural recording. The approach is to implement a stand-alone wireless link for a number of microelectrodes, which can be easily placed on the dura mater through a small hole in the skull by simple operation procedure.
The technical challenge is to implement an extremely low-power wireless link in a small form factor. The proposed system consists of multiple BioBolts and MasterBolt. Spatially distributed BioBolts in the region of interest record the neural activities and send signals to MasterBolt. We explore the intra-skin wireless communication between BioBolt nodes and MasterBolt, which allows extremely low power signal transmission under 5?ÃW. MasterBolt transmits all the collected neural signals from the neighboring BioBolts using single-channel FM modulation with a power budget of <1mW. This system gives flexibility and expandability for long-term chronic monitoring of neural signals.
Fully-implanted distributed wireless microsystems for epidural neural recording will provide a new tool to study and understand the collective brain activities for practical interface with computer, prosthetic devices and control of impaired body. The completed system developed in this project will expand the potential applications for brain-to-computer interfaces beyond simple cursor control and offer people with motor disabilities a good alternative to natural communication and movement. This project will train graduate and undergraduate students across the different disciplines to implement an important interface between brain and electronics.
We have developed a new platform concept, BioBolt, as a distributed wireless neural interface microsystem for minimally-invasive, fully-implantable epidural electrocorticographic (ECoG) recording. The BioBolt contains a stand-alone wireless link and can be easily placed on the dura mater through a small hole in the skull by simple operation procedure. Spatially distributed BioBolts in the region of interest can record neural activities and send signals through an intra-skin wireless communication link between the BioBolt nodes and a way-station. We have implemented an extremely low-power analog frontend circuits to amplify, digitize, and multiplex the neural signals in a small form factor. Total 32 channels could be simultaneously recorded with 600nW per channel. In addition to recording capability, we also implemented a stimulation function to complete the closed-loop system. The closed-loop neural interface system can monitor the brain activities as well as activate or deactivate the neurons by electrical stimulation. We have conducted in-vivo experiments for primates and rodents. The BioBolt could clearly distinguish β-waves from motor cortex in primate experiments. The recorded signals were transmitted from the monkey brain to a way-station located at 11cm apart from the implanted system through intra-skin wireless communication. We also performed experiments with an epilepsy induced rat. The epilepsy onset signals were successfully recorded, transmitted and identified by the custom algorithm. For closed-loop control stimulation, a trigger signal from an external processor could successfully induce on-chip electrical stimulation. This work has made significant achievement to realize a fully-implanted distributed wireless microsystem for neural recording, providing a new tool to study and understand the collective brain activities for practical interface with computer, prosthetic devices and control of impaired body. Neuroscience community can take advantage of the developed system to understand the correlation between neural signals and animal behavior as well as to implement prosthetic devices for restoring function to re-establish a link between the brain and the desired action through brain-computer interface (BCI) technology. The developed BioBolt system can be used for clinical use as an implantable devices monitor neural activities. A comprehensive closed-loop antiepilepsy system can address some practical limitations and unmet needs in the current technologies available in the market. There are two sectors for societal impact - diagnosis and treatment of intractable epilepsy and translational advancement of neuroprosthetics and neurorehabilitation to enable paralyzed patients suffering from spinal cord injury or stroke to communicate and interact with their environment. During the course of work, four Ph.D. students have been trained. The participating students were trained in low-power circuit design and characterization as well as in system integration and in-vivo animal experiments. Also, a couple of undergraduate students were involved and trained in the development and testing of intra-skin wireless telemetry. We displayed the BioBolt prototype devices in Tech Day Exhibition, an annual event that invites prospective students and their parents at the University of Michigan. We attracted many K12 students during the exhibition and contributed to captivate their interest in science and engineering. We presented six conference papers in IEEE international conferences and one journal article (and two more in preparation). We filed three patents related with the implementation of BioBolts.
