The current state of neuromodulation can be cast in a classical dynamic control framework such that the nervous system is the classical "plant", the neural stimulator is the controller, tools to collect clinical data are the sensors, and the physician's judgment is the state estimator. This framework characterizes the types of opportunities available to advance neuromodulation. In particular, technology can potentially address two dominant factors limiting the performance of the control system: "observability," the ability to observe the state of the system from output measurements, and "controllability," the ability to drive the system to a desired state using control actuation. Improving sensors and actuation methods are necessary to address these factors. Equally important is improving state estimation by understanding the neural processes underlying diseases. Development of enabling technology to utilize control theory principles facilitates investigations into improving intervention as well as research into the dynamic properties of the nervous system and mechanisms of action of therapies. In this paper, we provide an overview of the control system framework for neuromodulation, its practical challenges, and investigational devices applying this framework for limited applications. To help motivate future efforts, we describe our chronically implantable, low-power neural stimulation system, which integrates sensing, actuation, and state estimation. This research system has been implanted and used in an ovine to address novel research questions.

An implantable bi-directional brain-machine interface (BMI) prototype is presented. With sensing, algorithm, wireless telemetry, and stimulation therapy capabilities, the system is designed for chronic studies exploring closed-loop and diagnostic opportunities for neuroprosthetics. In particular, we hope to enable fundamental chronic research into the physiology of neurological disorders, define key electrical biomarkers related to disease, and apply this learning to patient-specific algorithms for therapeutic stimulation and diagnostics. The ultimate goal is to provide practical neuroprosthetics with adaptive therapy for improved efficiency and efficacy.

The monitoring of neuronal activity could potentially expand the diagnostic and therapeutic capabilities of neuroprosthesis. The challenge of designing sensing and control systems is two-fold: first, the signal input must be robust for chronic recording; second, the circuit architecture must be capable of achieving signal processing, algorithm control, and telemetry with a limited power budget. The first requirement should be met by measuring field potentials, which represent ensemble behavior in a neural network and can be measured chronically. For the second requirement, architecting an effective solution requires identification of the key information of interest and partitioning the signal chain to play to the strengths of analog vs. digital processing. For many neurological states of interest, information 'biomarkers' are encoded as low frequency power fluctuations within well-defined frequency bands of field potentials, similar to the amplitude modulation found in an AM radio. Recognizing this similarity, the feasibility prototype adapts a chopper-stabilized instrumentation amplifier to act as a superheterodyning AM receiver for brain signals. Since the physiological power fluctuations are generally orders of magnitude slower than the frequency at which they are encoded, the use of efficient analog preprocessing greatly reduces the overall energy requirements for implementing a complete mixed- signal system. Since the science of field potentials is rapidly evolving, the superheterodyning chopper is advantageous given its flexibility and immunity to process, temperature, and mismatch variations. This paper will discuss the design of a complete system prototype for a neurostimulator research tool; the design has a noise floor of under 2muVrms and a total system current of 25muW/processing channel (1.8V supply) while performing biomarker extraction, algorithmic processing and control, and data loop recording.

We describe two prototype micropower sensors that potentially help enable neuroprosthetics for the treatment of chronic disease. The first sensor is an EEG instrumentation amplifier for the measurement of neurological field potentials in physiologically relevant bandwidths. The second sensor is a three-axis accelerometer for measuring posture, activity, and tremor. Both sensor interfaces use dynamic offset cancellation techniques-chopper stabilization for the EEG amplifier, correlated-double-sampling for the accelerometer-to reject low frequency excess noise that might otherwise corrupt the key physiological signals. To be compatible with chronic implantation, each sensor interface must operate with less than 2 mu W of power from a single battery. Using accepted metrics, these sensors represent the state-of-the-art for noise efficiency.

This paper describes a micropower instrumentation amplifier that enables chronic biopotential sensing in battery powered applications. The amplifier is chopper-stabilized to eliminate excess noise from 1/f or popcorn processes, insuring the highest fidelity of signal measurement for diagnostic analysis. The circuit consumes 2.0μW of power from a 1.8V supply, with a noise floor of 0.94μVrms in a bandwidth from 0.05 to 100Hz; the resulting noise-efficiency factor of 3.6 is the lowest published to date. The specific implementation of chopper stabilization also provides rail-to-rail inputs and 100dB CMRR at low frequency. A digitally programmable on-chip high pass filter (0.05Hz, 0.5Hz and 2.5Hz) is used to suppress front-end electrode offsets while maintaining relevant physiological data. Although the focus of this paper is on biopotential sensing, the circuit architecture is also useful for a variety of micropower sensor interfaces using synchronous demodulation.