Abstract and Introduction
Objective: Deep brain stimulation (DBS) is a safe and effective therapy for movement disorders, such as Parkinson's disease (PD), essential tremor (ET), and dystonia. There is considerable interest in developing "closed-loop" DBS devices capable of modulating stimulation in response to sensor feedback. In this paper, the authors review related literature and present selected approaches to signal sources and approaches to feedback being considered for deployment in closed-loop systems.
Methods: A literature search using the keywords "closed-loop DBS" and "adaptive DBS" was performed in the PubMed database. The search was conducted for all articles published up until March 2018. An in-depth review was not performed for publications not written in the English language, nonhuman studies, or topics other than Parkinson's disease or essential tremor, specifically epilepsy and psychiatric conditions.
Results: The search returned 256 articles. A total of 71 articles were primary studies in humans, of which 50 focused on treatment of movement disorders. These articles were reviewed with the aim of providing an overview of the features of closed-loop systems, with particular attention paid to signal sources and biomarkers, general approaches to feedback control, and clinical data when available.
Conclusions: Closed-loop DBS seeks to employ biomarkers, derived from sensors such as electromyography, electrocorticography, and local field potentials, to provide real-time, patient-responsive therapy for movement disorders. Most studies appear to focus on the treatment of Parkinson's disease. Several approaches hold promise, but additional studies are required to determine which approaches are feasible, efficacious, and efficient.
Deep brain stimulation (DBS) is a safe and effective therapy for movement disorders, such as Parkinson's disease (PD), essential tremor (ET), and dystonia.[2,7,15] The devices have been in clinical use for decades, providing invariant stimulation at a fixed spatial distribution (electrode configuration), amplitude, frequency, and pulse width. This "open-loop" therapy relies on the determination of effective stimulation parameters by a clinician. As our understanding of the mechanisms underlying this therapy and movement disorders in general expands, the shortcomings of this system are increasingly evident.
Although DBS provides effective treatment of the motor symptoms of diseases such as PD, side effects of therapy can include cognitive impairment and changes in speech, gait, and balance. While most patients deem such side effects tolerable, the current approach likely does not restore basal ganglia function to the greatest extent possible, given its static approach to therapy within an inherently dynamic system. Moreover, the use of constant stimulation provides stimulation in excess of what is clinically warranted. With the most-commonly implanted, nonrechargeable DBS systems, this excess power usage shortens battery life and exposes patients to the risks of surgical replacement of the implantable pulse generator.
While cardiac pacemaker devices capable of sensing and responding to patient physiology have been in clinical use for over 50 years, similar efforts to develop a "closed-loop" DBS device have been delayed; this is likely due to the complexity of brain signals and uncertainty surrounding the clinical significance of recordable brain activity. These roadblocks are increasingly surmountable, with advances in technology, development of applicable algorithms, and a greater understanding of neurophysiology.
Closed-loop DBS represents a subset of bidirectional brain computer interfaces; a comprehensive review of such systems is beyond the scope of this paper, which will focus on relatively well-established approaches to the treatment of movement disorders. The goal is to provide an overview of the features of a closed-loop system, with attention paid to signal sources and biomarkers, general approaches to feedback control, and clinical data when available.
Neurosurg Focus. 2018;45(2):e2 © 2018 American Association of Neurological Surgeons