Automated deep brain stimulation programming with safety constraints for tremor suppression in patients with Parkinson’s disease and essential tremor

Abstract

Objectives. Deep brain stimulation programming for movement disorders requires systematic fine tuning of stimulation parameters to ameliorate tremor and other symptoms while avoiding side effects. DBS programming can be a time-consuming process and requires clinical expertise to assess response to DBS to optimize therapy for each patient. In this study, we describe and evaluate an automated, closed-loop, and patient-specific framework for DBS programming that measures tremor using a smartwatch and automatically changes DBS parameters based on the recommendations from a closed-loop optimization algorithm thus eliminating the need for an expert clinician. Approach. Bayesian optimization which is a sample-efficient global optimization method was used as the core of this DBS programming framework to adaptively learn each patient’s response to DBS and suggest the next best settings to be evaluated. Input from a clinician was used initially to define a maximum safe amplitude, but we also implemented ‘safe Bayesian optimization’ to automatically discover tolerable exploration boundaries. Results. We tested the system in 15 patients (9 with Parkinson’s disease and 6 with essential tremor). Tremor suppression at best automated settings was statistically comparable to previously established clinical settings. The optimization algorithm converged after testing 15.1±0.7 settings when maximum safe exploration boundaries were predefined, and 17.7±4.9 when the algorithm itself determined safe exploration boundaries. Significance. We demonstrate that fully automated DBS programming framework for treatment of tremor is efficient and safe while providing outcomes comparable to that achieved by expert clinicians.

Proceedings of the 10th annual deep brain stimulation think tank: Advances in cutting edge technologies, artificial intelligence, neuromodulation, neuroethics, interventional psychiatry, and women in neuromodulation

“…These strategies can provide quantifiable metrics for disease or symptom severity, treatment efficacy, target engagement, and side effect severity. Automated programming for tremor suppression in patients with Parkinson’s disease (PD) and essential tremor has also been successfully implemented and piloted using kinematic signals ( Haddock et al, 2018 ; Sarikhani et al, 2022 ). Bayesian optimization with safety constraints has further enabled safe and efficient programming with comparable tremor outcomes when compared to programming that is performed by expert clinicians ( Sarikhani et al, 2022 ).…”

“…Automated programming for tremor suppression in patients with Parkinson’s disease (PD) and essential tremor has also been successfully implemented and piloted using kinematic signals ( Haddock et al, 2018 ; Sarikhani et al, 2022 ). Bayesian optimization with safety constraints has further enabled safe and efficient programming with comparable tremor outcomes when compared to programming that is performed by expert clinicians ( Sarikhani et al, 2022 ). Image-guided programming aims to use patient-specific computational DBS activation models to choose stimulation settings in order to maximize computationally predicted stimulation effects in the target region and to avoid side-effects.…”

A Paradigm Shift in the Management of Patients with Parkinson’s Disease

“…For the intensified therapy options, the number of choices is actually increasing, as we will possibly have subcutaneous levodopa [50] and other forms of continuous delivery or a closed-loop deep brain stimulation [51], together with the existing levodopa-carbidopa (with or without entacapone) intestinal gel pump, deep brain stimulation, and apomorphine options [52]. Currently, there is an underuse of advanced therapies and a reason behind this is that physicians have difficulties to identify the right patient candidates [53].…”

In silico development and validation of Bayesian methods for optimizing deep brain stimulation to enhance cognitive control

“…In other DBS applications, Bayesian approaches have been suggested and demonstrated to be an useful tool to optimize DBS settings [41][42][43][44][45] in a closed loop neuromodulation [41]. Used Bayesian optimization to determine the best DBS parameters that reduced beta power in a basal ganglia-thalamocortical computational model of Parkinson's disease (PD).…”

“…Based on the participant's expressed preferences for stimulation settings [46], used Bayesian preference learning to find individualized optimal stimulation patterns. Recently, Bayesian algorithms have been demonstrated to be safe and viable options in humans, as in [43,44], with additional constraint to the objective function to be optimized for PD. Generalizing beyond movement disorders, we recently demonstrated Bayesian optimization in neurostimulation for pain [47].…”