Method for adaptive control of a medical device using Bayesian optimization
Systems and methods for adaptively controlling an electrical stimulation device, such as a closed-loop stimulation device, based in part on a Bayesian optimization of the operational parameters of the device are described. An adaptive dual control of the stimulation device can be provided. In a first control loop parameters are extracted from signals recorded from the subject by the stimulation device, and in a second control loop a Bayesian optimization is implemented with a hardware processor and memory to compute updated operational parameters for the stimulation device. As noted, the stimulation device is an electrical stimulation device, and may be a closed-loop stimulation device. Such devices can be used for deep brain stimulation (“DBS”), cardiac resynchronization therapy (“CRT”), and other electrophysiological stimulation applications.
1. An adaptive controller for controlling a controllable medical device, comprising:
an input that receives feedback data representative of a treatment response or effect in a subject;
a processor in communication with the input and programmed to:
receive the feedback data from the input and generate a posterior distribution therefrom;
estimate an acquisition function from the posterior distribution; and
generate updated control parameter settings based on the acquisition function;
a memory in communication with the input and the processor, wherein the memory stores instructions for generating updated control parameter settings, the feedback data received from the input, and the updated control parameter settings generated by the processor;
an output that communicates the updated control parameter settings to a controllable medical device comprising an electrical stimulation device having a pulse generator and at least one electrode in communication with the pulse generator;
wherein the processor is programmed to generate a response surface based on the feedback data using a probit function;
wherein the feedback data comprise user preferences between two different control parameter settings;
wherein the posterior distribution is generated using the response surface generated based on the feedback data using the probit function; and
wherein the processor receives control signals generated based on the updated control parameters settings and sends the controls signals to the pulse generator of the controllable medical device by the output; and
wherein the pulse generator of the controllable medical device receives the control signals from the output and in response thereto generates electrical signals and sends the electrical signals to the at least one electrode of the controllable medical device to generate an electrical stimulation based on the updated control parameter settings.
2. The adaptive controller as recited in claim 1 , wherein the feedback data received from the input further comprise behavior metrics, and wherein the processor is programmed to also generate the response surface based on the behavior metrics.
3. The adaptive controller as recited in claim 1 , wherein the feedback data received from the input further comprise at least one of physiological data or behavioral metrics.
4. The adaptive controller as recited in claim 3 , wherein the physiological data comprises at least one of neural signals, cardiac signals, or chemical signals.
5. The adaptive controller as recited in claim 4 , wherein the physiological data comprises chemical signals including at least one of insulin levels or glucose levels.
6. The adaptive controller as recited in claim 1 , wherein the user preferences comprise user preference responses to a questionnaire.
7. The adaptive controller as recited in claim 1 , wherein the electrical stimulation device comprises at least one of a peripheral nerve stimulator, a central nervous system stimulator, a cardiac pacemaker, or a cardiac resynchronization therapy (CRT) device.
8. A system, comprising:
a controllable medical device comprising an electrical stimulation device having a pulse generator and at least one electrode in communication with the pulse generator and in contact with a subject to deliver electrical stimulation according to control parameter settings to the subject patient; and
a controller, the controller comprising:
(i) an input that receives feedback data representative of a treatment response or effect in the subject patient of delivered electrical stimulation, wherein the feedback data comprises user preference relative to electrical stimulation delivered according to a first control parameter setting and different electrical stimulation delivered according to a second control parameter setting;
(ii) a processor in communication with the input and programmed to:
(A) receive the feedback data from the input;
(B) generate a response surface based on the feedback data using a probit function;
(C) generate a posterior distribution from the response surface;
(D) estimate an acquisition function from the posterior distribution; and
(E) generate updated control parameter settings based on the acquisition function;
(iii) a memory in communication with the input and the processor, wherein the memory stores instructions for receiving the feedback, generating the response surface, generating the posterior distribution, estimating the acquisition function, generating updated control parameter settings and storing control parameter settings and updated control parameter settings;
(iv) an output that communicates the updated control parameter settings to the controllable medical device;
wherein the controllable medical device receives the updated control parameters settings, and the pulse generator delivers electrical stimulation to the subject patient according to the updated control parameter settings.