Systems and methods for cooperative invasive and noninvasive brain stimulation
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11400300-B2
Publication Date
2022-08-02
Expiration Date
Abstract
Methods and systems for optimizing invasive and noninvasive brain stimulation are described herein. In a particular embodiment, methods and systems for a combinatorial, iterative approach to modify behavior are presented wherein deep brain stimulation (DBS) and other brain stimulation therapies are implemented in combination with monitoring the brain activity of an individual to optimize the effectiveness of the combinatorial approach to modify behavior. Methods described herein are iterative and systems described herein are utilized in iterative fashion. In a particular embodiment, modifying behavior provides a therapy for an individual in need thereof.
Core Innovation
The described invention provides systems and methods for optimizing cooperative invasive deep brain stimulation (DBS) and non-invasive brain stimulation using continuous brain activity monitoring and closed-loop adjustment. A stimulation device administers at least one stimulus configured to modulate brain electrical activity of an individual while the individual is performing a particular activity, and the stimulus provides a specific stimulation pattern to promote the ability of the individual to perform the particular activity.
The invention uses an apparatus worn on a head of the individual with at least one sensor configured to detect the particular activity performed and brain electrical activity associated with the particular activity, with continuous detection while the individual performs the particular activity. A specifically programmed computer system projects, in real time, the detected brain electrical activity onto a denoised optimal set of wavelet packet atoms to obtain a particular set of projections, and continuously assesses the brain electrical activity relative to representative brain electrical activity of a plurality of individuals performing the particular activity by applying at least one machine learning algorithm trained on the collected brain electrical activity.
The system continuously determines a relationship between the particular activity, the detected brain electrical activity, the real-time assessment relative to the representative activity using the machine learning algorithm, and the at least one stimulus. Based on that relationship, the system continuously causes adjustment of the specific stimulation pattern in the stimulus from the stimulation device to promote the ability of the individual to perform the particular activity. The description further relates brain activity with additional physiological and environmental parameters and generates visual indicators and feedback outputs based on detected changes.
Claims Coverage
The document contains two independent claims. Each claim centers on a closed-loop architecture that continuously detects activity-associated brain electrical activity, projects denoised wavelet packet representations, applies machine learning relative to representative data from multiple individuals, determines a relationship including the stimulus, and continuously adjusts the stimulation pattern to promote the ability to perform the particular activity.
Real-time closed-loop stimulation based on denoised wavelet packet projections and multi-individual machine learning
A system having a stimulation device that modulates brain electrical activity during performance of a particular activity, together with a head-worn apparatus for continuously detecting the particular activity and associated brain electrical activity, and a programmed computer system that continuously projects the detected brain electrical activity onto a denoised optimal set of wavelet packet atoms, continuously assesses the brain electrical activity relative to representative brain electrical activity of a plurality of individuals by applying at least one machine learning algorithm trained by the plurality, continuously determines a relationship between the particular activity, the detected brain electrical activity, the assessment relative to the representative activity, and the at least one stimulus, and continuously causes adjustment of the specific stimulation pattern based on the relationship to promote the ability to perform the particular activity.
Real-time method for continuous sensing, wavelet-packet projection, machine-learning assessment, relationship determination, and stimulation-pattern adjustment
A method that detects a particular activity performed by an individual and brain electrical activity of the individual associated with the particular activity, administers at least one stimulus to modulate the brain electrical activity while the individual performs the particular activity with a specific stimulation pattern to promote the ability, continuously detects and projects the brain electrical activity in real time onto a denoised optimal set of wavelet packet atoms to obtain projections, continuously assesses the brain electrical activity relative to a representative brain electrical activity of a plurality of individuals by applying at least one machine learning algorithm trained by the plurality, continuously determines a relationship between the particular activity, the detected brain electrical activity, the assessment relative to the representative activity, and the at least one stimulus, and continuously adjusts the specific stimulation pattern in the stimulus based on the relationship to promote the ability to perform the particular activity.
Across both independent claims, the core coverage is a real-time closed-loop stimulation approach. The system and method continuously derive denoised wavelet packet projections from activity-associated brain electrical activity, perform machine-learning assessment relative to representative data from multiple individuals, determine a relationship that includes the stimulus, and continuously adjust the specific stimulation pattern to promote ability during ongoing activity performance.
Stated Advantages
Promote the ability of the individual to perform the particular activity.
Continuously adjust the specific stimulation pattern in real time based on the determined relationship between particular activity, detected brain electrical activity, assessment relative to representative activity, and the at least one stimulus.
Optimize cooperative invasive DBS and non-invasive brain stimulation using continuous brain activity monitoring and closed-loop adjustment.
Documented Applications
Promoting walking ability in an individual having Parkinson’s disease.
Promoting the ability to resume walking in an individual having gate freeze.
Promoting the ability to perform an anxiety provoking activity in an individual having an anxiety disorder by reducing stress levels in response to the activity.
Application to disease or disorder contexts that impair the individual’s ability to perform the specified activity, including Parkinson’s disease, tremors, motor dysfunction, dyskinesia, gate freeze, epilepsy, migraine headaches, pain, anxiety, depression, mood swings, attention deficit disorders, sleep disorders, or cognitive decline disorders.
Interested in licensing this patent?