Deep Brain Stimulation Array Programming via Convex Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional deep brain stimulation (DBS) systems with high-density electrode arrays face challenges in efficiently and effectively programming and controlling stimulation settings due to the large number of possible combinations, which can be time-consuming and resource-intensive, and often rely on empirical data that does not account for individual patient anatomy.
Innovation Solution
The use of model-based and objective methods that employ convex optimization to determine optimal electrode stimulation settings based on patient-specific brain and lead geometry data, maximizing activation function values while minimizing side effects, and incorporating user-specific battery power constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional manual programming methods are used to test stimulation settings, then the system can be programmed with traditional four electrode DBS leads, but the process becomes time-consuming and resource-intensive when applied to high-density electrode arrays with many possible combinations
Solution Approach 1:
The system uses pre-computed efficacy probability maps stored in a lookup table that were generated from previous empirical patient data and simulations. Instead of performing time-consuming manual testing or new simulations for each patient, the system copies and applies pre-existing probability data to rapidly determine optimal electrode settings for high-density arrays.
Solution Approach 2:
Efficacy probability maps and simulation results are pre-computed and stored in a lookup table before clinical use. This preliminary action allows the system to quickly retrieve and apply optimal stimulation settings without performing time-consuming calculations or manual testing during the programming session.
2Reliability
If conventional brain mapping with empirical patient data is used, then the system can create efficacy probability maps, but the approach does not account for individual patient anatomy and requires extensive data compilation
Solution Approach 1:
The system introduces a pre-computed efficacy probability map as an intermediary between the electrode configuration and the stimulation outcome. This map serves as a lookup table that translates electrode settings into predicted efficacy without requiring direct compilation of empirical patient data for each new case, simplifying the process while maintaining reliability.
Solution Approach 2:
The system changes the parameter representation from raw empirical patient data to pre-computed efficacy probability values stored in a lookup table. This parameter transformation allows rapid retrieval of optimal settings based on electrode configuration without requiring complex data compilation or processing during clinical programming.
3Measurement precision
If computational neuron models with large numbers of simulations are used, then the system can predict optimal electrode settings, but the approach requires vast computational resources that may not be present in a clinical setting
Solution Approach 1:
Instead of performing computationally intensive neuron model simulations during clinical programming, the system copies pre-computed simulation results and stores them in a lookup table. This allows the clinical system to retrieve prediction data with minimal computational resource consumption while maintaining the accuracy benefits of extensive simulations.
Solution Approach 2:
Computationally intensive neuron model simulations are performed in advance to generate efficacy probability maps and populate the lookup table. This preliminary computational action transfers the heavy processing burden to a pre-processing stage, allowing the clinical system to operate with minimal computational resources while still accessing accurate predictions.
4Adaptability or versatility
If the number of electrodes in a DBS array is increased to enable current steering, then the system can target small or complex-shaped brain regions, but the number of possible stimulation combinations becomes unwieldy for manual programming
Solution Approach 1:
The system uses pre-computed efficacy probability data from the lookup table to directly determine optimal settings for high-density electrode arrays. This copying approach eliminates the need for clinicians to manually evaluate the large number of possible stimulation combinations, making high-density array programming as easy as traditional four-electrode leads.
Solution Approach 2:
The system performs self-service by automatically retrieving optimal electrode settings from the pre-computed lookup table based on the specific high-density array configuration and target region. This automated retrieval process eliminates the need for manual programming effort, allowing the system to handle complex high-density arrays without increasing operational difficulty.
Data Source
AI summary
Systems, methods and apparatus for determining and/or programming stimulation settings for a pulse generator capable of steering current to a deep brain stimulation array are disclosed. Individual patient brain geometry and lead specific geometry data can be used to generate a maximum activation function curve. Optimization methods can be used to find stimulation settings that are as close as possible to achieving the value of the maximum activation function curve.