DBS Contact Selection via LFP Spectral Coherency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for selecting stimulation contacts in deep brain stimulation (DBS) are time-consuming and dependent on behavioral outcomes during programming sessions, necessitating improved systems and methods for optimizing contact selection.
Innovation Solution
A computing device applies a spatial filter to local field potential (LFP) recordings, calculates power spectral density (PSD), performs parametric approximations, selects frequency bands, calculates spectral coherency matrices, and computes eigenvector centrality to identify optimal contacts for stimulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional behavioral outcome-based methods are used for contact selection, then therapeutic effectiveness can be optimized, but the programming session becomes time-consuming
Solution Approach 1:
The system performs preliminary analysis of LFP recordings and calculates eigenvector centrality metrics before final contact selection is made. This advance processing identifies candidate contacts based on spectral coherency patterns, so that when programming is needed, the selection is already largely determined by objective data rather than requiring lengthy behavioral testing of each contact.
Solution Approach 2:
The patent replaces the mechanical/behavioral testing process with an automated computational system that analyzes LFP signals and calculates spectral coherency matrices. Instead of manually testing each contact's behavioral outcome, the system uses algorithmic processing of neural signals to objectively identify optimal contacts, dramatically reducing programming time while maintaining or improving selection accuracy.
2Productivity
If automated computational methods are used for contact selection, then programming time is reduced, but dependence on behavioral outcome data is decreased
Solution Approach 1:
The system introduces LFP spectral coherency analysis as an intermediary between raw neural signals and contact selection decisions. Rather than directly using behavioral outcomes or raw LFP data, the patent creates intermediate metrics (spectral coherency matrices, eigenvector centrality values) that objectively characterize contact quality. This intermediary layer provides quantitative, objective criteria that complement or replace subjective behavioral assessments.
Solution Approach 2:
The patent transforms the selection criteria from behavioral outcome parameters to spectral coherency parameters. By changing the fundamental parameters used for evaluation—from subjective behavioral measures to objective frequency-domain characteristics—the system enables automated, rapid contact identification that is not dependent on lengthy behavioral testing while still providing scientifically grounded selection criteria.
Data Source
AI summary
The present disclosure provides systems and methods for selecting contacts for use in deep brain stimulation (DBS). A computing device includes a processor and a memory device communicatively coupled to the processor. The memory device includes instructions that, when executed, cause the processor to apply a spatial filter to local field potential (LFP) recordings for a plurality of contacts of a DBS lead, calculate a power spectral density (PSD) for each contact from the filtered LFP for that contact, calculate a parametric approximation for each PSD, select at least one frequency band based on the parametric approximations, calculate a spectral coherency matrix for each of the at least one selected frequency band, and calculate an eigenvector centrality for each spectral coherency matrix to facilitate identifying a contact for stimulation.