Real-Time Brain Signal Prediction for Closed-Loop Stimulus Timing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Influencing biological processes, particularly brain-related signals, is challenging, especially for faster changing signals, as existing technologies struggle to effectively manipulate these processes.

Innovation Solution

A closed-loop system comprising a recording assembly for brain-related signals, a stimulus generator, and a computer assembly that performs real-time curve fitting, prediction, and stimulus delivery based on predefined patterns in the brain signal.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If real-time curve fitting and prediction is performed on brain-related signals, then stimulus timing precision is improved, but computational complexity and processing time increase

Engineering Contradiction:

Improvestimulus timing precisionVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the brain signal into segments and performs curve fitting on specific segments rather than the entire signal. This allows real-time processing by focusing computational resources on relevant time windows, improving stimulus timing precision while managing computational complexity through selective processing of signal segments.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs curve fitting and prediction calculations in advance based on recorded brain signal patterns. By pre-computing the relationship between brain signal features and stimulus effectiveness, the system can quickly determine optimal stimulus timing without performing complex calculations at the moment of stimulus delivery, thus reducing real-time computational burden while maintaining precision.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If stimulus delivery is synchronized with brain signal phases, then biological process influence is improved, but response time for stimulus delivery decreases

Engineering Contradiction:

Improvebiological process influenceVSAvoidresponse time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements a feedback loop where brain signal recording continuously informs stimulus delivery timing. The system monitors brain signal phases in real-time and adjusts stimulus delivery to match optimal moments identified through curve fitting and prediction. This feedback mechanism ensures reliable biological process influence while minimizing response time by using ongoing signal data rather than relying on pre-programmed timing.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent employs dynamic adjustment of stimulus timing based on real-time brain signal characteristics. Rather than using fixed timing schedules, the system continuously adapts stimulus delivery to match the current state of brain oscillations detected through curve fitting. This dynamic approach optimizes biological process influence by synchronizing with actual brain state while reducing response time through adaptive rather than predetermined timing.

Inventive Principle:

Principle #15Dynamics

3Productivity

If curve fitting is performed on the most recent data segment, then real-time processing capability is improved, but data processing volume increases

Engineering Contradiction:

Improvereal-time processing capabilityVSAvoiddata processing volume

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The patent processes brain signal data in segmented portions rather than handling the complete signal at once. By dividing the continuous signal into recent data segments and performing curve fitting on these segments, the system achieves real-time processing capability while reducing the immediate data processing volume to manageable portions that can be handled efficiently by the computational system.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies curve fitting to the most recent data segment rather than the entire recorded signal. This partial action approach focuses computational resources on the most relevant time window for stimulus timing, improving real-time processing capability by avoiding the excessive processing burden of analyzing the complete signal history while maintaining sufficient accuracy for effective stimulus delivery.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12310739B2Assembly, method and computer program product for influencing a biological process

Publication Date: 2025.05.27 DEEP SLEEP TECHNOLOGIES BV

AI summary

The invention provides an assembly comprising:a recording assembly for recording a time-based brain-related signal;a stimulus generator for providing a stimulus, anda computer assembly, functionally coupled to said recording assembly and to said stimulus generator, said computer assembly comprising:a memory for storing at least a data segment of said time-based brain-related signal during recording of said time-based brain-related signal, anda computer program which, when running on said computer assembly, functionally real-time performs:retrieving a most-recent data segment of said stored data segment of said time-based brain-related signal;fitting at least one curve to said retrieved most-recent data segment;predicting a future continuation of said most-recent data segment using said at least one curve fitted to said most-recent data segment;detecting a predefined pattern in said predicted future continuation for predicting occurrence of said predefined pattern, and defining a predicted event time of said predefined pattern, said predicted event time being in the future with respect to said most-recent data segment, andactuating said stimulus generator for providing a stimulus within a predefined event time window of said predicted event time.