Seizure Detection Algorithm Using Dynamic Heart Rate Weighting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current seizure detection methods based on heartbeat data face challenges in distinguishing between pathological and non-pathological changes, leading to false positives due to factors like physical activity and external influences.

Innovation Solution

A medical device uses a seizure detection algorithm that accounts for physical activity and external factors by calculating a modified background heart rate using a weighting factor, which adjusts based on activity data such as accelerometer and temperature data, to differentiate between seizure events and non-seizure heart rate increases.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If seizure detection is based solely on heartbeat threshold monitoring, then detection simplicity is maintained, but false positive rate increases due to inability to distinguish pathological from non-pathological heartbeat changes

Engineering Contradiction:
Improvedetection algorithm complexityVSAvoidseizure detection accuracy
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent combines multiple data sources (heartbeat data from ECG sensor and activity data from accelerometer) into a unified detection framework. The algorithm merges these inputs to compute a confidence score that integrates cardiac information with motion context, thereby improving detection reliability while managing complexity through systematic integration.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces activity data as an intermediary element that mediates between raw heartbeat data and seizure detection decisions. This intermediary provides contextual information about patient movement and posture, allowing the system to distinguish between heartbeat changes caused by activity versus those caused by seizures, thus reducing false positives.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the detection algorithm incorporates activity data and multiple data sources, then false positive rate decreases, but device complexity increases

Engineering Contradiction:
Improveseizure detection accuracyVSAvoiddetection algorithm complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic adjustment of detection parameters based on real-time activity levels. The algorithm adapts its sensitivity and threshold settings according to the patient's current activity state, allowing it to maintain high reliability across varying conditions without requiring overly complex fixed-rule systems. This dynamic approach enables the algorithm to simplify processing during low-activity periods while maintaining accuracy during high-activity periods.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes detection parameters (such as confidence thresholds and weighting factors) based on activity level measurements. By adjusting these parameters dynamically according to the patient's state, the system maintains optimal detection performance across different scenarios without requiring a fundamentally complex algorithmic structure, thus managing device complexity while preserving reliability.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3977922B1Detecting seizures based on heartbeat data
Publication Date: 2024.03.06 LIVANOVA USA INC
  • EP3977922B1 patent drawingFigure 1
  • EP3977922B1 patent drawingFigure 2
  • EP3977922B1 patent drawingFigure 3

AI summary

A method comprising the steps of receiving heartbeat data of a patient; receiving accelerometer data of the patient; determining, using the accelerometer data, that the patient is transitioning from a sleep state to an awake state; modifying a weighting factor based on the transition; determining modified heartbeat data by applying the weighting factor to at least a portion of the heartbeat data; and detecting a seizure event based on the modified heartbeat data.