Hearing Device Health Issue Detection via Activity Pattern Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current hearing devices and systems lack an effective method for early detection of health issues, such as cognitive decline or dementia, which can lead to delayed intervention and reduced treatment success.

Innovation Solution

A method and computer program implemented in hearing devices that utilize sound detectors, user detectors, and processing units to analyze audio and user activity patterns over time, employing machine learning algorithms like neuronal networks to identify deviations from healthy patterns and provide early detection signals.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If hearing devices collect and analyze user activity data over time to detect health issues, then detection accuracy improves, but device complexity increases

Engineering Contradiction:
Improvehealth issue detection accuracyVSAvoidhearing device complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the health detection system into multiple independent components: a hearing device that collects audio and activity data, a user device that receives and stores data, and a processing system that performs machine learning analysis. This segmentation allows the complex detection algorithms to run on external devices while keeping the hearing device itself relatively simple.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a user device (smartphone, tablet, or server) as an intermediary between the hearing device and the health detection system. This intermediary handles the complex data processing and machine learning operations, enabling accurate health issue detection without requiring the hearing device itself to be overly complex.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If hearing devices process large amounts of audio and sensor data continuously, then detection reliability improves, but energy consumption increases

Engineering Contradiction:
Improvehealth issue detection reliabilityVSAvoidhearing device energy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent extracts the computationally intensive data processing and machine learning operations from the hearing device and relocates them to external user devices or servers. This allows the hearing device to continuously collect data with minimal energy consumption while maintaining high detection reliability through comprehensive data analysis performed externally.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system implements periodic data collection and transmission, where the hearing device gathers audio and activity data over time intervals and transmits batches of data to the user device for processing. This periodic approach reduces continuous processing demands and energy consumption while maintaining reliable detection through accumulated data analysis.

Inventive Principle:
Principle #19Periodic action

3Loss of time

If the system monitors user activity patterns over extended periods, then early detection capability improves, but data storage requirements increase

Engineering Contradiction:
Improvetime to detectionVSAvoiddata storage requirements
Core Design Contradiction:
Loss of timeVSQuantity of substance

Solution Approach 1:

The system performs preliminary local processing and filtering of data at the hearing device and user device levels before storing or transmitting to long-term storage. This preliminary action reduces the volume of raw data that needs to be stored while preserving the essential information needed for early health issue detection through pattern recognition.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4156717A1Method and computer program for early detection of a health issue, controller for a hearing device, hearing device, and hearing system comprising a hearing device
Publication Date: 2023.03.29 SONOVA AG
  • EP4156717A1 patent drawingFigure 1
  • EP4156717A1 patent drawingFigure 2
  • EP4156717A1 patent drawingFigure 3~4

AI summary

A method for early detection of a health issue of a user wearing a hearing device (12) is provided. The hearing device (12) comprises at least one sound detector (20) for generating an audio signal, at least one user detector for generating a user detection signal, a processing unit (23) for modifying the audio signal, and at least one sound output component (24) for outputting the modified audio signal. The method comprises: receiving (S2) the audio signal and/or the user detection signal for a predetermined time interval; determining (S4) a current activity pattern of the user by evaluating the audio signal and/or the user detection signal received over the predetermined time interval; evaluating (S6) the current activity pattern; determining (S8) whether there is an upcoming health issue of the user depending on the evaluation; and providing (S10), if it is determined that there is an upcoming health issue of the user, an output signal indicative of the health condition.