Fall Detection Using Kinetic Energy Signatures

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing fall detection devices worn on the wrist, particularly those using inclinometers and accelerometers, face challenges in differentiating between falls and trivial impacts, leading to high false alert rates and inconvenience, especially for daily activities.

Innovation Solution

A method utilizing an accelerometric sensor to estimate kinetic energy in three-dimensional space, combined with cardiovascular sensors, employs a qualitative signature comparison using Mahalanobis distance to differentiate between falls and trivial events, ensuring reliable detection without unnecessary alarms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a fall detector is worn on the wrist, then it is easy to wear and non-intrusive, but it generates false alerts due to inability to differentiate falls from trivial impacts

Engineering Contradiction:
Improveease of wearVSAvoidfalse alert rate
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent segments the detection task into multiple independent sensor measurements (acceleration magnitude, frequency content, duration) and processes them through separate thresholds and algorithms. This segmentation allows the system to distinguish fall characteristics from trivial impacts by analyzing multiple dimensions of motion data independently, reducing false alerts while maintaining wrist-worn convenience

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from simple quantitative threshold comparison to qualitative signature comparison by adding temporal and spectral dimensions to the acceleration data. By analyzing the time-windowed acceleration signature across multiple frequencies and comparing it against reference fall signatures, the system gains the ability to differentiate falls from trivial impacts without changing the physical placement of the device

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Device complexity

If a fall detector uses quantitative threshold comparison, then it is simple to implement, but it cannot differentiate between falls and trivial impacts leading to high false alerts

Engineering Contradiction:
Improveimplementation simplicityVSAvoidfall detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary actions by pre-computing and storing reference acceleration signatures for falls in a database before actual detection occurs. During operation, the system extracts the current acceleration signature and compares it against these pre-stored references using pattern matching algorithms, enabling accurate fall detection without complex real-time calculations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a copy of the acceleration signal by extracting its signature characteristics (magnitude, frequency, duration) and storing this simplified representation in a database. The actual detection process then involves comparing this copied signature against stored reference signatures, transforming the complex continuous signal into discrete comparable patterns that enable accurate differentiation between falls and trivial impacts

Inventive Principle:
Principle #26Copying

3Reliability

If the device is placed near the centre of gravity, then fall detection reliability is maximized, but it becomes inconvenient for daily activities like going to the toilet, getting dressed or sleeping

Engineering Contradiction:
Improvefall detection reliabilityVSAvoidconvenience for daily activities
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent makes the detection system dynamic by continuously adapting its sensitivity thresholds and activation criteria based on the wear context. The system monitors motion patterns to distinguish between intentional movements (toilet visits, dressing) and accidental falls, adjusting its detection parameters in real-time to maintain high reliability while accommodating daily activities

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the detection parameters by transitioning from fixed quantitative thresholds to adaptive qualitative pattern recognition. The system modifies its sensitivity settings based on the extracted acceleration signature characteristics and compares them against learned fall patterns, enabling reliable fall detection regardless of wear location while reducing false alerts during normal daily activities

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The solution achieves a high reliability in fall detection with a 91% success rate, reducing false alerts and improving user experience by distinguishing between falls and everyday activities, compared to prior wrist-worn devices which had a 65% reliability.

Implementation Method 1

a sensor (10) providing an accelerometric signal making it possible to obtain an approximation of the kinetic energy in the three dimensional space

Methodology Applied
Scientific EffectAccelerometry: Accelerometer

Data Source

PatentEP1974662B8Fall detector
Publication Date: 2014.07.16 CSEM CENTRE SUISSE D ELECTRONIQUE ET DE MICROTECHNIQUE SA

AI summary

The invention relates essentially to a method of detecting a fall of a person, characterized in that it comprises the following steps: - obtaining a signature St at an instant t, representing the distribution over a time window Δt of the kinetic energy, measured in three dimensions by an accelerometric sensor, - normalising the said signature, - comparing it with a reference signature obtained for a representative sample of falls, by means of similarity measurements, - analysing the result of the comparison, and - possibly, triggering a signal warning of a fall. A similar method based on cardiovascular parameters can be combined with the above processing to improve its reliability.