Fall Risk Assessment Using Punctuated Equilibrium Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting and preventing falls in seniors are inadequate, as they often rely on post-fall diagnostics and do not account for subtle balance changes influenced by lifestyle and health factors, making it difficult to assess fall risk effectively outside of real-time scenarios.
Innovation Solution
A system using machine learning algorithms, specifically the Hidden Markov Model, to classify postural states and calculate fall risk based on center of pressure data from load sensors, incorporating advanced metrics like time to first equilibrium, equilibria distance, and directional equilibria, to provide a composite balance score and fall risk classification without placing individuals in risky positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional fall risk assessment tests place individuals in challenging positions to gauge stability, then fall risk can be assessed, but the individual is placed in a risky position that may cause a fall
Solution Approach 1:
The patent uses an intermediary system (force plates with pressure sensors and machine learning algorithms) to indirectly assess fall risk by measuring center of pressure data during normal standing, rather than directly challenging the individual's balance. This mediator allows accurate assessment without exposing the individual to harmful fall risks.
Solution Approach 2:
The patent replaces the mechanical challenge-based assessment system with a sensor-based measurement system. Instead of mechanically challenging balance through challenging positions, the system uses force plates to measure subtle center of pressure fluctuations during normal standing, substituting mechanical challenge with electronic sensing and computational analysis.
2Device complexity
If postural stability is assessed using basic balance measures, then assessment is simple, but the predictive power for falls is insufficient
Solution Approach 1:
The patent segments the assessment into multiple components: center of pressure data collection, hidden Markov model processing, postural state classification, and composite balance score calculation. This segmentation allows the system to maintain simplicity in data collection while achieving high predictive accuracy through sophisticated multi-stage analysis.
Solution Approach 2:
The patent transitions from basic two-dimensional balance assessment to a multi-dimensional analysis by incorporating temporal dynamics through hidden Markov models. The system analyzes not just position but also the sequence and probability of postural states over time, adding a temporal dimension that significantly improves fall prediction accuracy.
3Productivity
If medical diagnostics and interventions are sought only after a fall or serious balance problem occurs, then resource utilization is efficient, but preventive intervention opportunities are lost
Solution Approach 1:
The patent enables preliminary action by detecting subtle balance changes that occur decades in advance of actual falls. The system identifies early warning signs through continuous monitoring of center of pressure data and predicts future fall risk, allowing preventive interventions to be initiated before falls occur, thereby improving prevention effectiveness while maintaining efficient resource utilization through targeted early intervention.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A person's fall risk may be determined based on machine learning algorithms. The fall risk information can be used to notify the person and/or a third party monitoring person (e.g. doctor, physical therapist, personal trainer, etc.) of the person's fall risk. This information may be used to monitor and track changes in fall risk that may be impacted by changes in health status, lifestyle behaviors or medical treatment. Furthermore, the fall risk classification may help individuals be more careful on the days they are more at risk for falling. The fall risk may be estimated using machine learning algorithms that process data from load sensors by computing basic and advanced punctuated equilibrium model (PEM) stability metrics.