Method of assessing human fall risk using mobile systems
A method for determining the motive instability of an individual using foot pressure, foot speed and foot direction data collected from sensors on shoes. The sensed data is used to determine the minimum number and the placement of pressure sensors in the shoe. The data from the sensors is processed to extract spatial and temporal parameters as desired. The data is grouped into segments based on a segmentation rule. The trend in each segment is determined. The variability of the trend in each segment is determined. The risk of fall is computed on the basis of the trend and variance. The computation is adjustable by emphasizing certain parameters in order to tailor the instability assessment to a specific individual.
1. A method for determining fall risk instability for a specific individual comprising: measuring data from sensors located in the shoes of an individual in motion; analyzing the measured data to determine trends in the data, the analyzing including, for a predetermined number of step cycles, correlating sensor data between each two consecutive ones of the step cycles; and using results of the correlations, determining a trend of a feature of interest; analyzing the measured data to determine variances in the trends; and determining fall risk instability based on the determined variances.
2. The method of claim 1 further comprising generating a warning signal when the determined fall risk instability exceeds a predetermined threshold.
3. The method of claim 1 wherein the data from the sensors, located in the shoes of the individual in motion provide data to determine the parameter step length, dual stance time, cadence, time for each taken step, gait, stride length, stance-to-swing phase ratio, pressure correlation, center line of pressure, or planar pressure distribution.
4. The method of claim 3 wherein each determined parameter is assigned a weighting factor based on the individual characteristics of the individual, or his environment.
5. The method of claim 1 , further comprising segmenting the measured data into classes according to a predetermined rule.
6. The method of claim 5 wherein the predetermined rule for segmentation may be a fixed time interval, type of activity, or geographical location.
7. The method of claim 1 wherein the measuring the data comprises extracting temporal features and spatiotemporal features from the measured data.
8. The method of claim 7 wherein the determining the trends comprises determining trends in the temporal features and spatiotemporal features.
9. The method of claim 8 where the determining the variances comprises determining variances in the trends of the temporal features and spatiotemporal features.
10. The method of claim 9 where the determining the fall risk instability comprises analyzing the variances in the trends of the temporal features and spatiotemporal features.
11. A method, comprising:
receiving sensor data from sensors in a shoe;
from the sensor data, detecting the beginning and end of each of a plurality of step cycles;
for a predetermined number of step cycles, correlating sensor data between each two consecutive ones of the step cycles;
using results of the correlations, determining a trend of a feature of interest;
using the results of the correlations, determining a variation from the trend; and
from the trend and variation, calculating a risk of instability.
12. The method of claim 11 , wherein the sensor data includes temporal and spatial data.
13. The method of claim 11 , wherein determining the trend includes performing a multi-pass interpolation using a pre-defined window to determine a relative average path.
14. The method of claim 11 , wherein calculating a risk of instability includes calculating a sum of a variance of temporal parameters and a sum of a variance of spatiotemporal parameters.
15. The method of claim 11 , further comprising: identifying a subset of the sensors, wherein the identification of a sensor to include in the subset of the sensors is based on the relative influence of the sensor data from that sensor on the determination of the feature of interest.
16. The method of claim 15 , wherein identifying the subset of the sensors includes using a clustering technique.
17. The method of claim 15 , wherein receiving the sensor data includes receiving sensor data from the subset of the sensors and not receiving sensor data from sensors outside of the subset of the sensors.
18. The method of claim 11 , wherein the sensor data is received through a wireless interface.
19. The method of claim 11 , wherein the sensor data is received and processed in the shoe.