Personalized fall detector
View Patent ↗A method and system for training a fall detection classifier using subject-specific movement data. Movement data responsive to a subject's movements during everyday activities are obtained over a predetermined data collection period. For each detected event in the movement data, values for one or more parameters that may (together or individually) indicate a fall are obtained. The obtained values are used to generate a subject-specific probability distribution for non-fall events. A fall event probability distribution is obtained using the non-fall event probability distribution. This fall event probability distribution can then be subsequently processed, with reference to a threshold value, to distinguish fall events from non-fall events.
1. A computer-implemented method executed on data processing hardware that causes the data processing hardware to perform operations comprising:
obtaining movement data representing movement of a subject over a predetermined data collection period, the movement data comprising one or more parameters, each respective parameter comprising a corresponding value;
detecting an event based on the movement data;
determining a non-fall event probability distribution based on a combination of the one or more parameters and the corresponding values, the non-fall event probability distribution indicating a probability of a non-fall event;
obtaining a threshold value based on the non-fall event probability distribution, the threshold value distinguishing between the non-fall event and a fall event; and
configuring an automatic fall detector with the threshold value, the automatic fall detector configured to use the threshold value to distinguish between the fall event and the non-fall event for subsequently obtained movement data.
2. The method of claim 1 , wherein the operations further comprise:
for each respective parameter of the one or more parameters, determining a corresponding parameter specific probability distribution,
wherein determining the non-fall event probability distribution is further based on a product of each corresponding parameter specific probability distribution.
3. The method of claim 1 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises determining a difference between the non-fall event probability distribution and a predicted fall event probability distribution.
4. The method of claim 1 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises determining an inverse or a complement of the non-fall event probability distribution.
5. The method of claim 1 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises dividing a predetermined estimation of a fall event probability distribution by the non-fall event probability distribution.
6. The method of claim 1 , wherein the operations further comprise determining, based on a predetermined estimation of a fall event probability, the combination of the one or more parameters.
7. The method of claim 6 , wherein determining the combination of the one or more parameters comprises determining that the predetermined estimation of the fall event probability satisfies a threshold value.
8. The method of claim 6 , wherein determining the combination of the one or more parameters comprises identifying a group of the one or more parameters having the greatest corresponding values.
9. The method of claim 1 , wherein obtaining the threshold value comprises processing the non-fall event probability distribution and a fall event probability distribution.
10. The method of claim 1 , wherein obtaining the threshold value is further based on a false alarm rate representing a rate of incorrectly classified non-fall events.
11. A system comprising:
data processing hardware; and
memory hardware in communication with the data processing hardware, the memory hardware storing instructions that when executed on the data processing hardware cause the data processing hardware to perform operations comprising:
obtaining movement data representing movement of a subject over a predetermined data collection period, the movement data comprising one or more parameters, each respective parameter comprising a corresponding value;
detecting an event based on the movement data;
determining a non-fall event probability distribution based on a combination of the one or more parameters and the corresponding values, the non-fall event probability distribution indicating a probability of a non-fall event;
obtaining a threshold value based on the non-fall event probability distribution, the threshold value distinguishing between the non-fall event and a fall event; and
configuring an automatic fall detector with the threshold value, the automatic fall detector configured to use the threshold value to distinguish between the fall event and the non-fall event for subsequently obtained movement data.
12. The system of claim 11 , wherein the operations further comprise:
for each respective parameter of the one or more parameters, determining a corresponding parameter specific probability distribution,
wherein determining the non-fall event probability distribution is further based on a product of each corresponding parameter specific probability distribution.
13. The system of claim 11 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises determining a difference between the non-fall event probability distribution and a predicted fall event probability distribution.
14. The system of claim 11 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises determining an inverse or a complement of the non-fall event probability distribution.
15. The system of claim 11 , wherein obtaining the threshold value based on the non-fall event probability distribution comprises dividing a predetermined estimation of a fall event probability distribution by the non-fall event probability distribution.
16. The system of claim 11 , wherein the operations further comprise determining, based on a predetermined estimation of a fall event probability, the combination of the one or more parameters.
17. The system of claim 16 , wherein determining the combination of the one or more parameters comprises determining that the predetermined estimation of the fall event probability satisfies a threshold value.
18. The system of claim 16 , wherein determining the combination of the one or more parameters comprises identifying a group of the one or more parameters having the greatest corresponding values.
19. The system of claim 11 , wherein obtaining the threshold value comprises processing the non-fall event probability distribution and a fall event probability distribution.
20. The system of claim 11 , wherein obtaining the threshold value is further based on a false alarm rate representing a rate of incorrectly classified non-fall events.