Alert system
A system is provided which, in at least some embodiments, can read the vital signs of the body of a user utilizing a sensing device such as a smartwatch or smart phone (for example utilizing the IOS, Android or Pebble operating systems) and apply algorithms to interpret the vital signs and then send a notification with an escalation process to nominated carriers if the patient is interpreted as having a fall or fit or seizure. In at least some embodiments doctors or other parties can log in to a secured dashboard and check a patient data in real time. Doctors or other parties can analyze the history of the patient. In at least some embodiments, users/patients can also use data to keep track of fall or fit or seizure episodes and monitor their progress.
1. A fall detection apparatus adapted to be worn by a user, the apparatus comprising:
a processor configured to:
receive data input comprising an acceleration signal and a time reference signal and determine a fall event from the data input;
an accelerometer configured for substantially continuous recording of the acceleration signal and communication of the acceleration signal to the processor, the acceleration signal quantifying acceleration of the fall detection apparatus relative to a reference frame; and
a timer configured for substantially continuous communication of the time reference signal to the processor;
wherein the processor is configured to implement a threshold-based algorithm comprising:
substantially continuously monitor the acceleration signal for detection of a low acceleration signal (LAS) within a low acceleration range (LAR), using the time reference signal, to compare a duration of the LAS to a predetermined first period of time, t 1 ;
substantially continuously monitor the acceleration signal for detection of a high acceleration signal (HAS) in a high acceleration range (HAR), subsequent to t 1 , and using the time reference signal, to compare a duration of the HAS to a predetermined second period of time, t 2 ;
wherein on detection of the HAS having a duration longer than the predetermined second period of time, t 2 , the processor is configured to generate a fall detection signal indicative of a fall by said user onto a surface;
wherein, on generation of the fall detection signal, the processor is configured to:
monitor the time reference signal and the acceleration signal during a third predetermined period of time, t 3 , subsequent to t 2 ; and
generate a user immobile signal based on the acceleration signal remaining in a predetermined very low acceleration range (VLAR) during the third predetermined period of time, t 3 ,
wherein the processor is configured to generate the immobile signal by processes including:
initiate a Time To Detect On The Floor (TTDOTF) time setting which corresponds to t 3 ,
analyze the acceleration signal for all time during t 3 ;
measure at least one Time On The Floor (TOTF) period defined as a time period that the acceleration signal is within an On The Ground acceleration Sensitivity (OTGS) setting during the TTDOTF time setting,
exclude time segments corresponding to the acceleration signal outside the OTGS during t 3 before and/or after the at least one TOTF period,
confirm a fall detection event in response to the user immobile signal by the processor being configured to determine that a sum of the at least one TOTF period is greater than a TOTF setting;
customize at least one of the LAR, the HAS and the predetermined VLAR for the user with reference to personal profile settings unique to said user.
2. The fall detection apparatus of claim 1 , further comprising a transmitter, and wherein when the fall detection event is confirmed by the processor a fall signal is transmitted by the transmitter to a remote location.
3. The fall detection apparatus of claim 2 , wherein transmitter has Bluetooth or other short-range radio or electromagnetic transmission capability.
4. The fall detection apparatus of claim 1 , wherein when the fall detection event is confirmed by the processor then the processor is further configured to communicate a fall signal locally.
5. The fall detection apparatus of claim 1 , wherein the acceleration signal is referenced against the reference frame.
6. The fall detection apparatus of claim 1 , wherein the reference frame is a surface upon which the user of the fall detection apparatus is supported.
7. The fall detection apparatus of claim 1 , wherein the fall detection apparatus is adapted to be a wrist-mounted fall detection apparatus.
8. The fall detection apparatus of claim 1 , wherein the processor is configured to use input from the threshold-based algorithm and use input of fall detection data from a trained neural network based machine learning algorithm to perform a weighting system function; the processor configured to adjust a weight in the weighting system function applied to the respective inputs over time based on user feedback including learned fall characteristics of the user over time thereby increasing reliability of fall detection decisions.
9. The fall detection apparatus of claim 1 , wherein said TTDOTF excludes time when the acceleration signal is greater than the OTGS setting.
10. A method of detecting a fall event of a user with a fall detection apparatus comprising:
processing via a processor data input comprising an acceleration signal and a time reference signal and determining a fall event from the data input;
substantially continuously recording the acceleration signal and communication of the acceleration signal to the processor, the acceleration signal quantifying acceleration of the fall detection apparatus relative to a reference frame;
substantially continuously communicating the time reference signal to the processor;
using the processor configured to implement a threshold-based algorithm comprising:
substantially continuously monitoring the acceleration signal for detection of a low acceleration signal (LAS) within a low acceleration range (LAR), using the time reference signal, comparing a duration of the LAS to a predetermined first period of time, t 1 ;
substantially continuously monitoring the acceleration signal for detection of a high acceleration signal (HAS) in a high acceleration range (HAR), subsequent to t 1 , and using the time reference signal, to compare a duration of the HAS to a predetermined second period of time, t 2 ;
wherein on detection of the HAS having a duration longer than the predetermined second period of time, t 2 , the processor is configured to generate a fall detection signal indicative of a suspected fall by said user onto a surface;
wherein, on generation of the fall detection signal, using the processor for monitoring the time reference signal and the acceleration signal during a third predetermined period of time, t 3 , subsequent to t 3 ; and
generating a user immobile signal based on the acceleration signal remaining in a predetermined very low acceleration range (VLAR) during the third predetermined period of time, t 3 ,
wherein the processor is configured to generate the immobile signal by processes including:
initiating a Time To Detect On The Floor (TTDOTF) time setting which corresponds to t 3 ,
analyzing the acceleration signal for all time during t 3 ;
measuring at least one Time On The Floor (TOTF) period defined as a time period that the acceleration signal is within an On The Ground acceleration Sensitivity (OTGS) setting during the TTDOTF time setting,
excluding time segments corresponding to the acceleration signal outside the OTGS during t 3 before and/or after the at least one TOTF period,
confirming a fall detection event in response to the user immobile signal by the processor being configured to determine that a sum of the at least one TOTF period is greater than a TOTF setting; and
customizing at least one of the LAR, the HAS and the predetermined VLAR the user with reference to personal profile settings unique to said user.
11. The method of claim 10 , further comprising using the processor to use input from the threshold-based algorithm and use input of fall detection data from a trained neural network based machine learning algorithm to perform a weighting system function, and adjusting a weight in the weighting system function applied to the respective inputs over time based on user feedback including learned fall characteristics of the user over time thereby to increase reliability of fall detection decisions.