Wnlab

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Android-based Fall Detection ======
Algorithm is designed as mobile application and operates on smartphones with Android OS (recommended version 2.3 and higher).\\
Make sure to complete all the steps specified in [[giraffplus:android-middleware|"GiraffPlus for android"]] to establish communication between your device and middleware.\\
Alternatively, you can choose to send alarms directly through Pushover app. This option is described in "Pushover alarm notifications" section below.
===== Application package ===
Current version of fall detection app:\\
[[https://www.dropbox.com/s/lssx1nlnkwjz704/AndroidBluetoothSPPedit.apk]] (for Android OS 2.3-4.0, last update 20-May-2014 15.30)\\
[[https://www.dropbox.com/s/6uoida1bq7vzfau/FallDetectorGiraffPlus_edit.apk]] (for Android OS 4.0-4.3, last update 18-July-2014 15.45)\\ \\
Simple guide on how to install android .apk file can be found [[http://www.talkandroid.com/guides/beginner/install-apk-files-on-android/|here]].
===== User interface ===
After installing the application, mobile device can now serve as wearable sensor adjusted to patient's body. Secondary users or caregivers can enter main menu and choose between two alternatives to manage fall alarm notifications.
==== DVPIS alarm notification ===
The first option will establish connection with GiraffPlus middleware and trigger an alarm in case of falling. This alarm is saved in database and visualized in DVPIS under "Physiological data" section. \\ \\
{{:giraffplus:falldetectiondvpis.png?190|}} {{:giraffplus:dvpis.png?600|}} \\ \\
However, no extra notification will be sent to a secondary user or caregivers and therefore, it is required to update DVPIS consistently to "catch" the alarm.
==== Pushover alarm notifications===
Alternatively Pushover application is deployed for managing alarms from the fall detection algorithm. This option allows secondary users or caregivers to receive triggered alarms directly on their mobile device including additional information such as primary user's current location. You can find detailed information on Pushover API [[https://pushover.net/api| here]] or ask a question directly on Basecamp. After installing Pushover application on secondary user's mobile device and registration, you obtain personal user key. Copy, paste and save this key into a following section of the user interface before starting fall detection algorithm.\\ \\
{{:giraffplus:screenshot_2014-05-20-15-32-47.png?190|}} {{:giraffplus:falldetectionpushover.png?190|}}  \\ \\
Every fall alarm detected by wearable sensor will be sent out to secondary user's mobile devices with the corresponding user key. Make sure to register your Pushover application with the same user key. An example of fall alarm generated by Pushover will look as following:\\ \\
{{:giraffplus:pushoverapp.png?225|}} \\ \\
===== Short algorithm description ===
Click the start button to initiate fall detection algorithm after alarm configuration is complete. Our method starts to receive data from accelerometer inbuilt into mobile device and calculate the overall acceleration value (empirically developed activity measure) //Act//. Collected measurements correspond to acceleration along 3 axis of the smartphone coordinate system.\\ \\
{{:giraffplus:acceleration_axes.jpg?300|}}\\ \\
If the impact is registered (//Act// ≥ threshold) during activity of daily living, we consider it a potential fall risk, pause monitoring process and check orientation of the phone. Using the same acceleration values, we determine devices’ Euler angles with respect to the earth’s gravitational attraction. A combination of thresholds for roll and pitch angles corresponding to a horizontal position of the body triggers an alarm signal informing the user about the fall. Before forwarding this message to a smart home database algorithm complements an alarm with additional data like patient's location or direction of the fall (optional).




===== Sensor set up ===
In order to provide reliable fall detection, mobile device should be adjusted to patient's body in the following horizontal position.\\ \\
{{:giraffplus:sensorposition.jpg?220|}}

===== Other comments ===
It is recommended to activate fall detector in the beginning of the day and stop monitoring before sleep or other inappropriate activity (e.g. taking a shower)\\
The primary user is able to leave the apartment without loosing connection to the middleware or Pushover functionality. In thi case, it is recommended to deploy mobile wireless network instead of wireless connection at home.