<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/xsl" href="https://devzone.nordicsemi.com/cfs-file/__key/system/syndication/rss.xsl" media="screen"?><rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:slash="http://purl.org/rss/1.0/modules/slash/" xmlns:wfw="http://wellformedweb.org/CommentAPI/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Thingy:52 As fall detection Device</title><link>https://devzone.nordicsemi.com/f/nordic-q-a/73485/thingy-52-as-fall-detection-device</link><description>Hello Everyone , 
 
 With variety of motion sensors on Thingy:52 , can it be used to detect if a person has fallen ? If &amp;#39;yes&amp;#39; then how accurate the results will be ? 
 If someone has tried it then please share your experience. 
 
 Thanks</description><dc:language>en-US</dc:language><generator>Telligent Community 13</generator><lastBuildDate>Wed, 07 Apr 2021 11:32:56 GMT</lastBuildDate><atom:link rel="self" type="application/rss+xml" href="https://devzone.nordicsemi.com/f/nordic-q-a/73485/thingy-52-as-fall-detection-device" /><item><title>RE: Thingy:52 As fall detection Device</title><link>https://devzone.nordicsemi.com/thread/303459?ContentTypeID=1</link><pubDate>Wed, 07 Apr 2021 11:32:56 GMT</pubDate><guid isPermaLink="false">137ad170-7792-4731-bb38-c0d22fbe4515:8fad3528-0391-4038-a48b-9d6500ac4f6f</guid><dc:creator>JONATHAN LL</dc:creator><description>&lt;p&gt;Hi,&lt;br /&gt;&lt;br /&gt;&lt;br /&gt;We don&amp;#39;t have any example demo that you can work on but there is some options to explore. To best detect a fall using the sensor date from the Thingy52 you could maybe use something like &lt;a href="https://docs.edgeimpulse.com/docs/nordic-semi-nrf52840-dk"&gt;this example&lt;/a&gt; form Edge impulse on the device.&amp;nbsp;&lt;br /&gt;&lt;br /&gt;it can in theory be very accurate and might even be able to detect different types of fall, but that would require a huge set of sample data and training a couple models. Exactly how and what is the best solution is hard to say. Maybe a simple accelerometer check is enough.&lt;br /&gt;Regards,&lt;br /&gt;Jonathan&lt;/p&gt;&lt;div style="clear:both;"&gt;&lt;/div&gt;</description></item></channel></rss>