Class leading performance.
Automatic + Continuous
& Prevention Matters.
falls (USA) in hospitals each year, of which 30% to 51% results in injury.
Direct medical costs due to falls in hospitals.
Falls are the leading cause of injuries for older Americans which result in financial penalties for hospitals and dissatisfaction for the nursing staff.
Continuous Detection Matters.
Self-correct miss detections and false triggers.
Contact-based systems such as fall detection pads or wearable devices are only designed to detect a possible event “one-time”. If it misses a fall motion, it will not be able to self-correct. With continuous detection, it can self-correct for any type of miss detection or even false triggers by continuously scanning and monitoring.
Therefore, even with 98% miss-detection, the “2%” chance that Xandar Kardian’s systems may miss can be self-corrected and detect a fall by scanning again, again and again. Similarly, the 6% false trigger event can also self-correct and cancel out alerts automatically.
DETECTS BOTH FAST / HARD AND SLOW / SLIDING FALLS
Fall Detection Logic (3 step process)
Falls can happen anywhere, and it can happen in various ways. Wearable devices, including neck pendants are based on 3 axis sensors which includes accelerometers, gyroscope and sometimes even a barometer sensor to get atmospheric pressure. All of these systems are designed to detect for “fast / rapid” type of falls. Think of slipping on a banana peel!
We believe that this type of logic completely misses slow / sliding type falls if the patient is able to grab onto something before falling to the ground. This is why our logic continuously scans for ToF (time of flight) motion and confirms the fall by locating the heartbeat and breathing signals emitted from the body from ground level of the given space.
A patient may be classified as “high fall risk”, which automatically sets the system to look for early bed exit signs. It includes looking for increase of large motion from the bed followed up with a sit-up. Unlike other competitors, Xandar Kardian is also able to look for “leg falling to the side of the bed”, which we call “Sit-up + Fidgeting Detection”. This critical information tells the nurses that the patient didn’t sit-up to grab a TV remote control, smartphone or water, but instead, have full intention of getting out of the bed. The system continues to scan and monitor their motion and alerts the nurse when they exit the bed, but also monitor their motion – including possible falls as they make their way to the bathroom.
* Survey was conducted in collaboration with Philips during Philips Healthworks program from October to December 2019.
Fall Prevention Logic:
Designed by Nurses.
28 nursing staff from 28 different hospitals were surveyed. *
28/28 Answered - In-hospital Fall is one of top 3 in-hospital patient safety concern.
23/28 Confirmed that “bed exit” alert was extremely important. Bathroom + Bed represented 89% of falls.
18/28 Stated that “stress” and “financial burden” affected the staff.
18/28 Voted that “2 minute” alert time speed for bed exit was most ideal.
Further research revealed that 45.2% of falls in hospitals were toileting related. ** In fact, this co-relates to research that shows 41.3%*** of falls occurred from 11pm to 7am, supporting the assumption that most falls occur in the middle of the night while trying to “walk over” to the bathroom without calling for nurse’s assistance.
The feedback from nurses and historical data formed the foundation of Xandar Kardian’s fall prevention system logic.
Improve staff experience
Reduce admin burden. No staff or patient action / intervention required. General ward + bathroom monitoring.
Improve patient experience
Continuous fall detection promotes better patient experience, quality of life.
Lower cost of care
Fall prevention reduces financial burden/penalties and patient read missions.
Compliant Cloud Monitoring.
For long term care patients and residents, some private health data may be linked to a patient which requires certain privacy protection protocols. Xandar Kardian has partnered with MEDSTACK.CO for its HIPAA compliance requirements. By hosting all the data with Medstack, it will be able to support stringent security expectations and integrate patient EHR data.
Cloud computing can also mean that every alert is sent, monitored, stored and pushed out in real-time. Historical data can also be pulled up rapidly from any secure device, making review of events simple and easy.
Detection Range Explained.
By regulation, ir-uwb radar signals can only go to a maximum of 10 meters (33ft). Fall detection system relies of picking up correct vital signs being picked up from the ground level of a given space. This is done, again, to ensure that a human being has fallen on the floor, no matter how they have fallen. The downside to this method is that the further away the person falls from the radar device, the signal may deteriorate, which also decreases the performance of fall accuracy and false trigger rates.
Please keep in mind however, that Xandar Kardian solutions are designed to continuously monitor, self-correcting potential miss detections or false triggers as time goes by.
* % ratio based on first detection scan only. Miss detections & false triggers can self-correct during further continuous / repetitive scanning.
Fall prevention requires just in time reaction
Real-time push notification.
Whether it is fall prevention or detection, quick reaction requires real-time information about what is happening in a particular room. Although Xandar Kardian’s systems have SDK/API for digital nurse call systems (upon compatibility), we believe that nurses can better react with push alerts on their smartphone or wearable device. As many hospitals move to BYOD, Xandar Kardian has made it available for Android or iOS “nurse apps” to be used securely for fall management purposes.
Right decisions at the right time.
A recent survey*, asked nurses “what are the current fall alarm issues faced by nursing/care giver staff?”. Overwhelming majority answered that there is not enough staff and that the systems did not provide early enough notification (bed exit). One survey feedback stated “95% of falls occur while staff is assisting other patients.”
With Xandar Kardian, nurses can make the right decision at the right time by quickly glancing over the current situation. Information such as “sit-up warning since 0:36” vs. “bed exit since 0:06” can help nurses to react to the room with bed exit rather than a sit-up. Even with the same sit-up detection situation, the nurse can decide to enter the room of a patient that have waited longer as they are more likely to get out of the bed earlier.
* Survey was conducted in collaboration with Philips during Philips Healthworks program from October to December 2019
Related Patents & Journal Publications
Wall Mounted Fall Detection using Radar
IP Number: 10-2100639
Granted Date: 2020-04-08
Fall Detection using Multiple Radars
IP Number: 10-2038081
Granted Date: 2019-10-23
International: USA (16/601,789), Europe (19199841.8)