Detecting Medical Emergencies in Isolated Older Adults Living Alone in Rural Areas

摘要 独居农村的孤寡老人有可能在得不到帮助的情况下陷入医疗困境。一个农村 房子可以与邻居隔离，邻居可以很容易地检查他们健康。当这些人成为 老年人，他们可以有很高的偏好留在他们的家，只要他们认为他们可以照顾 自己一个孤苦的老人在偏远的农村， 回家在一个孤立的农村房子，许多天可以通过之前，有人决定开车到他们的位置， 看看老人家。该项目为孤立的农村老年人开发了一种低成本的监测解决方案， 安全地过独立的生活第一阶段SBIR项目将开发一个无线室内跟踪系统， 通过典型的农村房屋的几面墙来探测一个人的位置。将没有电缆运行或需要 在房子里安装一系列复杂的传感器安装成本低廉。系统将 有电池备份和协议，以在长时间停电时运行。它也将有一个手机备份 通讯选项，如果电话线被关闭。系统将识别老年人24的运动 每天在房子里呆上几个小时。机器学习（ML）算法可以从日常生活中检测出异常。 该产品的目的是作为一个可选的配件，在家庭报警系统在使用的今天， 家庭警报系统的多个供应商。它将与可穿戴按钮并行操作，以发出警报信号。 一旦系统识别出发生了遇险事件，它将以与 可穿戴按钮按压。将遵循相同的警报协议。在这些系统中，操作员将首先尝试 用供应商的家用警报系统的扬声器电话与人交谈。如果他们无法沟通 与人，他们开始通过当地人的电话名单工作，以检查家庭。第一阶段将 开发和测试无线室内跟踪系统和机器学习算法。

Abstract Isolated older persons living alone in a rural house are at risk of being in medical distress without help. A rural house can be isolated from neighbors who can easily check on their well-being. As these people become elderly, they can have a high preference to stay in their home as long they believe they can care for themselves. An elderly person alone and in distress can be in a life and death situation in an isolated rural home. In an isolated rural house, many days can pass before someone decides to drive to their location to check on the elderly person. This project develops a low-cost monitoring solution for isolated rural elderly to safely lead independent lives. The phase I SBIR project will develop a wireless indoor tracking system to detect a person's location through several walls of a typical rural house. There will be no cables to run or need to install a complex array of sensors through the house. The cost of installation inexpensive. The system will have battery backup and protocols to operate over long power outages. It will also have a cell phone backup communication option should phone lines be down. The system will identify motions of the elderly person 24 hours a day in the house. Machine Learning (ML) Algorithms can detect abnormalities from their daily routine. The product is intended to be an optional accessory to in-home alert systems in use today and work with multiple vendors of in-home alert systems. It will operate in parallel with wearable buttons to signal an alert. Once the system identifies a distress event has occurred it will activate the alert system the same way as a wearable button press. The same alert protocol would be followed. In these systems an operator would first try to talk to the person with a speaker phone of the vendors' in-home alert system. If they cannot communicate with the person, they start working through a call list of local people to check on the home. The phase I will develop and test the wireless indoor tracking system and the Machine Learning (ML) Algorithms.