In recent years, the emerging framework of the Internet of Things has been leading the technological landscape in a number of different fields and applications, from autonomous and connected vehicles to wearable devices. The healthcare system is benefiting from this continuously evolving environment since it leverages the opportunities offered by the ubiquitous and pervasive presence of connected objects and smart services. This attitude has given rise to the concept of eHealth, thus enabling new approaches and solutions for healthcare. In this framework we propose SmartPants, an IoT-based wireless system specifically designed for the remote rehabilitation of lower limbs in poststroke patients. The platform consists of multiple nodes used to monitor physical therapy and a software platform that provides real-time feedback on the execution by recognizing the type of exercise currently being performed by the patient. Our experimental results, evaluated through appropriate metrics, show that the proposed movement recognition algorithm provides very good results in terms of classification performance, independent of the considered classifier, with an average true positive rate of about 91 percent and an overall accuracy of around 96.5 percent.

中文翻译：

---

当eHealth与IoT相遇时：用于中风后家庭康复的智能无线系统

近年来，新兴的物联网框架在从自动驾驶和联网汽车到可穿戴设备的许多不同领域和应用中一直引领着技术格局。医疗保健系统受益于这种不断发展的环境，因为它利用了连接对象和智能服务无处不在的普遍存在所提供的机会。这种态度引起了eHealth的概念，从而为医疗保健提供了新的方法和解决方案。在此框架中，我们提出SmartPant，这是一种基于IoT的无线系统，专门用于中风后患者下肢的远程康复。该平台由用于监视物理治疗的多个节点和一个软件平台组成，该软件平台通过识别患者当前正在执行的运动类型来提供有关执行情况的实时反馈。我们的实验结果通过适当的指标进行了评估，结果表明，所提出的运动识别算法在分类性能方面提供了非常好的结果，而与所考虑的分类器无关，平均真实阳性率约为91％，总体准确度约为96.5％ 。