Wireless signal technology performs a key role in the research area of medical science to detect diseases that are associated with the human gesture. Recently, wireless channel information (WCI) has received vast consideration because of its potential practice of detecting the human behavior. In this article, we present the convolutional neural network (CNN) model to classify WCI-based image data and determine the involuntary movement (tic disorder) diseases. Motor and vocal are two aspects of tic disorder and depend on the amount of complication, both aspects classified into the simple and complex group, and each group has several symptoms. Using WCI data of symptoms from the simple and complex group of motor aspects, we form a dataset to train the CNN model. Experimental results show that CNN provides satisfying result in classification, and accuracy is more than 97%.

中文翻译：

---

一种利用无线信道信息检测抽动障碍的深度学习方法

无线信号技术在医学研究领域发挥着关键作用，可检测与人类手势相关的疾病。最近，无线信道信息 (WCI) 因其检测人类行为的潜在实践而受到广泛关注。在本文中，我们提出了卷积神经网络 (CNN) 模型来对基于 WCI 的图像数据进行分类并确定不自主运动（抽动障碍）疾病。运动和声音是抽动障碍的两个方面，取决于并发症的数量，这两个方面分为简单组和复杂组，每组都有几个症状。使用来自简单和复杂运动方面的症状的 WCI 数据，我们形成一个数据集来训练 CNN 模型。实验结果表明，CNN 在分类方面提供了令人满意的结果，