Visual analysis system to detect and predict the signs of anxiety in healthcare

这个项目的目的是在牙科实践中识别焦虑的非语言线索。本项目以牙科诊所为例，说明紧张的健康环境。这个项目的目的是通过视频分析牙科病人的行动和行为，以预测和检测他们的焦虑程度。该项目的框架是基于深度学习方法来分析牙科患者的运动和身体动作。该流程是基于视频处理技术的动作识别。本研究的主线可以概括如下：1)更新当前关于行为表达和动作预测的视觉分析系统的文献综述；2)确定适当的机器视觉方法来检测和预测有待扩展的焦虑；3)改进框架，使其在实时和现实场景中有效地执行。4)实施开发的系统，并在牙科实践模拟中进行实验。

The intention of this project is to identify non-verbal cues of anxiety in dental practice. This project uses the dental practice as an example of a stressful health environment. The aim of this project is to analyse the action and behaviour of dental patients through videos to predict and detect their level of anxiety. The framework of this project is based on a deep learning approach to analyse the movement and body actions of dental patients. This flow is based on video processing techniques for action recognition. The main lines of this research study can be summarised as follows: 1) Updating the current literature review on visual analysis systems on behaviour expression and action prediction 2) Identify the appropriate machine vision approach to detect and predict anxiety to be expanded 3) Improve the framework to perform efficiently in real time and in real-world scenarios. 4) Implement the developed system and perform experiments in dental practice simulations.