Font Size: a A A

Design Of Finger Rehabilitation Training System In Mixed Reality

Posted on:2022-10-07Degree:MasterType:Thesis
Country:ChinaCandidate:X LiFull Text:PDF
GTID:2492306314973179Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
China has entered an aging society since the beginning of the 21st century.The contradiction between the increasing rehabilitation needs of the elderly and the lack of professional rehabilitation physicians and nursing staff has become increasingly prominent.In recent years,with the continuous progress and development of computer technology,Microsoft has launched a new generation of mixed reality product HoloLens 2,and the human-computer interaction experience under mixed reality has been further improved.With the promotion of 5G technology,the hologram of mixed reality under low latency and high bandwidth will have a more eye-catching performance,and the future application prospects are very broad.Traditional rehabilitation training is mainly based on one-on-one guidance by professional rehabilitation doctors,supplemented by robotic arm training,which is single and boring,with low training efficiency and high cost,making it difficult to promote on a large scale.The rehabilitation training system proposed in this paper is a system developed by HoloLens 2 with the game as the carrier.In order to better realize personalized rehabilitation training,the system will record patients’ training data during the game,and provide patients with various forms of feedback including visual and auditory feedback,which can objectively and reasonably evaluate patients’current training status.The system is designed with an introductory guide to facilitate the patients to be familiar with HoloLens 2’s interactive mode and understand the rules of the game.The system also designs trajectory routes with different difficulty levels,and patients will perform standard tests and match the game difficulty level for the first time.During the game,the system will collect 3D data of the patient’s current motion trajectory,and use LSTM to predict the patient’s motion trajectory.The predicted result is used for analysis of reinforcement learning and provides real-time operation guidance for the patient.At the same time,during the game,we will objectively and reasonably evaluate the completion of the patient’s game task based on the difference between the patient’s real motion trajectory and the system’s displayed trajectory,and then dynamically adjust the difficulty of the game,which also provides an important basis for formulating the next stage of rehabilitation training plan.
Keywords/Search Tags:mixed reality, rehabilitation training, HoloLens 2, LSTM, reinforcement learning
PDF Full Text Request
Related items