- 學位論文
基於深度影像步態分析實現感官刺激導引復健助行系統
Sensory Cues Guided Rehabilitation Robotic Walker Realized by Depth Image based Gait Analysis
摘要
本論文提出一感官刺激為基礎之引導助行系統以協助改善帕金森氏症患者之行走步態。一般常見的帕金森氏症患者之異常步態可分為凍結步態與小碎步。基於許多研究證實視覺與聽覺刺激對於帕金森氏症患者具有相當顯著的復健效果，故於本論文中引用影像與聲音之感官刺激結合輔助行走機器人提供帕金森氏症患者復健以達到改善步態的目標。本系統透過建置於助行機器人內部之深度影像攝影機實現一非侵入式、即時三維下肢影像追蹤實現行走步態分析功能。本系統中提出一適應性步態之感官刺激引導復健功能，經由患者之下肢影像追蹤與步態分析之結果提供符合患者穩定步態之視覺影像刺激，並經由內建於助行機器人中的投影機投射視覺刺激影像於地面與具有符合患者行走頻率之提示聲。再者，我們提出可調整式視覺刺激提示功能以達到同時改善患者步態且減少患者於復健時持續跨步超出下肢負荷所造成的負擔。本系統測試實驗中，我們透過動作捕捉技術平台驗證下肢影像追蹤之準確性。此外，我們邀請七位受試者包含四位帕金森氏症患者以及三位健康年長者進行三天的復健系統測試，實驗結果顯示患者使用本系統復健並獲得顯著的步態改善成果。
並列摘要
In this thesis, we propose a sensory cues guided robotic walker for improving gaits of Parkinson Disease (PD) patients. A completely non-intrusive, real time 3D leg pose tracking and gait analysis are proposed by using depth camera which mounted on the rear of robotic walker. It has been studied that the sensory cues can serve as effective stimuli to the PD patients for gait improvement. In our work, the sensory cues include visual and auditory cue are incorporated into robotic walker. The adaptive gait sensory cues guider provides the visual cue projected on ground by projector installed on walker and rhythmic audio cue to stimulate patients’ walking gait. An adaptive gait rehabilitation mechanism is proposed to offer an appropriate visual and auditory cue based on walking gait. The adjustable visual cue is proposed to improve their gait and reduce their uncomfortableness simultaneously. In experiment, the accuracy rate of proposed 3D leg pose tracking was evaluated by a motion capture system. In addition, seven subjects (4 PD patients and 3 healthy elders) were invited to test the system for three days. The result shows subjests’ gait performance has substantially improved by using our system.