探讨头戴式虚拟现实（VR）设备在全膝关节置换术后康复的应用，调查受试者对佩戴头戴式VR设备的感受并寻求改进建议。

采取前瞻性横断面研究的方法，选取2022年6至10月在北京大学第三医院因全膝关节置换术后功能障碍住院康复的50例患者作为研究对象。其中男16例，女34例，年龄（66.86±8.96）岁。采用定量研究方法，术后分别对患者进行人工评定和头戴式VR设备评定，测量其膝关节活动度（ROM）、直抬腿训练持续时间、运动时和静息时的疼痛，分析比较人工评定和VR评定的差异。采用质性研究方法，对11例受试者进行半结构化访谈，使用主题框架法对访谈结果进行分析。

定量研究中，人工测量与使用头戴式VR设备测量的膝关节最大屈曲ROM、最大伸直ROM、运动状态的数字评定量表（NRS）评分、静息状态NRS、直抬腿训练持续时间等指标差异均无统计学意义（均P＞0.05）。两种测量方法得到的膝关节最大屈曲ROM［组内相关系数ICC=0.839，95%置信区间（95%CI）：0.732~0.905，P<0.001］、最大伸直ROM（ICC=0.970，95%CI：0.946~0.983，P＜0.001）、运动状态NRS（ICC=0.944，95%CI：0.901~0.969，P＜0.001）、静息状态NRS（ICC=0.968，95%CI：0.942~0.982，P＜0.001）、直抬腿训练持续时间（ICC=0.968，95%CI：0.938~0.984，P＜0.001）指标结果具有高度一致性。将质性研究得到的半结构化访谈内容进行分析得出3个主题：感知易用性、感知有用性和改进建议，其中感知易用性包括设备缺陷、可操作性、学习难度和接受程度；感知有用性包括影响适配体验的因素和设备功能；改进建议包括多功能化、标注目标角度、丰富虚拟背景环境和训练项目。

头戴式VR设备可作为全膝关节置换术后康复的辅助工具，在评定膝关节ROM、NRS以及直抬腿训练持续时间上VR设备与康复专业人员评定无明显差别；相较于其他姿态检测设备有着便于携带和实时反馈的特点。

To explore the use of virtual reality (VR) headsets in rehabilitation after total knee arthroplasty (TKA), and to investigate participants' feelings about wearing VR headsets and seek their suggestions for improvement.

A prospective cross-sectional study was conducted on 50 patients who were hospitalized and recovered from dysfunction after TKA at Peking University Third Hospital from June to October 2022, including 16 males and 34 females with a mean age of (66.86±8.96) years. Using quantitative research methods, the patients were evaluated manually or with head-mounted VR devices after surgery. The range of motion (ROM), straight leg raise exercise duration, and pain during exercise and rest were measured, and the differences between manual assessment and VR assessment were analyzed. Additionally, semi-structured interviews were conducted with 11 selected participants, and the interview results were analyzed by thematic framework analysis.

In the quantitative study, there were no significant differences between the manual measurement and and VR assessment with regard to the maximum knee flexion ROM, maximum extension ROM, numerical rating scale (NRS) score, resting state NRS, and straight leg raise exercise duration (P＞0.05 for all). The intraclass correlation coefficients (ICCs) for the maximum flexion ROM of the knee (ICC=0.839, 95% confidence interval (CI): 0.732-0.905, P＜0.001), maximum extension ROM of the knee (ICC=0.970, 95% CI: 0.946-0.983, P＜0.001), movement state NRS (ICC=0.944, 95% CI: 0.901-0.969, P＜0.001), resting state NRS (ICC=0.968, 95% CI: 0.942-0.982, P＜0.001), and straight leg raise exercise duration (ICC=0.968, 95% CI: 0.938-0.984, P＜0.001) showed high consistency between the two methods. The analysis of the semi-structured interview content from the qualitative study revealed three themes: perceived ease of use, including device defects, learning difficulty, and acceptance; perceived usefulness, including operability, factors that affect the fit experience, and device functions; and improvement suggestions, including multifunctionality, labeling the target angle, enriching virtual background environments, and training programs.

VR headsets can serve as an auxiliary rehabilitation therapy tool for post-TKA recovery. In rehabilitation assessments, there is no significant difference between VR devices and rehabilitation professionals in terms of knee joint ROM, NRS, and straight leg raise exercise duration. Compared with other attitude detection devices, head-mounted VR devices have the characteristics of easy portability and real-time feedback.