Skip to main content
  • Study Protocol
  • Open access
  • Published:

Effectiveness of a digital rehabilitation program based on computer vision and augmented reality for isolated meniscus injury: protocol for a prospective randomized controlled trial



The lack of access to physical therapists in developing countries and rural areas poses a significant challenge in supervising postsurgical rehabilitation, potentially impeding desirable outcomes following surgical interventions. For this reason, this study aims to evaluate the feasibility, safety, and effectiveness of utilizing a digital rehabilitation program based on computer vision and augmented reality in comparison with traditional care for patients who will undergo isolated meniscus repair, since to date, there is no literature on this topic.


This study intends to enroll two groups of participants, each to be provided with informed consent before undergoing randomization into either the experimental or control group. The experimental group will undergo a digital rehabilitation program utilizing computer vision and augmented reality (AR) technology following their surgical procedure, while the control group will receive conventional care, involving in-clinic physical therapy sessions weekly. Both groups will adhere to a standardized rehabilitation protocol over a six-month duration. Follow-up assessments will be conducted at various intervals, including preoperatively, and at 2 weeks, 6 weeks, 12 weeks, and 24 weeks postoperatively. Imaging assessments and return-to-play evaluations will be conducted during the final follow-up. Clinical functionality will be assessed based on improvements in International Knee Documentation Committee (IKDC) and Visual Analog Scale (VAS) scores.

Registration number



The lateral and medial menisci protect articular cartilage by providing shock absorption, distributing load, and lubricating the articular surface. Injury to the meniscus is prevalent among young and athletic individuals, causing biomechanical alterations of the joint, resulting in increased joint contact stress and possibly precipitating early degenerative changes and osteoarthritis [1, 2]. Studies have established that meniscectomy increases the likelihood of accelerated knee osteoarthritis; thus, arthroscopic procedures with suturing or meniscoplasty (central meniscal resection) are usually employed to preserve as much meniscal tissue as possible, depending on the shape of the damaged meniscus [1,2,3,4].

Postoperatively, physical therapy plays a pivotal role [5]. However, a significant challenge arises from the limited access to skilled therapists in developing countries and remote rural areas [6,7,8]. Patients in these resource-constrained regions often face long commutes to healthcare facilities, reduced compliance with rehabilitation protocols, and a lack of monitoring during the postoperative phase [6]. Consequently, recognizing the limitations inherent in conventional clinic-based physical therapy practice, digital rehabilitation programs will emerge as a promising supplemental solution. [7, 9]

Advanced technologies such as augmented reality and computer vision will have the potential to offer interactive digital therapy experiences [10]. Currently, the utilization of the VR system facilitates home-based rehabilitation training and increases its interactivity and enjoyment. [11,12,13,14,15] However, in VR, patients interact with a virtual environment that simulates real-life activities. The risk associated with this technology is that potentially dangerous situations may not be appropriately identified; however, images portrayed in both AR and virtual reality overlap with real-world images, thus allowing patients to be aware of potential dangers. [10] The incorporation of both virtual and real-world elements combined with real-time interaction and standard rehabilitation protocols can be leveraged to promote the recovery of an injured joint [15]. Telerehabilitation using technologies such as computer vision (CV) offers the potential for improving access to rehabilitation programs. The use of markerless human pose estimation based on computer vision in telerehabilitation is a promising research area, as it offers the advantage of closely monitoring movement without the need for external markers to capture motion data [16]. In our study, AR technology will utilize readily available devices such as smartphones, eliminating the need for additional hardware and providing greater convenience, while computer vision will facilitate real-time monitoring of patient movements and exercise postures.

The protocol of the digital rehabilitation program is based on the concept of accelerated rehabilitation after meniscal surgery proposed in 1996, suggesting that early postoperative weight bearing and knee range of motion (ROM) could reduce the risk of joint adhesions and muscle atrophy [5]. In recent years, accelerated rehabilitation programs for early weight bearing and active ROM after meniscus repair have shown positive results in patients with longitudinal meniscus tears [1, 2, 17]. Further research is warranted to investigate the safety and efficacy of early weight bearing and ROM training in patients who undergo complex meniscal tear repair, as the rehabilitation strategies may vary depending on tear pattern. [2, 18]

As a result, this study will utilize computer vision and smartphone-based augmented reality to overcome the limitations associated with conventional postoperative rehabilitation methods, with the goal of enhancing orthopedic postoperative outcomes. This study will involve delivering standardized exercise protocols using a digital platform, providing patients with continuous and effective physical therapy training supplemented by real-time feedback. Notably, there is currently no literature reporting the application of augmented reality and computer vision for postoperative recovery in patients with isolated meniscus injuries.

Aim of study

This study will recruit adult participants diagnosed with isolated longitudinal meniscus injuries confirmed through arthroscopic examination and scheduled for meniscus repair. The primary objective of this research is to assess the safety, effectiveness, and feasibility of implementing a digital rehabilitation program that incorporates computer vision and augmented reality as part of the postoperative rehabilitation process.


Study design

This is a single-center, prospective, randomized controlled study. After receiving informed consent, participants will be randomly allocated into two groups: the experimental group and the control group will adopt the standardized rehabilitation protocol (Table 1). The experimental group will engage in a digital rehabilitation program that can be completed from home, whereas the control group will attend weekly physical therapy clinic sessions for exercise guidance. The flowchart of the study is shown in Fig. 1.

Table 1 The standardized postoperative rehabilitation plan
Fig. 1
figure 1

Research flowchart


SPSS 23.0 (IBM, New York, NY, USA) will be used to generate random numbers. Subjects with odd numbers will be assigned to the experimental group, and subjects with even numbers will be assigned to the control group. The envelope method will be used for hidden grouping, which will be implemented by a third party. The random allocation scheme will be stored in an opaque envelope, which will be opened in succession according to the order of inclusion, thereby determining the assigned group for each patient. After that, the envelopes will be given to the study implementer, who will then use inclusion and exclusion criteria to decide whether each patient can be included in the study.

Blinding method

For this study, blinding of the interveners and patients was not feasible. Blinding should be applied to the data collectors, data analysts, and outcome assessors in the study.


Inclusion criteria:

  1. (1)

    Isolated meniscus injury diagnosed by magnetic resonance imaging (MRI)

  2. (2)

    Confirmation of a longitudinal tear pattern (including bucket handle tears) under arthroscopy, with a repairable tear.

  3. (3)

    Suture meniscus repair with or without partial meniscectomy

  4. (4)

    Participants have sustained their meniscus injury due to physical activities, such as playing basketball.

  5. (5)

    Willing to participate in this clinical trial and receive follow-up.

Exclusion criteria:

  1. (1)

    Preoperative MRI diagnosis with other ligament or chondral lesions;

  2. (2)

    A discoid meniscus was diagnosed under arthroscopy.

  3. (3)

    Trauma and surgery to other weight-bearing joints in the lower extremities, such as torn ligaments in the ankle and necrosis of the femoral head in the hip, can indirectly affect knee load and movement.

  4. (4)

    Knee osteoarthritis, defined by Kellgren–Lawrence grade II or higher or Outerbridge classification grade II and above. [19, 20]

Sample size

The sample size is calculated based on the calculation method of a pilot study [21], and the effect size is calculated as follows: d = 0.8; α = 0.05; Power = 0.8. The minimum sample size per group is 26, and the estimated loss of follow-up/exit rate is 20%. Thus, a maximum of 32 participants will be recruited for each group.

Intervention measures

Surgical procedure

All surgeries will be performed by 1 designated surgeons from our center. These designated surgeons are required to undergo a meniscal surgical technique evaluation and register with the center before being permitted to participate in the study and perform surgeries. Arthroscopic surgery will be performed under general anesthesia. If repairable meniscal lesions are found in the patient, the lesions will be first cleared using a shaver for freshening. All patients will receive a full-endoscopic meniscal repair technique, and if the meniscus is not repairable, the partial meniscectomy can be performed to reshape the meniscus.

Postoperative rehabilitation protocol

A standardized postoperative rehabilitation protocol will be adopted according to the clinical guidelines in the American Academy of Orthopaedic Surgeons (AAOS) postoperative rehabilitation manual18 and previous studies1,2,4,14 (Table 1). [1, 2, 4, 18, 22]

Experimental group

Patients will receive daily exercise plans via a software platform for 12 weeks. This software can be installed on the patients' smartphones. Once the patient initiates the digital therapy platform, the system will utilize the smartphone's camera to capture the patient's position in real time (Fig. 2). As the patient performs the physical therapy exercises, the system will provide demonstration videos, verbal and auditory feedback on the patient's exercise posture and duration. The digital platform will also record the patient's exercise frequency and duration. This data will then be transmitted to the software platform for detailed analysis by the research team.

Fig. 2
figure 2

The camera system

Smartphone camera requirements

It is recommended to have a photo resolution of 1280 × 720 pixels or higher and a video frame rate of 30 frames per second or higher for the smartphone camera.


The environment in which the camera is placed should have appropriate lighting conditions. Adequate natural light or suitable illumination can enhance image quality and recognition accuracy. The camera should be placed in a stable position to avoid image blurring. The camera's position should be adjusted to capture the patient in the frame. Ensure that the camera's field of view is unobstructed.

Human pose detection model

The human pose detection model (Fig. 3) will detect critical anatomical landmarks on the human body using images or videos captured by a smartphone camera. By employing a convolutional neural network (CNN) algorithm, this model can identify various joints, including the shoulder, elbow, wrist, hip, knee, and ankle. It will then generate a 3D skeletal model (Fig. 4), tracking dynamic changes in the patient's body position during motion.

Fig. 3
figure 3

Human pose detection model recognizing knee flexion

Fig. 4
figure 4

3D skeletal model reconstruction

When the model detects exercise postures that deviate from preset values, it will provide auditory and verbal feedback via the software to remind the patient to adjust their position based on the instructional video. Hence, the patient will receive interactive feedback during the training process.

Control group

Patients will be referred to the clinic weekly for 12 weeks postoperatively, during which they will receive instructions on physical therapy exercises. Every week, patients will learn the exercises at the clinic and will then be asked to take the instruction sheet home to complete the remaining exercises.


To increase compliance, patients will receive adequate education on software usage before the start of the trial. Throughout the trial, patients will receive detailed instructions via handouts and reminders via texts and calls. Support will be provided to patients to answer questions throughout the trial. Compliance data will be monitored based on software usage data and patient attendance data in the clinic. The research team will collect data on patients' training frequency and duration to objectively assess participant compliance within the study.


Clinical knee function evaluation

The estimated follow-up time was 1 day before surgery and 2, 6, 12, and 24 weeks after surgery, and the specific evaluation items included knee ROM (2, 6, 12, 24 weeks after surgery), weight-bearing progress (when fully weight bearing) and knee function and pain scores [23] (Lysholm score, Tegner score, IKDC knee subjective function score, and VAS score). At the final follow-up assessment, return-to-play (RTP) status will be evaluated based on the rehabilitation literature available for meniscus repair. (Table 2). [4, 24, 25]

Table 2 Return-to-play assessment

The clinical function of patients was assessed and evaluated based on IKDC and VAS score improvement via the criteria of minimal clinically important difference (MCID), patient accepted symptom state (PASS), and significant clinical benefit (SCB) [25].According to the study of Gowd et al. [25], the MCID threshold, which represents the minimal clinically important difference, is defined as an improvement of at least 10.6 points in the IKDC score after treatment or intervention. This threshold signifies that a clinically meaningful change in knee joint function is recognized. The SCB threshold, indicating significant clinical benefit, is set at an IKDC score of 27.3 points or higher after treatment or intervention. This threshold is used to confirm that the treatment has brought about a significant clinical benefit. Lastly, the threshold for patient accepted symptom state (PASS) is established at an IKDC score of 57.9 points or higher. This threshold signifies that patients subjectively consider their symptom state to be acceptable.

Radiographic evaluation

Evaluation time point and project: X-ray: 1 day before surgery, 1 day, and 6 months after surgery; the anteroposterior and lateral position of the knee joint was examined; MRI (T1-weighted and T2-weighted images of axial, coronal and sagittal positions): 1 day before surgery, 6 months after surgery. The imaging assessment was conducted in a blinded manner, and the evaluators were not informed of the identity and grouping of the patients. Imaging evaluation should be performed without intervention.

Complications assessment

Postoperative complications, such as postoperative knee infection, deep venous thrombosis of the lower limbs, stiffness, and arthrofibrosis, will be recorded.

Statistical analysis

Quantitative data will be analyzed using SPSS 21.0 (IBM, New York, NY, USA). The demographic, social, and preclinical characteristics of the subjects in both groups will be described. Differences in these variables between the intervention and control groups will be analyzed using either the Chi-square test or the T test, with continuous variables recorded as the mean and standard errors and categorical variables as rates (incidence). For continuous variables, the Shapiro‒Wilk test will be performed to verify that the variables in each group obeyed a normal distribution. For repeated measures, 2-way ANOVA analysis will be conducted. Two independent samples mean comparison T tests were used to compare the normally distributed continuous variables between groups. For nonparametric variables, the Kruskal‒Wallis test was used to determine the differences between the groups. Chi-square tests and Fischer exact tests were used to determine differences in categorical variables. Statistical significance was set at p < 0.05.

Adverse event management

In this study, inadequate healing of the meniscus may be observed following suturing. After a 6-month postoperative MRI review and clinical evaluation, some patients may require an arthroscopic review to confirm meniscus healing. If the repair of the meniscus is unsuccessful, then partial meniscectomy will be undertaken.

Availability of data and materials

The datasets used or analyzed during the current study are available from the corresponding author upon reasonable request.


  1. Sherman SL, DiPaolo ZJ, Ray TE, Sachs BM, Oladeji LO. Meniscus injuries. Clin Sports Med. 2020;39:165–83.

    Article  PubMed  Google Scholar 

  2. O’Donnell K, Freedman KB, Tjoumakaris FP. Rehabilitation protocols after isolated meniscal repair: a systematic review. Am J Sports Med. 2017;45:1687–97.

    Article  PubMed  Google Scholar 

  3. Cavanaugh JT, Killian SE. Rehabilitation following meniscal repair. Curr Rev Musculoskelet Med. 2012;5:46–58.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Wedge C, Crowell M, Mason J, Pitt W. Rehabilitation and return to play following meniscus repair. Sports Med Arthrosc Rev. 2021;29:173–9.

    Article  PubMed  Google Scholar 

  5. Mariani PP, Santori N, Adriani E, Mastantuono M. Accelerated rehabilitation after arthroscopic meniscal repair: a clinical and magnetic resonance imaging evaluation. Arthrosc J Arthrosc Relat Surg. 1996;12(6):680–6.

    Article  CAS  Google Scholar 

  6. Antonio CT, et al. Components of compulsory service program for health professionals in low- and middle-income countries: a scoping review. Perspect Public Health. 2020;140:54–61.

    Article  CAS  PubMed  Google Scholar 

  7. Stander J, Grimmer K, Brink Y. Learning styles of physiotherapists: a systematic scoping review. BMC Med Educ. 2019;19:2.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Gorgon EJR, Barrozo HGT, Mariano LG, Rivera EF. Research evidence uptake in a developing country: a survey of attitudes, education and self-efficacy, engagement, and barriers among physical therapists in the Philippines. J Eval Clin Pract. 2013;19:782–90.

    Article  PubMed  Google Scholar 

  9. Zhang Z-Y, et al. Digital rehabilitation programs improve therapeutic exercise adherence for patients with musculoskeletal conditions: a systematic review with meta-analysis. J Orthop Sports Phys Ther. 2022;52:726–39.

    Article  PubMed  Google Scholar 

  10. Berton A, et al. Virtual reality, augmented reality, gamification, and telerehabilitation: psychological impact on orthopedic patients’ rehabilitation. J Clin Med. 2020;9:E2567.

    Article  Google Scholar 

  11. Zeng X, Zhu G, Zhang M, Xie SQ. Reviewing clinical effectiveness of active training strategies of platform-based ankle rehabilitation robots. J Healthc Eng. 2018;2018:2858294.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Chalian M, et al. The QIBA profile for MRI-based compositional imaging of knee cartilage. Radiology. 2021;301:423–32.

    Article  PubMed  Google Scholar 

  13. Byra J, Czernicki K. The effectiveness of virtual reality rehabilitation in patients with knee and hip osteoarthritis. J Clin Med. 2020;9:2639.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Gazendam A, Zhu M, Chang Y, Phillips S, Bhandari M. Virtual reality rehabilitation following total knee arthroplasty: a systematic review and meta-analysis of randomized controlled trials. Knee Surg Sports Traumat Arthrosc Off J ESSKA. 2022;30:2548–55.

    Article  Google Scholar 

  15. Li L. Effect of remote control augmented reality multimedia technology for postoperative rehabilitation of knee joint injury. Comput Math Methods Med. 2022;2022:9320063.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Hellsten T, Karlsson J, Shamsuzzaman M, Pulkkis G. The potential of computer vision-based marker-less human motion analysis for rehabilitation. Rehabil Process Outcome. 2021;10:11795727211022330.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Lind M, Nielsen T, Faunø P, Lund B, Christiansen SE. Free rehabilitation is safe after isolated meniscus repair: a prospective randomized trial comparing free with restricted rehabilitation regimens. Am J Sports Med. 2013;41:2753–8.

    Article  PubMed  Google Scholar 

  18. Harput G, Guney-Deniz H, Nyland J, Kocabey Y. Postoperative rehabilitation and outcomes following arthroscopic isolated meniscus repairs: a systematic review. Phys Ther Sport. 2020;45:76–85.

    Article  PubMed  Google Scholar 

  19. Kohn MD, Sassoon AA, Fernando ND. Classifications in brief: Kellgren–Lawrence classification of osteoarthritis. Clin Orthop Relat Res. 2016;474:1886–93.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Slattery C, Kweon CY. Classifications in brief: outerbridge classification of chondral lesions. Clin Orthop Relat Res. 2018;476:2101–4.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Browne RH. On the use of a pilot sample for sample size determination. Stat Med. 1995;14:1933–40.

    Article  CAS  PubMed  Google Scholar 

  22. Andrew Green, Roman Hayda, Andrew Hecht. Postoperative Orthopaedic Rehabilitation. American Academy of Orthopedic Surgeons. 2017; ISBN :9781496360281.

  23. Tanner SM, Dainty KN, Marx RG, Kirkley A. Knee-specific quality-of-life instruments: which ones measure symptoms and disabilities most important to patients. Am J Sports Med. 2007;35:1450–8.

    Article  PubMed  Google Scholar 

  24. van Melick N, et al. Evidence-based clinical practice update: practice guidelines for anterior cruciate ligament rehabilitation based on a systematic review and multidisciplinary consensus. Br J Sports Med. 2016;50:1506–15.

    Article  PubMed  Google Scholar 

  25. Gowd AK, Lalehzarian SP, Liu JN, Agarwalla A, Christian DR, Forsythe B, Cole BJ, Verma NN. Factors associated with clinically significant patient-reported outcomes after primary arthroscopic partial meniscectomy. Arthrosc J Arthrosc Relat Surg. 2019;35(5):1567–75.

    Article  Google Scholar 

Download references


Funding for this research project is provided by the Sichuan Science and Technology Department's approved project on "Application research of individualized and precise injury prevention and treatment of knee joint based on the development of intelligent wearable devices and remote 5G technology sports and rehabilitation platform." The fund number is 2022YFS0372.

Author information

Authors and Affiliations



LW and XC are the primary designers of the research. All authors have made suggestions for the research design and will be involved in the execution of the research.

Corresponding authors

Correspondence to PengCheng Li or Jian Li.

Ethics declarations

Ethics approval and consent to participate

This study has been registered at the Chinese Clinical Trial Registry (ChiCTR) website; Registration number: ChiCTR2300070582. This study was approved by the Clinical Trial Ethics Review Committee of West China Hospital of Sichuan University.

Competing interests

Di Liu is an employee of Jiakang Zhongzhi Technology Company.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, L., Chen, X., Deng, Q. et al. Effectiveness of a digital rehabilitation program based on computer vision and augmented reality for isolated meniscus injury: protocol for a prospective randomized controlled trial. J Orthop Surg Res 18, 936 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: