Evaluating a Computer Vision-Based Rehabilitation Program for Osteosarcoma Patients: Randomized Controlled Trial (Preprint)

Abstract

UNSTRUCTURED Abstract Background: Osteosarcoma can affect the function of the lower limbs when found in the distal femur and proximal tibia. Following total knee arthroplasty, patient may exhibit instability, limited range of motion, difficulty walking, reduced ability to bear weight, joint pain, joint dysfunction, and a decrease in daily life performance. This study aims to evaluate the efficacy of a computer vision-based intelligent rehabilitation program to manage patients post-operatively. Method: 96 patients with osteosarcoma requiring knee arthroplasty were recruited. Patients were randomized into either the intelligent rehabilitation group (IRG) or natural rehabilitation group (NRG) with an equal distribution. The main endpoints included the range of motion (ROM) of the knee joint and Knee injury and Osteoarthritis Outcome Score (KOOS). Secondary outcomes include 6-minute walk test (6MWT), Timed Up and Go test (TUG), the Toronto Extremity Salvage Score (TESS), SF-36 Scale Role Limitations due to Emotional Problems (RE), and Mental Health (MH) at three months and six months post-operatively. Results: In a 6-month study periodl, participants successfully concluded the trial.The IRG showed significant improvement in ROM (p=0.015), KOOS, 6-minute walk test (6MWT) (p=0.037), Timed Up and Go (TUG) test (p=0.041), Treatment Effectiveness and Satisfaction Survey (TESS) score (p=0.039), SF-36 Role Limitations due to Emotional Problems (RE) score (p<0.001) , and SF-36 Mental Health (MH) score (p<0.001) . Conclusion: This study provided evidence that the participants in the IRG showed considerable enhancements in joint function relative to the NRG. These findings confirmed the effectiveness of the computer vision-based intelligent rehabilitation in the post-surgery recovery of knee arthroplasty in osteosarcoma.