This study aims to evaluate the efficacy and safety of an AI-based home rehabilitation application ("Exersite Rehab") for patients who have undergone hip or knee arthroplasty(Total Knee Arthroplasty (TKA), Total Hip Arthroplasty (THA) or hip fracture surgery). Following surgery, effective rehabilitation is crucial for functional recovery; however, many patients face barriers to consistent in-hospital exercise. This interventional study divides participants into two groups to compare outcomes. The experimental group will perform a 12-week home-based rehabilitation program guided by an AI application that provides real-time posture correction and monitoring. In contrast, the control group will perform the same exercise protocol using traditional printed educational brochures.The primary objective is to determine if the AI-based digital healthcare solution significantly improves physical function, measured by the Functional Ambulatory Category(FAC), compared to conventional methods. Secondary outcomes including pain levels(Visual Analog Scale (VAS)) and joint-specific functional scores (Knee Society Score (KSS), Harris Hip Score (HHS)) will also be assessed over the 12-week period.

• Study Overview and Randomization This study is conducted as a multi-center, randomized, open-label trial. After providing informed consent and completing baseline assessments, eligible participants are randomly assigned to either the Experimental Group or the Control Group at a 1:1 ratio. The primary focus is to evaluate the recovery of physical mobility through different home-based rehabilitation methods.
• Technical Intervention Protocols The Experimental Group utilizes the "Exercite Rehab" mobile application, which incorporates AI-based computer vision technology. The application recognizes the participant's joint movements in real-time through the device's camera, providing immediate corrective feedback on exercise posture and automatically logging the completion of prescribed routines. In contrast, the Control Group follows the same clinical exercise protocol using traditional printed educational materials. Participants in this group manually record their daily exercise performance in a provided physical logbook.
• Longitudinal Monitoring and Follow-up Following the initial in-hospital education session, participants transition to a 12-week home-based phase. Research personnel conduct remote monitoring at specific intervals (Weeks 4 and 8) to ensure safety, assess exercise adherence, and track symptomatic changes via telephone or the application's monitoring dashboard. A final face-to-face clinical evaluation is performed at Week 12 to determine the overall efficacy of the intervention compared to the conventional method.

Inclusion Criteria:

• Adult patients aged 19 years or older.
• Patients scheduled to undergo hip or knee arthroplasty at Gangnam Severance Hospital or affiliated hospitals of Chung-Ang University Medical Center (Chung-Ang University Hospital, Gwangmyeong Hospital, Hyundai Hospital).
• Patients with a preoperative Functional Ambulatory Category (FAC) stage of 4 or 5 (For hip fracture cases, pre-fracture FAC stage will be considered).
• Patients capable of downloading and using mobile applications on smart devices (smartphones or tablets) independently without researcher assistance.
• Patients who fully understand the study's purpose and procedures and have voluntarily signed the informed consent form.

Exclusion Criteria:

• Individuals unable to perform exercises due to severe musculoskeletal pain.
• Individuals with severe dizziness or poor vision that prevents them from following exercise videos provided by the digital treatment.
• Individuals with visual impairment that causes difficulty in recognizing the monitor/device screen.
• Individuals requiring long-term inpatient treatment due to unstable vital signs.
• Individuals with partial or total lower-limb amputation due to trauma, disease, or congenital defects.
• Individuals classified as vulnerable subjects (as defined by clinical research ethics guidelines).
• Any other individuals deemed ineligible for participation at the discretion of the principal investigator.

Individual participant data (IPD) will not be shared with other researchers to ensure the privacy and confidentiality of the participants, in accordance with the Institutional Review Board (IRB) guidelines and personal information protection protocols of Gangnam Severance Hospital. Only de-identified, aggregated study results and statistical analyses will be made available through scientific publications and conference presentations. Access to the raw data is restricted to the primary investigators and authorized research personnel involved in the "Dr. Answer 3.0" project.

This is an open-label study where masking of participants and care providers is not feasible due to the nature of the interventions (AI-based mobile application vs. printed educational brochure). Both groups will be clearly aware of the assigned rehabilitation method. However, to minimize bias, the statistical analysis will be conducted using de-identified data.