LLM-Guided Rehabilitation in Degenerative Knee Disease (LLM-RehabKnee)

The Effect of ChatGPT-5, Gemini 2.5 Pro, and DeepSeek V3.1 Guided Rehabilitation on Clinical Outcomes in Individuals With Degenerative Knee Disease

This randomized, four-arm clinical trial will investigate the effects of large language model (LLM)-assisted exercise prescriptions integrated with conventional physiotherapy on pain, function and quality of life in adults with degenerative knee disease. Participants will be randomly assigned to conventional physiotherapy alone or to conventional physiotherapy plus an exercise program planned with the assistance of ChatGPT-5, Gemini 2.5 Pro, or DeepSeek V3.1. All participants will receive supervised outpatient physiotherapy two times per week for 8 weeks (16 sessions in total). Outcomes will be assessed at baseline and at the end of the 8-week intervention period by physiotherapists blinded to group allocation.

Study Overview

• Status: Completed
• Conditions: Degenerative Knee Disease
• Intervention / Treatment: Behavioral: ChatGPT-5-assisted exercise planning; Behavioral: DeepSeek V3.1-assisted exercise planning; Behavioral: Gemini 2.5 Pro-assisted exercise planning; Behavioral: Supervised conventional physiotherapy

Detailed Description

Degenerative knee disease, including knee osteoarthritis, is a common cause of pain, disability and loss of independence in middle-aged and older adults. Exercise-based rehabilitation is a core component of evidence-based management, but the content and progression of exercise programs often vary between clinicians. Large language models (LLMs) such as ChatGPT-5, DeepSeek V3.1 and Gemini 2.5 Pro can generate exercise suggestions based on clinical information and may support physiotherapists in designing and progressing individualized programs. However, their impact on clinical outcomes in degenerative knee disease has not been evaluated in randomized controlled trials.

This single-center, randomized, parallel-group clinical trial will be conducted in adults with degenerative knee disease referred to the physiotherapy clinic of Kirsehir Ahi Evran University. After written informed consent and baseline assessments, participants will be randomly assigned (1:1:1:1) to one of four groups. All groups will receive a standardized outpatient physiotherapy program including hot pack, conventional TENS, therapeutic ultrasound and strengthening, range of motion, proprioceptive and functional exercises for the knee joint. In the three experimental groups, the content and progression of the exercise program will be planned with the assistance of standardized prompts to ChatGPT-5, DeepSeek V3.1 or Gemini 2.5 Pro, and all LLM-generated suggestions will be reviewed for safety and appropriateness by physiotherapists before being implemented. Sessions will be delivered two times per week for 8 weeks (16 supervised visits) under the supervision of a physical medicine and rehabilitation specialist and physiotherapists.

Outcomes will be evaluated at baseline (week 0) and after the 8-week intervention by a physiotherapist blinded to group allocation. Main outcome domains include knee pain, knee-related patient-reported outcomes, objective functional performance tests, isometric muscle strength, knee range of motion, psychosocial factors and health-related quality of life. Data will be analyzed according to CONSORT guidelines for randomized trials.

• Study Type: Interventional
• Enrollment (Actual): 84
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Turkey (Türkiye)
  • Merez
    • Kırşehir, Merez, Turkey (Türkiye), 40100
      • Kırşehir Ahi Evran University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age between 40 and 65 years.
• Clinical and radiographic diagnosis of degenerative knee disease in the target knee.
• Knee pain in the target knee for at least 3 months.
• Ability to bear weight on the target limb and perform the planned functional tests.
• Sufficient cognitive and physical capacity to participate in an 8-week supervised exercise program.
• Willingness to participate and ability to provide written informed consent.

Exclusion Criteria:

• Intra-articular corticosteroid or similar injection to the target knee within the previous 3 months.
• Knee surgery on the target knee within the previous 6 months.
• Severe malalignment of the knee or concomitant severe neurological disease and/or balance disorder that would interfere with safe testing or exercise.
• Pregnancy or any systemic disease that would contraindicate participation in the rehabilitation program according to the physician.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Quadruple

Number of Arms

4

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Conventional physiotherapy; Participants receive supervised conventional physiotherapy including hot pack, TENS, ultrasound and a standardized knee exercise program two times per week for 8 weeks (16 sessions).	Behavioral: Supervised conventional physiotherapy; Supervised outpatient physiotherapy including hot pack (20 minutes), conventional TENS (100 Hz, 60 µs, 20 minutes), therapeutic ultrasound (1 MHz, 1.5 W/cm², 5 minutes) and a standardized knee exercise program (strengthening, range of motion, proprioceptive and functional exercises), delivered two times per week for 8 weeks (16 sessions in total).
Experimental: Conventional physiotherapy plus ChatGPT-5-assisted exercise planning; Participants receive supervised conventional physiotherapy plus an exercise program whose content and progression are planned with the assistance of the ChatGPT-5 large language model, under physiotherapist supervision, two times per week for 8 weeks (16 sessions).	Behavioral: ChatGPT-5-assisted exercise planning; Exercise selection and progression for the knee rehabilitation program supported by standardized prompts to the ChatGPT-5 large language model. All LLM-generated suggestions are reviewed for safety and appropriateness by physiotherapists before being implemented in supervised sessions.
Experimental: Conventional physiotherapy plus DeepSeek V3.1-assisted exercise planning; Participants receive supervised conventional physiotherapy plus an exercise program whose content and progression are planned with the assistance of the DeepSeek V3.1 large language model, under physiotherapist supervision, two times per week for 8 weeks (16 sessions).	Behavioral: DeepSeek V3.1-assisted exercise planning; Exercise selection and progression for the knee rehabilitation program supported by standardized prompts to the DeepSeek V3.1 large language model. All LLM-generated suggestions are reviewed for safety and appropriateness by physiotherapists before being implemented in supervised sessions.
Experimental: Conventional physiotherapy plus Gemini 2.5 Pro-assisted exercise planning; Participants receive supervised conventional physiotherapy plus an exercise program whose content and progression are planned with the assistance of the Gemini 2.5 Pro large language model, under physiotherapist supervision, two times per week for 8 weeks (16 sessions).	Behavioral: Gemini 2.5 Pro-assisted exercise planning; Exercise selection and progression for the knee rehabilitation program supported by standardized prompts to the Gemini 2.5 Pro large language model. All LLM-generated suggestions are reviewed for safety and appropriateness by physiotherapists before being implemented in supervised sessions.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in KOOS total score	Knee Injury and Osteoarthritis Outcome Score (KOOS) total score (0-100), calculated as the mean of the five KOOS subscales (Pain, Symptoms, Activities of Daily Living, Sport/Recreation, and knee-related Quality of Life). Higher scores indicate better knee-related health. The primary analysis will compare change in KOOS total score from baseline to week 8 between the four groups.	Baseline (week 0) to end of treatment (week 8)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Change in knee pain intensity at rest, at night and during activity (Numeric Pain Rating Scale)	Numeric Pain Rating Scale (0-10) scores for knee pain at rest, at night and during activity. Higher scores indicate more severe pain. Mean change in each pain condition from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in IKDC subjective knee score	International Knee Documentation Committee (IKDC) subjective knee form score (0-100). Higher scores indicate better symptoms, sports activity and knee function. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in Lysholm knee score	Lysholm knee score (0-100). Higher scores indicate better knee function. Change from baseline to week 8 will be analyzed between the four groups.	Baseline (week 0) to week 8
Change in 30-second chair stand test performance	Number of full stands completed in 30 seconds from a standard chair. Higher numbers indicate better lower limb functional performance. Change in repetitions from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in stair climb test time	Time (seconds) needed to ascend and/or descend a standardized set of stairs. Lower times indicate better functional performance. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in maximal isometric quadriceps strength	Maximal voluntary isometric strength of the quadriceps muscle of the target knee, measured with a handheld dynamometer. Higher values (e.g., Newtons) indicate greater muscle strength. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in maximal isometric hamstring strength	Maximal voluntary isometric strength of the hamstring muscles of the target limb, measured with a handheld dynamometer. Higher values indicate greater muscle strength. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in knee range of motion	Active knee flexion and extension range of motion measured with a universal goniometer in degrees. Higher flexion and closer-to-zero extension deficit indicate better joint mobility. Changes in range of motion from baseline to week 8 will be analyzed.	Baseline (week 0) to week 8
Change in Pain Catastrophizing Scale score	Pain Catastrophizing Scale total score (0-52). Higher scores indicate greater levels of pain catastrophizing. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in Tampa Scale for Kinesiophobia score	Tampa Scale for Kinesiophobia total score (17-68). Higher scores indicate greater fear of movement or re-injury. Change from baseline to week 8 will be analyzed between groups.	Baseline (week 0) to week 8
Change in SF-12 physical and mental component summary scores	Short Form-12 (SF-12) physical and mental component summary scores (0-100). Higher scores indicate better health-related quality of life. Change from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8
Change in Four Square Step Test performance	Four Square Step Test (FSST): the time (seconds) taken to step sequentially over four canes arranged in a cross, moving forward, sideways, and backward. Lower times indicate better dynamic balance and functional mobility. Change in completion time from baseline to week 8 will be compared between groups.	Baseline (week 0) to week 8

Collaborators and Investigators

• Sponsor: Kirsehir Ahi Evran Universitesi

Study record dates

Study Major Dates

• Study Start (Actual): November 26, 2025
• Primary Completion (Actual): May 15, 2026
• Study Completion (Actual): May 15, 2026

Study Registration Dates

• First Submitted: November 25, 2025
• First Submitted That Met QC Criteria: November 25, 2025
• First Posted (Actual): December 5, 2025

Study Record Updates

• Last Update Posted (Actual): May 28, 2026
• Last Update Submitted That Met QC Criteria: May 26, 2026
• Last Verified: May 1, 2026

More Information

Terms related to this study

• Keywords: rehabilitation; physiotherapy; Degenerative knee disease; ChatGPT; large language model; DeepSeek; Gemini
• Other Study ID Numbers: 2025/1056

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Individual participant data (IPD) will not be shared because there is currently no approved institutional or ethical framework for external IPD sharing in this study.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No