AI predicts walking recovery after hip replacement surgery

Artificial intelligence can help predict how well patients with hip osteoarthritis will be able to walk again after surgery. Researchers at Karlsruhe Institute of Technology (KIT) have developed an AI model that analyzes movement patterns. This gait biomechanics analysis also allows rehabilitation programs to be tailored to the patient’s personal needs. Researchers believe this approach, developed for the hip joint, could be extended to other joints in the future. They published their results in the journal Arthritis Research & Therapy. (DOI: 10.1186/s13075-025-03709-2)

In 2024, approximately 200,000 people in Germany will undergo hip replacement surgery, making the procedure one of the most common orthopedic surgeries in German hospitals. Most often, such operations are performed to treat osteoarthritis of the hip, which is the result of wear of the cartilage surfaces of the femoral (femoral) head and hip socket. Patients respond differently to total hip arthroplasty in terms of mobility and pain relief.

Understanding these differences is the goal of a joint project involving the Traumatology and Orthopedics Clinic of the University of Frankfurt (Klinic für Unfallchirurgie und Orthopädie) and the Institute of Sports Science (IfSS) at KIT. This project (Improving surgical outcomes in hip osteoarthritis based on biomechanical and biomarker discoveries (HOBBID)) is sponsored by the German Research Foundation.

KIT researchers have developed an AI model to analyze the movement patterns of patients with hip osteoarthritis, using gait biomechanics data obtained before and after surgery. The data was acquired and processed by Frankfurt Medical University and provided to KIT for AI-based analysis.

Make highly complex biomechanical data available to applications

“Biomechanical data that describe the movement of biological systems in a mechanical, anatomical, and physiological way are extremely complex,” said Dr. Bernd J. Stetter, head of the Musculoskeletal Health Technology Research Group at IfSS and corresponding author of the study. “We are using AI models to make data available to applications, which is a step toward personalized treatment.

Stetter said the model was trained specifically to use hip prostheses, but could be used for other joints and diseases in the future. Such AI models have the potential to support physician decision-making, convey realistic expectations to patients, and enable personalized post-surgery rehabilitation.

Identify various gait change patterns

For this study, researchers analyzed the gait biomechanics of 109 patients with unilateral hip osteoarthritis before total hip arthroplasty. Sixty-three of these patients were re-evaluated after surgery, and 56 healthy individuals served as a control group. Three-dimensional joint angle and joint loading data were obtained from musculoskeletal modeling for all participants. AI-based analysis reveals that people with hip osteoarthritis can be assigned to three groups with different patterns of gait change. Certain biomechanical gait parameters, such as hip angle and load, have been shown to be particularly useful in determining which group an individual belongs to. The three groups also differed in age, height, weight, walking speed, and severity of osteoarthritis.

The three groups responded differently to surgery. In some patients, the improvement in gait biomechanics with the hip prosthesis was significant. For others it wasn’t. In other words, some were able to walk almost normally afterwards, while others continued to show clear deviations from the control group.

Our model allows us to predict who will benefit particularly strongly from surgery and who will subsequently require additional intensive care. Because the algorithm is explainable and transparent, we expect this model to have a high level of clinical acceptance. ”

Dr. Bernd J. Stetter, corresponding author of the study