Do we still need to screen our patients?-Orthopaedic scoring based on motion tracking

Area of Science:

• Orthopaedic surgery
• Biomedical engineering
• Rehabilitation medicine

Background:

• Orthopaedic scores are crucial for assessing movement disorders but traditionally require manual scoring by experienced clinicians.
• Wearable systems offer a promising avenue for convenient motion tracking data collection in medical applications.

Purpose of the Study:

• To demonstrate the feasibility of automated orthopaedic scorings using motion tracking data.
• To apply this methodology to the Harris Hip Score and the Knee Society Score as examples.

Main Methods:

• Instrumental gait analysis was performed on 78 patients post-hip or knee arthroplasty.
• 744 gait features were extracted, and hierarchical multiple regression analysis with tenfold cross-validation was employed.
• A 70%/30% data split was used for training and testing regression models.

Main Results:

• Orthopaedic scores were reproduced with excellent coefficients of determination (R²).
• Regression models utilized 4-6 gait features and the 'pain' parameter.
• High accuracy was achieved in training, testing, and cross-validation phases.

Conclusions:

• Automated orthopaedic scoring based on motion tracking provides objective joint function evaluation for hip and knee.
• This technology, combined with wearable sensors, can assist healthcare professionals with telemedical scoring for numerous patients.
• The method reduces the reliance on trained clinicians for manual scoring, enhancing efficiency and accessibility.