Out-of-the-lab Biomechanical Assessments in Cerebral Palsy

Cerebral palsy (CP) is a motor disorder caused by early brain damage. It primarily affects motor control, leading to significant deviations from normal gait patterns in individuals with CP. Common symptoms include increased muscle tone (spasticity), reduced baseline muscle tension (hypotonia), impaired precise movement (ataxia), and fluctuating muscle tone with uncontrolled (arm) movements (athetosis). The treatment of CP involves a multifaceted approach, combining conservative and surgical measures such as physiotherapy, occupational therapy, medications, and orthopaedic interventions. For medical decision-making, clinical 3D gait analysis serves as an essential tool to ensure optimal care for patients with CP. It allows for objective measurements of movement patterns, gait velocity, step length, step width, and other parameters that are crucial for diagnosis and treatment planning. It helps physicians and therapists to gain a detailed understanding of a patients movement patterns so that more informed treatment decision can be made. It allows for regular monitoring of the progress and timely adjustments to the therapy plan if necessary. Overall, 3D gait analysis contributes to enhancing the quality of life for individuals with CP by providing a solid foundation for personalized treatment. A significant drawback of instrumented 3D gait analysis lies in its limited availability. Currently, it is accessible only to a few institutions due to its reliance on expensive and complex motion-capture hardware. Analyses typically occur in specialized laboratories with trained personnel. To ensure that more patients with CP benefit from these technologies in their medical treatment, cost-effective and user-friendly alternatives are needed. Recent advancements in deep learning and computer vision have led to new technologies that allow 3D motion analysis based on simple video recordings, such as those captured by smartphones. These markerless motion capture technologies show promising potential for cost-effective and lab- independent gait analyses. However, their accuracy and reliability in individuals with CP require thorough investigation. This project aims to fill this gap through a comprehensive study that simultaneously validates the effectiveness and reliability of a cost-effective markerless motion analysis system for gait assessment. Specifically, we will examine the recently developed OpenCap system (www.opencap.ai) from Stanford. OpenCap combines straightforward smartphone technology with cloud computing, providing 3D motion analysis without the need for expensive hardware. Additionally, the data collected in this project will be used to (re)train the underlying pose estimation algorithms of OpenCap, enhancing their precision in recognizing movements specific to CP patients. To verify whether the optimized pose estimation algorithms indeed yield better results, the analyses will be repeated using these models.