Cerebral Palsy Clinical Planning

Cerebral palsy (CP) is a debilitating disorder that causes a wide range of gait pathologies. CP affects approximately 800,000 Americans, and is the most common movement disability in children and adolescents. The financial burden of CP on families and the healthcare system is large, with the lifetime cost per person with CP estimated to be more than $1 million beyond normal living costs.

Diagnosis and treatment of gait pathologies in individuals with CP is challenging.

Diagnosis and treatment is challenging for a number of reasons, including the heterogeneous nature of the disorder, the complexity of the treatment history of individuals with CP, and the lack of tools to aid treatment planning. Children with CP often receive one or many invasive musculoskeletal surgeries, but walking doesn’t always improve. A major contributing factor is the subjective nature of clinical decisions: two surgeons might look at the same set of gait analysis data from a given patient and decide on different surgical plans.

Large clinical data sets hold the potential to improve outcomes for CP patients.

A large amount of data is collected in clinical centers. However, they are isolated, with each individual clinical lab using its own collection and annotation standards.  Furthermore, clinicians have few rigorous ways to use follow-up data on outcomes to improve their practice. Aggregating patient data from multiple clinical centers would generate sufficient data to make statistically significant inferences and predictions and offer a powerful, new approach to treatment design for CP patients.

The Mobilize Center will integrate different data science techniques and bring new insights into treatment planning for CP patients.

To improve treatment planning for CP, the Mobilize Center will:

  • Aggregate patient data from clinical centers, providing a subject pool that is complex and heterogeneous in terms of pathology, treatments, and the amount and type of data available
  • Integrate statistical learning and biomechanical modeling techniques to create “trained systems” to give treatment recommendations for children with cerebral palsy. Our systems will utilize data from our clinical data consortium and be validated against existing outcome data.
  • Develop novel approaches to effectively communicate a model’s predictions and recommendations to clinicians, helping them improve their decision-making process

While this framework will be developed with CP as the application, it will be applicable to a wide range of diseases.