Generating Music from Movement to Improve Gait of Individuals with Spastic Cerebral Palsy

Cerebral palsy (CP) is a neurological condition that impacts 764,000 people in the United States [1]. Individuals with spastic CP often experience gait disorders. A high dose of repetitive exercises can be beneficial for patients with neurological conditions, including patients with spastic cerebral palsy, in improving muscle strength and overcoming impaired selective motor control. Therapies that are currently being used to address impaired gait behavior are potentially unmotivating and limited to therapy time. They are used roughly three times a week for an hour as they require patients to physically go to a hospital or a physical therapy center. This is a relatively low frequency and thus requires a long period of time to yield improvement in gait behavior. Thus, the goal of this project is to provide a new form of physical therapy that uses music to encourage patients to engage in these repetitive exercises. We propose and have developed a generative music technology intended to motivate high repetition and encourage gait pattern improvements. This system gives users the means to actively create music as opposed to following along to an existing song or pattern. Users can generate different types of music and chords as they walk to encourage these repetitive patterns. To meet these user needs, we established technical requirements addressing performance, reliability, and enjoyability.

To fulfill these goals and requirements, we’ve built a full system containing a footswitch, knee angle-detection sleeve, and iOS app, which all communicate with each other wirelessly. The footswitch enables precise detection of initial contact and toe-off through the use of two force resistive sensors: one on the heel and one on the toe. We then mapped these gait events to music for user feedback and enjoyment. The knee sleeve provides data on knee angle by measuring the angle between two Inertial Measurement Units (IMUs), one located on the shin and one located above the knee. This provides user information about knee flexion angle which is then used to provide positive feedback to our users. We tested our device against our technical requirements and made design changes to ensure that we met each requirement.