Gait analysis in spastic hemiplegia and diplegia cerebral palsy using a wearable activity tracking device- a data quality analysis for deep convolutional neural networks

Cerebral Palsy (CP) is a movement disorder in children and affects their course of physical activity as they move to adulthood. Clinical gait analysis is the most acceptable technique to quantify and understand the defects in their gait. We developed a prototype of a wearable physical activitytracking device for gait analysis in cerebral palsy. Our device is a manifestation of a recently proposed architecture on Wearable Internet of Things (WIoT). We demonstrate the design of our device and illustrate two sets of imitated data collected by a volunteer for spastic hemiplegia and diplegia, which are two different types of CP. We sampled the data every 120ms to preserve the pattern of motion. We prepared our data in one and two dimensions to present the data to a deep convolutional neural network. The data shows some variations between different groups and opens a channel for future research.