Problem

• Gait asymmetry is defined as a dissimilarity in the kinematic motions of the legs during the walking cycle

• Having an asymmetric gait can potentially lead to further injury if not appropriately managed

• It is currently difficult to monitor subtle changes in gait asymmetry outside of a laboratory setting

Solution

• Design an algorithm that can predict gait asymmetry with the use of only acceleration signals

• This algorithm could be incorporated into a wearable sensor for use outside of the lab

• For increased efficiency, the work was divided into the following two projects:

1. The Ground Truth Project

• Utilized vertical ground reaction forces (GRF) to create a model capable of predicting an asymmetry index (AI) 

• Model data was used as a “ground truth” reference for the acceleration project

  2. The Acceleration Project

• Identified gait events in acceleration signals

• Segmented acceleration signal into gait cycles without the aid of GRF