Assessing whether a sensor-equipped insole for detecting changes in gait patterns can be used for preventive therapeutic purposes.
Background and Project Overview
Early diagnosis of deviations from standard gait patterns is vital for treatment and developing appropriate preventive therapeutic measures such as gait training. Detecting such deviations usually requires a therapist or special measurement systems not available to private users. However, to enable patients to work more independently on their therapy goals and make feedback on deviations in gait patterns available in everyday life situations, simple mobile systems would be of great help. Such systems should be designed to be easy to understand, simple to operate, and provide simple feedback so that they can be used autonomously by patients.
The current project expands on work done in the SONIGAIT II using the "stappone" insole. This insole was developed by the partner company stAPPtronics and records pedobarographic data (i.e., data on pressure between the foot and the floor). The accompanying app records an individual's movement behaviour. During SONIGAIT II, an interface that enables real-time data acquisition has been developed already. In addition, machine learning methods have been utilized to derive ground reaction forces and acceleration data of the soles. Finally, feedback systems were added that inform walkers about deviations from standard gait with the help of acoustic signals ("sonification").
Goals and Methods
SONIGAIT II InnoScheck builds upon a previous project, which involved the development and testing of a mobile auditory biofeedback system that assists users in correcting deviations from standard gait. The objective of this project is to further evaluate the effectiveness of such a system as an additional therapeutic tool.
First, it is necessary to validate the sonification machine learning models used to generate the auditory feedback, in order to better understand how effective they are and where modifications need to be made. The automatic classification from the machine learning models will be compared with the assessment of therapists to validate and improve the received auditory feedback.
Once users receive reliable auditory feedback, it is important to understand how they perceive such feedback and how they use it to correct their gait patterns. For this, data will be collected from healthy subjects wearing the insoles and performing different walking and posture exercises. While they perform the exercises, researchers fully track their posture and gait with a state-of-the-art motion capture system and examine how well people respond to the auditory feedback.
The results of these two main experiments will provide valuable insights for our researchers, advance mobile gait analysis technology, and help our company partners to improve their products.