GaitViewer: Semantic Gait Data Analysis and Visualization Tool

Clinical gait analysis studies human locomotion by characterizing movement patterns with heterogeneous acquired gait data (e.g., spatio-temporal parameters, geometry of motion, measures of force). Lack of semantic integration of these heterogeneous data slows down collaborative studies among the laboratories that have different acquisition systems. In this work we propose a semantic integration methodology for gait data, and present GaitViewer a prototype web application for semantic analysis and visualization of gait data. The proposed semantic integration methodology separates heterogeneous and mixed numerical and meta information in gait data. Ontology concepts represent the separated meta information, while numerical information is stored in a NoSQL database. Parallel coordinates visual analytics technique are used as an interface to the analytics tools proposed by the NoSQL database. We tailor GaitViewer for two common use-cases in clinical gait analysis: correlation of measured signals for different subjects, and follow-up analysis of the same subject. Finally, we discuss the potential of a large-scale adoption of frameworks such as GaitViewer for the next generation diagnosis systems for movement disorders.