Analysis of Big Data in Running Biomechanics: Application of Multivariate Analysis and Machine Learning Methods


  • Angkoon Phinyomark University of Calgary
  • Sean T. Osis University of Calgary
  • Reed Ferber University of Calgary


Much of the biomechanical research over the past 20 years has investigated the influence of potential injury risk factors in isolation. More likely, multiple biomechanical and clinical variables interact with one another and operate as combined risk factors to the point that traditional biomechanical analysis techniques (that is, using discrete variables, such as peak angles, together with a statistical hypothesis test, such as analysis of variance) cannot capture the complexity of these relationships. To identify these complex associations, advanced multivariate analysis and machine learning methods are necessary. However, to build accurate classification and prediction models, an adequate number of samples is needed, which grows exponentially with the number of variables used in the analysis. Therefore, to directly meet this need we have developed the infrastructure and established a worldwide and growing network of clinical and research partners all linked through the world's first automated 3-dimensional (3D) data collection and analysis system: 3D GAIT. Similarly, traditional data analytics may not be able to handle these large volumes of data. Hence, the appropriate multivariate analysis and machine learning methods must be developed. This paper begins with a brief introduction to our 3D data collection system, followed by a discussion of existing multivariate and machine learning methods that can be applied to big data analytics. Next, we provide a comprehensive overview of our proposed methods for 3D kinematic data during running from our database. Finally, important challenges and future research directions are presented.




How to Cite

A. Phinyomark, S. T. Osis, and R. Ferber, “Analysis of Big Data in Running Biomechanics: Application of Multivariate Analysis and Machine Learning Methods”, CMBES Proc., vol. 39, no. 1, May 2016.