Validación de las técnicas basadas en sensores inerciales para el seguimiento de la actividad física : técnicas basadas en sensores inerciales para el seguimiento de la actividad física.

Abstract

Physical activity in humans is constant in every daily action performed, from relaxing activities to actions that require high energy expenditure. In many cases, the executed movement is deficient or incorrect, leading to accidental or future-developed pathologies. This is why biomechanical study allows evaluating and understanding characteristic movement patterns in different parts of the human body, providing a useful tool in diagnosing a particular subject and thus enabling future correction of motor actions in the observed limb. Given the importance of developing techniques focused on monitoring physical activity, the present work aims to validate the technique of collecting experimental data through inertial sensors in relation to Photogrammetry. To achieve this, a protocol for preferred movement is established, targeting the everyday motion exerted by cyclists, as it describes a stable and sequential movement pattern. Our analytical focus is on two points of the lower limb: the femur and the fibula. For this purpose, we turn to the development of programs in Matlab that allow us to control experimental data and compare the used techniques. These are based on numerical mathematical models of kinematic position, primarily quaternions, which enable the description of instantaneous position over time. Furthermore, an additional data treatment is developed to simulate the executed movement using OpenSim software, in order to establish the observation and analysis of the movement virtually represented by musculoskeletal models.