Accuracy and Precision of Microelectronic Measuring Systems (MEMS)

Abstract

Microelectronic Measuring Systems (MEMS) are being used to capture kinematic data in real-world environments. The benefits of using MEMS are their small size, relatively low cost (compared to an Optical Motion Capture System) and the ability to capture real-time data in almost any environment. The accuracy and precision of MEMS can be influenced by elements in their surrounding environment such as building materials (i.e., reinforced steel) and structural components (i.e., elevators). Recognizing the influence of the environment on MEMS output is important if the MEMS are to be used in real-world environments where subjects could navigate between various environments. MEMS can also be affected by dynamic motion therefore testing of the MEMS in the same conditions in which they are to be used will help to identify any issues prior to data collection. The overall purpose of this thesis was to determine if the outputs of four Shimmer 2r MEMS were accurate and precise enough in static and dynamic conditions to use in a future study to assess gait activities of daily living in individuals with a unilateral transtibial amputation. In order to understand the effect of the environment on the MEMS, accuracy and precision were assessed in a rural environment (to reduce the effect of building materials and structural components) as well as the clinical environment where they will eventually be used for research. The MEMS were also evaluated in static and dynamic conditions to better understand how motion affected accuracy and precision. The results of this study confirmed that the clinical environment affected the MEMS outputs. During the dynamic condition, the gyroscope output of one MEMS sensor was significantly different than the other devices indicating recalibration or possible exclusion from future studies. Prior to using MEMS in research, it is advisable to investigate the effects of the environment on the sensor outputs as well as assess the performance of the individual sensors.