Intra-cyclic analysis of the backstroke swimming technique with an inertial measurement unit

Over the past decades, the use of video to monitor and analyze the performance in swimming has become the gold standard of motion analysis. Due to the high cost of a professional video system as well as the time and expertise required to evaluate performance relevant parameters, these systems are only available to athletes at international level. To enable training analysis also for athletes at national level, more cost-effective systems such as Inertial Measurement Units (IMU) are becoming more and more important. The advantage of these systems is that they are easily accessible and offer the possibility to automatically analyze the movement of a swimmer. The current study addresses backstroke swimming and transfers the knowledge gained through video analysis to the data measured with an IMU. The focus is on intra-cyclic characteristics such as the body's side-to-side roll, angular velocity and forward acceleration. Ten athletes from regional to national level swam 100 m backstroke with an IMU positioned on the lower back and were simultaneously recorded on video. The IMU data obtained was linked to the video to identify key positions during a swimming stroke cycle and to find similarities and differences between the swimmers. The findings are the basis for the development of an automatic pattern recognition system that provides coaches and scientists with direct, real-time feedback on the execution of swim movements. In addition, it provides information on which parameters should be specifically trained to improve performance.