Eventos Anais de eventos
COBEM 2019
25th International Congress of Mechanical Engineering
LINEAR MODELS BACKSTROKE START TIME PREDICTION USING EMG
Submission Author:
Karla de Jesus , AM , Brazil
Co-Authors:
Karla de Jesus, Leandro dos Santos Coelho, Helon Vicente Hultmann Ayala, Kelly de Jesus, ricardo fernandes, J. Paulo Vilas-Boas
Presenter: Karla de Jesus
doi://10.26678/ABCM.COBEM2019.COB2019-1040
Abstract
The start phase is a junction of explosive movements intended to boost the swimmers from the block/wall, representing an important part of the short distance swimming events. We aimed to apply linear models to predict 15 m backstroke start time using electromyographic (EMG) data. Following a four-week start familiarization with each start variant, 10 male backstroke swimmers randomly performed six maximal 15 m trials with feet parallel and partially emerged, but three with a horizontal handgrip and three with a vertical handgrip (2 min rest in-between trials). Surface EMG of Biceps Brachii, Triceps Brachii, Rectus Femoris, Biceps Femoris, Gastrocnemius Medialis and Tibialis Anterior was recorded and processed using the time integral EMG (iEMG). Eight video cameras (four surface and four underwater) were used to determine backstroke start hands-off, take-off, flight, entry and underwater phases. A Data-driven approach based on regression linear models using iEMG of each backstroke start phase and muscle in both start variants can be useful to predict dynamical behavior related to start component in swimming analysis.. The implementation of the linear mathematical model requires optimizing its parameters according to measured data. Preliminary results show that the obtained linear model is able to capture the relationship present in the data. Future submission of the full paper will include the importance of each predictor, different types of learning algorithms for linear models such as Least absolute shrinkage and selection operator, ridge regression, elastic net, and the test on nonlinear model structures.
Keywords
muscle activation, linear models, swimming start, performance
