Tracking of strength training: validation of a motion-recognition algorithm and a pilot towards 1RM, muscle loading and fatique index using a smart watch

Ubiquitous wrist-worn devices have become a noteworthy research topic in motion
recognition of strength training exercises. This study was designed to develop a watchOS
and iOS application and assess its exercise recognition and repetition counting accuracy.
As a pilot, a method to estimate the 1 repetition maximum and muscle stress and fatigue
was explored. To test the exercise recognition and repetition counting accuracy, a
workout consisting of nine sets with five randomly ordered resistance training exercises
and repetition amounts was conducted when wearing an Apple watch. Overall mean
%error in exercise recognition was 3.5% and 0.92% in repetition counting. In the future
this app can also be used for estimation of 1 repetition maximum, muscle stress and
fatigue.