Abstract

Determining foot touchdown and foot takeoff is an essential part of gait analysis protocols and is usually accomplished using force plate data and thresholds for the vertical component of the ground reaction force (GRF). It has been proposed that gait events can also be estimated from kinematic data, and several algorithms have already been published. However, many of these algorithms were designed and specified for application in walking gait only. Hence, the purpose of this paper is to present our newly developed foot contact algorithm (FCA) and compare its estimation accuracy for touchdown and takeoff to a selection of existing algorithms. Forty one subjects were analyzed in a laboratory setup recording ground reaction forces and 3D-kinematic data of the lower limbs during running in three different running shoes. Prediction error of touchdown and takeoff was quantified in relation to gold standard events determined from force-plate data. Out of the evaluated algorithms, only FCA was able to estimate touch down and take off with similar precision, and is recommended when analyzing running gait. Furthermore, we found no evidence that different running shoes influenced the prediction error for any of the used algorithms. Verf.-Referat